HEALTH SECTOR REVIEW INSTITUTIONAL ANALYSIS UNDER DECENTRALIZATION

HEALTH SECTOR REVIEW INSTITUTIONAL ANALYSIS UNDER DECENTRALIZATION JULY 2014 1 AUTHORS: MICHAEL JONES BUDIHARDJA SINGGIH I MADE SUWANDI 2 Ackn...
Author: Dale Shelton
3 downloads 0 Views 2MB Size
HEALTH SECTOR REVIEW

INSTITUTIONAL ANALYSIS UNDER DECENTRALIZATION

JULY 2014

1

AUTHORS: MICHAEL JONES BUDIHARDJA SINGGIH I MADE SUWANDI

2

Acknowledgements The Health Sector Review (HSR) on Institutional Analysis under Decentralisation has been supported by the Australian-Indonesian Partnership for Health Systems Strengthening (AIPHSS). Funding was generously provided by the Australian Government’s Department for Foreign Affairs and Trade (DFAT). The Review Team would like to express its sincere thanks to all those who provided useful information and thereby helped in the formulation of the team’s conclusions and recommendations. We are particularly grateful to Minister of Health Dr. Nafsiah Mboi, who made time available to meet the team, and Ministry of Health Secretary General Dr. Supriyantoro who also made himself available for meetings with the Review Team. The team benefitted significantly from discussions with Director General of Medical Services Dr. Akmal Taher and a wide range of other contributors from the Ministry of Health, other government agencies, and development partners in Jakarta. The Review Team would like to acknowledge the importance of the leadership of Deputy of Human Resources and Cultural Affairs Dra. Nina Sardjunani, supported by Director of the Health and Nutrition Department Dr Hadiat and Ir Pungkas Bahjuri Ali, in integrating the review recommendations into the 2014-2019 National Mid-Term Development Plan (RPJMN 2014-2019). The Review Team would also like to acknowledge the support of the HSR Secretariat led by Dr Arum Atmawikarta and secretariat staff Ms. Chiquita Abidin and Ms. Fitri Inayati, all of whom were guided by the team leader of the implementing service provider for the AIPHSS program, Ahmer Akhtar.

3

Table of Contents Executive Summary ..................................................................................................................... 12 1.Achievements so far.................................................................................................................. 22 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12

Introduction ....................................................................................................... 22 Social determinants of health ............................................................................ 23 The relationship between institutional arrangements and health outcomes ....... 24 Health sector oversight and governance ........................................................... 25 Cross-sector relationships, coordination and accountabilities ............................ 28 Transactions...................................................................................................... 30 Health care delivery ........................................................................................... 31 Equity and access ............................................................................................. 42 Health expenditure ............................................................................................ 43 Political commitment ......................................................................................... 44 Health sector achievements .............................................................................. 45 Headline achievements ..................................................................................... 48

2. 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13

Remaining challenges ............................................................................................ 50 Introduction ....................................................................................................... 50 Social determinants of health ............................................................................ 50 The relationship between institutional arrangements and health outcomes ....... 50 Health sector oversight and governance ........................................................... 51 Cross-sector relationships, coordination and accountabilities ............................ 55 Transactions...................................................................................................... 56 Health care delivery ........................................................................................... 60 Equity and access ............................................................................................. 61 Health expenditure ............................................................................................ 63 Political commitment ......................................................................................... 63 Organisational development .............................................................................. 64 Service quality and standards............................................................................ 68 Summary ........................................................................................................... 69

3. 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.10 3.12 3.13

New and emerging challenges ................................................................................ 70 Introduction ....................................................................................................... 70 Social determinants of health ............................................................................ 71 The relationship between institutional arrangements and health outcomes ....... 71 Health sector oversight and governance ........................................................... 72 Cross-sector relationships, coordination and accountabilities ............................ 73 Transactions...................................................................................................... 75 Health care delivery ........................................................................................... 76 Equity and access ............................................................................................. 77 Health expenditure ............................................................................................ 77 Political commitment ......................................................................................... 80 Organisational development .............................................................................. 81 Service quality and standards............................................................................ 90 Mitigation of new and emerging challenges ....................................................... 90

4. 4.1 4.2

Targets to be achieved by 2019 .............................................................................. 98 Main issues ....................................................................................................... 99 Strategies ........................................................................................................ 101 4

5. 5.1 5.2

Policy options and strategies ................................................................................ 102 Recurrent strategic remedies........................................................................... 102 Policy options, strategy and indicators............................................................. 103

6.

Programmes ........................................................................................................ 110

7.

Indicators .............................................................................................................. 124

Annexes Annex 1: Landscape by flows..................................................................................................... 131 Annex 2: Stakeholders ............................................................................................................... 133 Annex 3: Contract types and how do we provide incentives to develop efficiency and quality? .. 159 Annex 4: Health care as a market (inc. market failure and gaming)............................................ 166 Annex 5: Determining the strength of individual Puskesmas ...................................................... 178 Annex 6: Conceptual model used to describe institutional relationships ..................................... 183 Annex 7: Summary of actions .................................................................................................... 185

5

List of Figures Figure 1: Contracted relationships................................................................................................ 19 Figure 2: WHO Health Systems Blocks and health outcomes ...................................................... 25 Figure 3: Outline structure of the SKN.......................................................................................... 27 Figure 4: Actors in the health care landscape .............................................................................. 29 Figure 5: Beds available ............................................................................................................... 34 Figure 6: Beds available/100,000 ................................................................................................. 34 Figure 7: The primary health care cube20 ..................................................................................... 38 Figure 8: Allocation of public health spending .............................................................................. 79 Figure 9: Road map: implementation of UHC ............................................................................... 88 Figure 10: Contracted relationships............................................................................................ 102 Figure 11: Institutional relationships ........................................................................................... 131 Figure 12: Ministry of Health context .......................................................................................... 133 Figure 13: Ministry of Health structure to 2014 ........................................................................... 134 Figure 14: Ministry of Health structure from 2014 ....................................................................... 135 Figure 15: Correlation between health expenditures and outcomes ........................................... 176 Figure 16: Analytical model ........................................................................................................ 183

6

List of Tables Table 1: High-level indicators against 2009 RPJMN targets ......................................................... 23 Table 2: Hospital stewardship ...................................................................................................... 32 Table 3: Hospitals by type18 ......................................................................................................... 33 Table 4: Human assets by broad category and location18 ............................................................ 35 Table 5: Human assets in Puskesmas18 ....................................................................................... 38 Table 6: Coverage of Health Insurance, 2012 .............................................................................. 43 Table 7: Health Minimum Service Standards .............................................................................. 46 Table 8: Headline achievements .................................................................................................. 49 Table 9: District/City income source in local budget (2010) .......................................................... 56 Table 10: Range of expenditure in 7 sample Provinces by disposal (2010) .................................. 72 Table 11: Consolidation of funds into BBJS ................................................................................. 86 Table 12: Policy options and strategies ...................................................................................... 103 Table 13: Selected priority programme areas for the MoH ......................................................... 110 Table 14: Selected priority programme areas for multiple agencies ........................................... 114 Table 15: Other programme actions ........................................................................................... 116 Table 16: Summary of proposed indicators for health systems governance ............................... 124 Table 17: Indicator 1, Governance: ............................................................................................ 125 Table 18: Matrix of indicators for RPJMN 2015-2019; Institutional Analysis ............................... 126 Table 19: Characteristics of the previous and new MoH structures ............................................ 140 Table 20: Ministry of Health features .......................................................................................... 143 Table 21: Main contract types .................................................................................................... 160 Table 22: Typical criteria for judging clinical quality .................................................................... 162 Table 23: Examples of models of health care system................................................................. 167 Table 24: Dimensions of universal coverage in 'high-performing' countries ................................ 177 Table 25: Illustration of how Puskesmas may be judged ............................................................ 178 Table 26: Definitions of the model components .......................................................................... 183 Table 27: Summary of actions.................................................................................................... 185

7

Abbreviations and Acronyms ABPD

Anggaran Pendapatan Belanja Daerah (Subnational Budget)

ADB

Asian Development Bank

ADINKES

Asosiasi Dinas Kesehatan (District Health Office Association)

API

Annual Parasite Index

ARSADA

Asosiasi Rumah Sakit Daerah (District Public Hospitals Association)

ARSANI

Asosiasi Rumah Sakit Nirlaba Indonesia (Indonesian Non-Profit Hospitals Association)

ARSVI

Asosiasi Rumah Sakit Vertikal Indonesia (Indonesian Vertical Hospital Association)

ASKES

Asuransi Kesehatan (Civil servant’s Health Insurance)

Badan POM

Badan Pengawas Obat dan Makanan (National Agency of Drug and Food Control),

Bappenas

Badan Perencanaan Pembangunan Nasional (National Planning and Development Board)

BinWas

Pembinaan dan Pengawasan (Guidance and Oversight Mechanism)

BKKBN

Badan Koordinasi Kependudukan dan Keluarga Berencana Nasional (National Population and Family Planning Board)

BLUD

Badan Layanan Umum Daerah (Sub-National Public Service Unit)

BOK

Bantuan Operasional Kesehatan (Operational Health Assistance)

BPJS

Badan Penyelenggara Jaminan Sosial (Social Security Agency)

BPOM

Badan Pengawas Obat dan Makanan (National Drug and Food Control Agency)

BPRS

Badan Pengawas Rumah Sakit (Hospital Supervisory Agency)

Bupati

Head of the District Government

CCM

Country Co-ordinating Mechanism

CCT

Conditional Cash Transfer

CHE

Catastrophic Health Expenditure

DAK

Dana Alokasi Khusus (Special Allocation Fund)

DFAT

Department for Foreign Affairs and Trade, Government of Australia

DG BUK

Directorate General Bina Upaya Kesehatan (DG of Health Services)

DHO

District Health Office

DJSN

Dewan Jaminan Sosial Nasional (National Social Security Council)

DRG

Diagnosis-Related Group

GAVI

Global Alliance for Vaccines and Immunisation 8

GDP

Gross Domestic Product

GFATM

Global Fund to Fight AIDS, Tuberculosis and Malaria

GIZ

Gesellschaft für Internationale Zusammenarbeit

GoI

Government of Indonesia

GP

General Practitioner, General Practice

HCIS

Human Capital Information System

HMIS

Health Management Information System

IDAI

Ikatan Dokter Anak Indonesia (Indonesian Paediatrician Association)

IDB

Inter-American Development Bank

IDB

Islamic Development Bank

IDHS

Indonesia Demographic and Health Survey

IDI

Ikatan Dokter Indonesia (Indonesian Medical Association)

IHR

International Health Regulation

IMR

Infant Mortality Rate

Jamkesda

Jaminan Sosisal Kesehatan Daerah (Regional Government’s Health Insurance)

Jamkesmas

Jaminan Kesehatan Masyarakat (Community Health Insurance)

Jamsostek

Jaminan Sosial Kesehatan (Workers’ Social Security)

JKN

Jaminan Kesehatan Nasional (National Health Insurance)

KARS

Komite Akreditasi Rumah Sakit (Hospital Accreditation Committee)

LPUK

Lembaga Pengembangan Uji Kompetensi (Competency Test Development Agency)

MDG

Millennium Development Goal

MMR

Maternal Mortality Rate

MNCH

Maternal, New-born and Child Health

MoE

Ministry of Education

MoF

Ministry of Finance

MoHA,

Ministry of Home Affairs

MPW

Ministry of Public Works

MSS

Minimum Service Standards

MSW

Ministry of Social Welfare

MTKI

Majelis Tenaga Kesehatan Indonesia (Indonesian Health Workforce Assembly)

NCD

Non-Communicable Disease

NHIC

National Health Information Centre 9

NSPK

Norma, Standar, Prosedur dan Kriteria (Public Service Standards)

PAN Kemenag

Kementerian Negara Pendayagunaan Aparatur Negara dan Reformasi Birokrasi (Ministry for Administrative and Bureaucratic Reform)

PDGI

Persatuan Dokter Gigi Indonesia (Indonesian Dentist Association)

PERSI

Persatuan Rumah Sakit Indonesia (Indonesian Hospitals Association)

PHBS

Pola Hidup Bersih dan Sehat (Clean and Healthy Lifestyle Programme)

PHC

Primary Health Care

PHO

Provincial Health Office

PKH

Program Keluarga Harapan (Family Hope Programme)

PNPM

Program Nasional Pemberdayaan Empowerment Programme)

PODES

Potensi Desa (Village Potential Survey)

Poskesdes

Pos Kesehatan Desa (Village Health Post)

Posyandu

Pos Pelayanan Terpadu (Integrated Health Post)

PPNI

Persatuan Perawat Nasional Indonesia (Indonesian Nurse Association)

ProLegNas

Program Legislatif Nasional (Programme for National Legislation)

PTT

Pegawai Tidak Tetap (contract physician scheme)

Puskesmas

Pusat Kesehatan Masyarakat (Community Health Centre)

RISKESDAS

Riset Kesehatan Dasar (Basic Health Research Survey)

RPJMN

Rencana Pembangunan Jangka Panjang Nasional (National Long-Term Development Plan)

RS

Rumah Sakit

SJSN

Sistem Jaminan Sosial Nasional ( National Social Health Insurance)

SKN

Sistem Kesehatan Nasional (National Health System)

SLA

Service Level Agreement

SOE

State-Owned Enterprise

SOP

Standard Operating Procedure

SPM

Standar Pelayanan Minimum (Minimum Service Standards)

SUSENAS

Survei Sosial Ekonomi Nasional (Household Socioeconomic Survey)

TB

Tuberculosis

TNI

Indonesian Military

TFR

Total Fertility Rate

TNP2K

Tim Nasional Percepatan Penanggulangan Kemiskinan (National Team for the Acceleration of Poverty Reduction)

UHC

Universal Health Coverage

UKBM

Upaya Kesehatan Bersumberdaya Masyarakat (Community-Based Health

Masyarakat

(National

Community

10

Efforts) UKS

Usaha Kesehatan Sekolah (School Health Programme)

UNDP

United Nations Development Programme

UNICEF

United Nations Children's Fund

UPT

Unit Pelaksana Teknis (Technical Implementation Unit)

USAID

United States Agency for International Development

Walikota

Mayor of a City Government

WASH

Water, Sanitation and Hygiene

WB

World Bank

WHO

World Health Organisation

11

Executive Summary 1. Headline Achievements Stimulating both demand- and supply-side factors in the health economy, producing sustainable growth in access to services, is an important policy direction in Indonesia’s health economy going forward. This policy is being driven by the combination of the creation of new organisations/systems such as the Social Security Agency (BPJS), legislative interventions designed to enrich access to health services, the adoption of universal health coverage (UHC) by 2019, and the planned increase in the supply of skilled professionals working in the health care system. Fundamental to this, free access to many health providers, both private and public, together with a comprehensive benefits package, has made community health insurance (Jamkesmas) attractive to the majority of people in Indonesia. It would appear that the informal culture around the health-care landscape has crystallised in a number of ways. First, at a political level, commitments have been made to expand public health care substantially, apparently with minimal political resistance. As a result, the plan to introduce UHC by 2019 has attracted a high degree of political consensus. Second, at the high level of the executive branch, acknowledgement of the importance of refocusing the attention of the Ministry of Health (MoH) on a more strategic level has been understood. At the service delivery end of the national health system (SKN), moves to professionalise health-care management are underway, and the private and public sectors are seeking opportunities to collaboration in mutually beneficial ways. National health insurance (JKN) will become the major player in the reimbursement of health providers for delivering personal health-care services. The MoH plans to lead a strengthening of public/population health care and is hoping to achieve this objective in a number of ways. First, the MoH needs to be reorganised in line with Presidential Decree No. 81/2010, in order to develop an organisational structure that is more “fit for purpose” in terms of scale, mix of functions, and better reflects the new partnerships arising from the review of the decentralisation framework and the expansion of BPJS. Deliberations around the new structure are complete and the resultant plan has been submitted to the Ministry of Administrative and Bureaucratic Reform (PAN Kemeneg) for endorsement. In addition, a new Directorate General (DG) of Public Health will be established (still in process for endorsement by PAN Kemeneg). The previous DG (DG Gizi and KIA-based on MoH Decree No. 1144/Menkes/Per/VIII/2010) was also responsible for Nutrition, Maternal and Child Health affairs. Second, a new national health system (SKN) has been developed in Presidential Decree No. 72/2012 that promotes primary health care as a building block of the Indonesia health-care system. In addition to these changes, Indonesia has achieved substantial progress during the life of the 2009 RPJMN, evidence of which can be seen in several forms listed below. Institutional arrangements under decentralisation 12



The redistribution of health functions between tiers of government was agreed and issued through PP No. 38/2007 and the MoH has since developed public service standards (NSPK) as guidance for provinces, districts/cities in the execution of their delegated health functions. However, these NSPK have not been fully implemented as yet, which is explained in the section on “Remaining Challenges”.



Permenkes No. 741/2008 on minimum service standards (SPM) has been issued and defines 18 standards that district/city governments must adhere to.



Health institutions at national and sub-national levels have been reorganised. The MoH has been reorganised based on MoH Decree No. 1144/Menkes/Per/VIII/2010 following Presidential Decree No. 81/2010, with the aim of mitigating the anomalies arising from the tension between supervising the delivery of personal health care and public/population health care, by forming the DG of Health Services (DG BUK)The result of this restructuring will be a clearer and stronger referral mechanism from Puskesmas to Hospitals, clearer stewardship of institutional health services at provincial and district/city levels, a clearer structural organisation for hospitals in the provinces and districts/cities, and a clearer structural organisation for Puskesmas at the district/city level. These reforms are designed to introduce greater clarity at all levels of the SKN.

Development of a road map for JKN 2012-19 and the introduction of UHC Indonesia has made substantial progress on UHC through the establishment of a clear policy framework. The Government passed two key laws: (i) Law on the National Social Security System No. 40/2004; and, after considerable delay, (11) Law on Establishing the Social Security Agency (BPJS) Law No. 24/2011 in October 2011. Health insurance for the poor and near-poor (Jamkesmas) has been expanded to reach 76.4 million people (32 percent of the population). With the passing of the second law (Law No. 24/2011), the Government has developed a guide for the implementation of UHC. This is known as the Road Map towards National Health Insurance Universal Coverage (NHIC) 2012-2019. Particularly important is the creation of the Social Security Agency (BPJS). Significantly, the road map targets make explicit reference to enriching community, patient and staff experience within the system. “At least 75 percent of the members shall be satisfied with the services provided at the NHIC and by health care providers contracted by the NHIC” by 2014 (85 percent by 2019), and “At least 65 percent of the health workers and health care providers shall be satisfied with and/or receive appropriate payment by the NHIC” by 2015 (85 percent by 2019)1, and

1

Note that the quotes have been changed slightly to correct grammar but the intention remains in accordance with the original text.

13

“The NHIC will be managed in a transparent, efficient and accountable manner.” These statements represent a determination to alter the cultural framework within which the SKN operates. The roadmap towards NHIC 2012-2019 contains many targets and milestones; some are in progress, some have still to be achieved, but many have been achieved. Growth of BPJS In terms of organisational development, BPJS has been established, existing national social protection schemes are being consolidated into the BPJS portfolio, and local schemes will be inducted in due course. BPJS is displaying a degree of technocratic professionalism and drive, has expanded the number of people covered and is working to stimulate demand and provider quality. Systems have been developed that enable rapid and high volume member recruitment and also produce real-time performance and activity reporting. Financial commitment to health The GoI has expressed its commitment to health programmes through an increase in expenditure. In absolute expenditure terms, per capita spending in 2011 increased from US$16.4 to US$32.4 (public), and from US$98.3 to US$126.9 (total). Supply-side readiness Efforts through BPJS, MSS/SPM, NPSK and village health posts (Poskesdes) have all increased the drive to achieve a health-care system more focused on primary care and the delivery of quality services in the right place, at the right time, and deploying the appropriate level of expertise. Growth in the supply of services (both primary and secondary) has also been seen through an increase in available in-patient beds, an increase in the number of clinical practitioners, and an increase in clinical learning opportunities. Headline achievements • • •

• •

Increase in the number of citizens insured. The creation of new organisations/systems to stimulate demand (BPJS/UHC). Legislative interventions designed to enrich access by, paving the way towards UHC by 2019, and increasing the supply of skilled professionals in the health-care system. These interventions have the effect of stimulating both demand- and supply-side factors in the health economy, producing sustainable growth in access to services. In absolute expenditure terms an increase in per capita spending in 2011 from US$16.4 to US$32.4 (public), and US$98.3 to US$126.9 (total).2 Free access to many providers, both private and public, and a comprehensive benefits package, have made Jamkesmas attractive to the majority of Indonesians.

2

WHO (2014), http://apps.who.int/gho/data/node.country.country-IDN

14



• •

At the political level, commitments have been made to expand public health care substantially, with minimal political resistance. The plan to introduce UHC by 2019 has attracted a high degree of political consensus. At the high level of the executive branch, acknowledgement of the need to refocus the attention of the MoH at a more strategic level has been understood. BPJS is displaying a high degree of technocratic professionalism and drive.

As can be seen above, significant achievements have been recorded in all aspects of the SKN architecture in response to the needs and challenges determined by the environment in which the SKN operates. However, more remains to be done. 2. Challenges both Remaining and Emerging Remaining challenges are largely about: (i) equity (geographical, health status, economic and access), about capacity to absorb growth (both supply and demand), (ii) the clarity and focus of the roles of institutions, (iii) technical capacity (information systems, managerial and clinical skills), (iv) accessing “hard-to-reach” groups, (v) co-ordination between institutions, and (vi) engaging with the private sector. A recurring theme of this analysis is that many obstacles prevent the implementation of highlevel policy in practice and in the field, through the absence or weakness of systems, data and capacity. The policy framework is clear, but the collateral impact of larger structural characteristics affects the mechanisms available for the SKN to implement policies. In particular, this refers to decentralisation and its effect on the policy objectives of achieving equity. Furthermore, the relative status of the various institutions involved in the protection of health and the delivery of health care, limits the ability of the lead political institution, the Ministry of Health (MoH), to determine how providers are reimbursed, accredited, supervised and governed. Illustrations of these obstacles which appear in the main text include: “The MoH, first has very few levers of influence other than regulatory instruments or financial sanctions and, second, the decentralised structure militates against direct intervention in operational affairs.” “The BPJS is tasked with delivering UHC by 2019. This seems an ambitious target, based on the scale of work required to register an additional 100 million people, the majority of whom operate in the informal sector.” “There are repeated references to a lack of clarity, and both overlap and conflict of roles, between tiers both at the governance level and at the service delivery level.” “Puskesmas are expected to achieve 100 percent accreditation by 2019 to secure recognition of quality-assured service provision. The only sanction for noncompliance to quality standards would be withdrawal of services from the Puskesmas. If this sanction were exercised, many communities would be left without access to care of any kind. Since the accrediting agency is also the body responsible for Puskesmas, it is unlikely that this sanction would ever be applied in practice.”

15

Institutional arrangements under decentralisation There is a lack of clarity concerning the relationship between the MoH and district/city health authorities. Whilst efforts to resolve this issue are ongoing, they have not yet wholly resolved the anomalies. The tension between local and national priorities is still present, and manifests itself in the lack of conformity of local authorities in observing central programmes using local financial resources. Execution is often limited to programmes that attract funding from central sources. In addition, the MoH frequently undertakes functions that are supposed to have been delegated to the local authorities, adding to the confusion around boundaries, responsibilities and accountabilities. The absence of operating procedures that govern the relationship between the MoH and provincial health offices/district health offices (PHOs/DHOs) creates tensions between local and national priorities. It is often unclear where authority can or should be exercised. a. Relationship between the MoH, BKKBN, BPOM, BPJS and other health-related ministries/agencies at the sub-national level: Branches of the National Population and Family Planning Board (BKKBN) exist at the subnational level. However, following decentralisation, many have been merged with existing related institutions. The working relationship with PHOs/DHOs is limited to a coordinating relationship that is highly dependent on the (inconsistent) leadership of each institution. Similarly, the National Drug and Food Control Agency (BPOM) is represented down to the province level and practical relationships vary widely. Other health-related ministries/agencies are represented down to the district/city level, and have functional coordination with PHOs/DHOs, but in practice this is poorly implemented. b. Guidance and oversight mechanism (Binwas) exercised at the sub-national level PP No. 79/2005 was developed to enable guidance and supervision from central to local government levels, but the MoH and MoHA have no coordinating mechanism to implement guidance and oversight mechanisms (Binwas) in local governments. The practical application of leverage through legislative and social controls remains weak. The role of the provincial governor as the representative of the central government exercising Binwas in district/city governments remains anomalous, and the absence of clear and unequivocal sanctions that could be applied to districts/cities failing to conform to national policies hinders the effective use of Binwas. Development of the road map for JKN 2012-2019 and introduction of UHC The relationship between BPJS and sub-national governments is problematic. There is no obvious mechanism to harmonise reimbursements in the wider health economy, or to deploy BPJS engagement in the development of local strategies to deliver community-wide health gains. Likewise, there appears to be no mechanism to mitigate the risk of the compression of fiscal space or an excess of funds in the BPJS risk pool. If the BPJS risk pool becomes decapitalised during a financial year, there seems to be no mechanism to correct for the shortfall. Equally, if the mechanism raises more assets in the pool than can be usefully 16

disbursed, there is no obvious mechanism to adjust premiums downwards. The former eventuality is more likely. Given that the premium will be a fixed rate amount, not adjusted for risk, and one that will be made compulsory, the premium is tantamount to a tax. There is a well-known “insurance effect” whereby subscribers (whose risk is not assessed) seek to gain benefits from their expenditure and so seek a return on their investment. This effect is especially prevalent amongst those who perceive that they are low risk and so are not getting value for money from their payments. The capacity to absorb growth in both demand and supply has not been risk-assessed, so it is not known whether systems, competence, willingness or capability exist to absorb the impact of UHC within a decentralised framework. Growth of BPJS BPJS branches are found at both the provincial and district/city levels. However, the functional relationship is more with hospitals and Puskesmas (as direct service providers) than with PHOs or DHOs as supervisory agencies. There remains an unclear relationship between the MoH, PHOs and DHOs on the one hand and BPJS on the other. In terms of absorptive capacity, BPJS is registering, on average, about 20,000 members per day. Its priority in the short term is to successfully consolidate existing schemes and recruit a greater share of established schemes. The remaining challenge up to 2019 will be to recruit 100 million members who are not currently enrolled in any scheme, and to ensure the stable and predictable payment of premiums. This is an extremely challenging goal and one that may not be achievable in practice. Financial commitment to health The flow of financial resources between elements of the health-care sector, and the framework determining the value of the resources, is opaque. Whilst theoretically there is a population-based allocation system, in fact a large share of the allocation is a product of historical and/or arbitrary agreements. The situation will become less predictable in the future as an activity-based reimbursement mechanism is introduced from 2014, implying post-hoc expenditure (at least for marginal costs) allied with a population-based capitation system for primary care. Health-care markets are notoriously inelastic, where need does not necessarily translate into demand, demand does not necessarily translate into supply, and supply does not necessarily translate into price- and cost-control levers. As a consequence, markets may not balance supply, demand and price effectively. The immediate impact of this inelasticity is an increase in demand-side forces, expressed as growth in the use of private-sector services. This is allied with the increased registration of beneficiaries in the various social welfare programmes creating expanding health-care demand. This means that the health-care system must be able to absorb a greater volume and co-ordinate and manage a developing mixed-economy of care.

17

The current conditions of increasing demand (created by UHC, the epidemiological shift and other factors) and increased supply (created by strong private-sector growth, capital investment in secondary care and by the development of Puskesmas) will lead to increased global costs. Whether Indonesia has sufficient fiscal space to absorb that growth in demand within the planned increase in expenditure remains to be seen. Supply-side readiness Primary and ambulatory care still represents a low share of total health expenditure. More than 60 percent of all funds are spent at the in-patient level, compared with 30 to 45 percent in most OECD countries where more care is provided at the primary and out-patient level. Indonesia’s health system outputs and inputs are relatively low. The DPT3 immunisation rate in Indonesia in 2013 is low in contrast to the overall the East Asia Pacific region.3 A similar pattern is observed in skilled birth attendance rates, which are far lower than the average for the region and for lower middle-income countries. In terms of physical health system inputs, Indonesia has a low ratio of doctors and hospital beds per 1,000 population compared with its regional peers. The unclear relationship between PHOs and DHOs is most acutely seen in the referral mechanism. There should be a clear procedure regarding when patients should be handled in DHO facilities and when they should be referred to PHO facilities. A clear mechanism set under guidance from the MoH would help to avoid an imbalance in patient referrals between facilities. As primary care is developed, and direct financing is implemented, there is a risk that “weaker” Puskesmas will have no mechanism to rectify their limitations. Equally, BPJS will have no mechanism to judge quality. Therefore, it is in the interests of central government, sub-national governments, health service commissioners (including BPJS and government specifiers of what services are to be provided) and providers, to be able to judge the vulnerability of Puskesmas that do not have sufficient scale to be autonomous (unlike hospitals). Failing this, such Puskesmas will be unable to assume the expanded and developed role that national policy envisages. Bearing in mind that service distribution is already inequitably distributed throughout the country, and that areas of poorer public provision are generally (but not entirely) those that also experience poorer levels of private provision,4 there is a risk of a widening gap between supply and demand. The impact of this is twofold. First, those who are most in need will have a relatively poorer experience of access to care compared with expectations and, second, insurance premiums will produce less benefit/value for those in the less wellprovided areas. What is necessary is an increase in capital investment and an increase in the deployment of human assets, together with enabling/encouraging the private sector to develop services in under-served areas.

3

WHO. National Institute of Health Research and Development, Ministry of Health, Indonesia & World Bank (2013), Assessing Supply-Side Readiness for Universal Health Coverage in Indonesia, Jakarta, Indonesia 4

18

Headline challenges Many suggestions and recommendations are presented throughout this analysis, including issues around leadership development, procurement practices, information management and institutional assessment. However, two themes appear frequently. The first is on how the MoH exercises a leadership role in the context of a decentralised constitutional structure, and how bilateral relationships can be managed in a transparent way (with the MoE, BPJS, the MPW etc). The second is on how governance is conveyed down through a hierarchical structure from the centre to village-level, whilst respecting the legitimacy of provincial and district/city parliaments. The solution suggested for the first issue is through a network of contract and service-level agreements (contracts with health providers, and service-level agreements with other government entities, on the grounds that government cannot contract with itself). These would be implemented via a joint ministerial decree, and negotiated through co-ordinating ministers (see Section 5 on Policy Options and Strategies). The benefits of such a solution would be as follows: 1.

Contracts will have the effect of aligning the three up-to-now misaligned systems, namely information, money and authority flows.

2.

Such a solution is consistent with the principles of decentralisation, a unified constitutional form, and with the structure of the government executive. This is because contracts/SLAs are the product of negotiated agreements, not the blunt instrument of the exercise of authority. As such, contracts/SLAs do not compromise the role of sub-national representative governments.

3.

Such a solution is politically acceptable both at sub-national and central levels.

The solution suggested for the second issue is for the requirement of autonomous/semiautonomous government entities (e.g., hospitals, DHOs, PHOs) to prepare tri-annual business plans and annual operational plans for submission up to the next hierarchical tier. Figure 1: Contracted relationships

Service agreements Contracts

level

Business plans

The benefits of such a solution would be as follows: • • •

The hierarchical principle is retained Such a solution creates an opportunity to promote quality and equity. Such a solution promotes continuing dialogue with a manageable span of control.

19

3. Targets to Be Achieved by 2019 Targets are organised into four areas, all of which are associated with institutional form, as follows: • • • •

Governance and system-level Input-level Output-level Outcome-level

The governance and system-level targets are aimed at securing the major policy objectives of the MoH, namely confronting NCDs and infectious diseases, achieving equity, and providing accessible services to all citizens. The targets require the GoI to put in place policies and plans to provide clear guidance on the achievement of these objectives, and that the GoI focuses on the creation or updating of plans and policies. These targets also require the creation, updating or clarification of relationships, information sources and guidance. The input-level targets focus on the resourcing, capacity, and structure of the SKN. These targets are concerned with human resources, the stability of the risk pool, and the existence of guidelines. These are all targets that reflect stability and institutional responses to prevailing conditions. The output-level targets reflect issues around access/coverage, efficiency and resource quality and distribution, and assume that outcomes are at least partly a product of outputs of the system and which in turn are partly a product of inputs guidelines. If the logic chain above is reliable, then the outcome-level targets can be presented as targets that secure access to high-quality services by all citizens. If these targets are achieved, then it will be possible to confidently assert that the mechanisms implemented are producing the desired results. The main targets are as follows: • • • • • • • •

5

Full implementation of UHC. Growth in annual public health expenditure to 4 percent of GNP. Growth in primary care and public health expenditure to levels typical of the Asia Pacific Region. Extension of sub-national public service units (BLUD) to Puskesmas. Investment in Puskesmas to go towards refurbishing the proportion of 38 percent of centres in a moderately or severely damaged condition.5 Deploy physicians in the about 2,000 Puskesmas that do not have any in place (PODES, op. cit). Create Puskesmas in the 430 sub-districts that have none. Reduction of management costs to around 10 percent of total expenditure.

PODES, 2011

20

• • •

Fifty percent of primary care to be delivered by private sector providers. All PHOs and DHOs to prepare business plans. Service-level/performance agreements in place between the main actors in the healthcare landscape.

21

1.

Achievements so far

1.1

Introduction

A number of significant achievements have been recorded since the 2009 RPJMN. However, it is difficult to attribute any causal relationship between health outcomes and organisational/institutional reform and development. More relevant have been the changes recorded from the implementation of the decentralisation legislation since 2000. Organisational/institutional reform is an instrument to enable other parts of the health-care sector to perform in a more efficient, more effective way, or to produce greater access to, or quality of, care. The 2009 RPJMN identified health care as being amongst a raft of deserving areas for organisational/institutional reform. Within the commentary, the following references appeared in the 2009 RPJMN: Efforts to increase health services have been continuous. In 2005, the budget allocation for health reached Rp 7.7 trillion while in 2008 it reached Rp 17.9 trillion. Most of the additional budget allocation was used for enhancing health services at public health centres (Puskesmas) and at integrated services centres (Posyandu) that are financed by amongst others the community health insurance (Jamkesmas), as a program aimed to serve disadvantaged communities. In 2008, this program was able to provide health services to 76.4 million persons. In order to broaden the access of the general public to medicines, steps were taken to significantly lower the price of generic medicines. A portion of the continuously increasing health budget allocation was used to recruit new medical doctors and paramedics and to assist regional governments to rehabilitate Puskesmas and to build new hospitals in various regions. A portion of the budget funds was also used for the Program to Revitalize Family Planning that had been somewhat neglected in the initial period of reform and decentralization. To a degree, the expansion of the health sector has yielded various results, such as the decline in the prevalence of malnutrition amongst children. ….Such a foundation is reflected in the implementation of the programs on the Jamkesmas (Social Health Insurance) The reallocation of oil fuel subsidies to education and health programmes in 2005‐08 is also evidence of such endeavours. ….conditional cash transfers for the financing of education and health will be provided for very poor communities For the future, the 2009 RPJMN declared: Development of health will aim to increase the access of people to health services, indicated amongst others by an increase in life expectancy rates, a decline in the infant mortality rate, and a reduction in the maternal mortality rate.

22

Table 1: High-level indicators against 2009 RPJMN targets Indicator

2009

Achievement

Target in 2014

2010

2011

2012

2019

Life expectancy

70.7

70.9

71.1

71.1

72.0

Maternal mortality rate (MMR)

228

n.a

n.a

n.a

% of trained health workforce delivery

84.3

84.8

86.38

88.64

90

Infant mortality rate (IMR)

34

34

34

32

24

Total fertility rate (TFR)

2.6

2.4.

n.a

2,6

2.1

Access to clean water

47.7

44.19

42.76

na

68

HIV prevalence

66.2

57.5

n.a

79.5

90

API (Annual Parasite Index) decline

1.85

1.96

1.75

1.69

1

% community with health insurance

n.a

59.1

63.1

64.58

80.10

359

118

The highlighted indicators above can be strongly linked with organisational/institutional actions. There have been many policy initiatives over the past decade that were either designed to increase access, quality or demand for health-care services, or that improved health and health care as a collateral impact of those initiatives. Amongst the raft of reforms are some critical legislative initiatives, including the following: • • • • • • • •

Decentralisation Laws in 1999 and 2004 National Social Security System Law No. 40/2004 Medical Practices Law No. 29/2004 Health Law No. 36/2009 Social Welfare Law No. 11/2009 Hospital Law No. 44/2009 Social Security Providers Law No. 24/2011 Medical Education Law (still under discussion in parliament)

Apart from these and other legislative interventions, the landscape in which the SKN operates has also altered in terms of the societal, technical, economic and political features prevailing. As a result, the internal operations of the SKN in terms of its organisational arrangements, its tasks and transactions, its capacity and its culture, have also had to adapt. 1.2

Social determinants of health

Many factors in the health landscape have an impact on health outcomes, or on healthsector functioning. Amongst these are the social determinants of health, the direct impact of water and sanitation and hygiene improvements (WASH), nutrition improvements, and 23

indirect factors, such as girls’ literacy (impact on fertility),6 transport infrastructure (for access to services), 7 and employment/economic security (for general well-being and financial access).8 The social determinants of health are defined as: “… the circumstances in which people are born, grow up, live, work and age, and the systems put in place to deal with illness. These circumstances are in turn shaped by a wider set of forces: economics, social policies, and politics.”9 In Indonesia, these wider forces mentioned include rapid population growth, a population with a reducing median age, changing patterns of employment, and an economy with a widening wealth gap. The impact of these changes in the context of health and health care on the demand-side is that they will increase the risk of serious illness and premature death due to health inequity and the poor becomes poorer due to catastrophic expenditure. Meanwhile, on the supplyside these changes will increase the burden on the health system to provide the required health facilities to meet the demand. Directly influencing many of the social determinants of health in Indonesia is largely beyond the reach of the MoH. However, the MoH, in partnership with other agencies, can influence the supply-side of health care, and some aspects of the demand-side forces. Consequently, the MoH can help to create an environment and structure that enable a greater degree of alignment between supply and demand. 1.3

The relationship between institutional arrangements and health outcomes

The reason why institutional arrangements are significant in terms of achieving positive health outcomes is well illustrated by the WHO.

6

Akmam, W (2002), Women's Education and Fertility Rates in Developing Countries, Eubios Journal of Asian and International Bioethics 12 (2002), 138-143. 7 World Bank (2008), Transport for Health Access, http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTTRANSPORT/EXTTSR/0,,contentMDK:20835940~menuPK:22747 62~pagePK:210058~piPK:210062~theSitePK:463716,00.html 8 Stronks, K et al (1997), The interrelationship between income, health and employment status, Int. J. Epidemiol. (1997) 26 (3): 592-600. doi: 10.1093/ije/26.3.592 9 WHO (2011), Rio Political Declaration at the World Conference on Social Determinants of Health, Rio de Janeiro

24

Figure 2: WHO Health Systems Blocks and health outcomes

10

The governance of the entire national health system (SKN and related partners) is a concept that aggregates the many ways individuals and institutions, public and private, manage their common affairs. It is a continuing process through which conflicting or diverse interests may be accommodated and cooperative action may be taken. It includes formal institutions and regimes empowered to enforce compliance, as well as informal arrangements that people and institutions either have agreed to, or perceive to be in their interest.11 The exercise of governance involves ensuring strategic policy frameworks exist and are combined with effective oversight, coalition-building, regulation, attention to system design and accountability. 1.4

Health sector oversight and governance

The health system in Indonesia is delivered and planned at three levels of government: central, provincial, and district/city. Based on Government Regulation No. 38/2004, the authority of each level is arranged within the concurrent principle. The institutional architecture of the national health system (SKN) is organised through a central and provincial/district/city health system. Based on the Decentralisation Law of 1999 (replaced in 2014 by Law No. 23 on Regional Government Autonomy), health is a decentralised sector. Depending on the function of government, there are at least three functions: policy-making and regulation, financing, and service delivery. The main actors in the system and the roles

10 11

WHO (2011) Commission on Global Governance (1995)

25

and responsibilities they fulfil in the overall governance/management structure at central, provincial, and district levels are as follows: At the central level: Health matters are planned in various agencies. In the executive, several units are involved in health policy and management: the Ministry of Health, under the Office of the Vice-President the National Team for Accelerating Poverty Reduction (TNP2K), the National Development Planning Agency (Bappenas), the National Population and Family Planning Board (BKKBN), and the Social Security Agency (BPJS). The MoH, in its role as the RPJMN implementing agency, operates in a complex environment. It might seem that the MoH is bureaucratic and detached from the processes of health-care delivery, but it is the only direct instrument of ministers and the nodal point where the political system interacts with health. The key question is whether or not the bureaucracy helps or impedes the technocracy. In the MoH Strategic Plan there are six main strategic directions. Strategy No. 6 is as follows: “Improve accountability, transparency, efficiency and effectiveness of the health system management to strengthen decentralised health systems.”12 Some health-service activities are financed through other ministries. Based on MoF data, the MoH received about 60 percent of the total government budget for health in 2012. The share of the MoH budget will decrease in the coming years because the social insurance budget will be moved to BPJS. In addition, the Ministry of Public Works (MPW) is allocated around 12 percent of the national budget for health devoted to water and sanitation. The MoH has no direct authority over district/city governments. However, various central budgets held by the MoH are disbursed to public hospitals, Puskesmas, some centrallygoverned services and to the PHOs and DHOs. Technical regulation in health is developed by the minister of health. However, all technical ministries including the MoH have the authority (and duty) to provide technical guidance and supervision to local governments in delivering their autonomous functions (Law No. 32/2004 on Regional Government). The final responsibility for health matters rests with the President through the minister of health. It is the responsibility of the MoH to provide guidance, supervision and technical facilitation to district/city governments, especially the health Dinas. In practice, this guidance is rarely exercised by the MoH since the MoH mainly focuses on its own programmes (even though most of these are becoming the functions of local government). At the provincial level: The provincial government’s governor is elected by popular vote.13 The governor has strong executive powers on health matters. The Office of Health and Family Planning resides within provincial government. The legislative instrument is the provincial parliament, whose members are elected by popular vote. The executive and legislative may pass provincial regulations on health matters, while the governor is empowered to issue governor regulations.

12

MoH, 2010 [paraphrased] Although at the time of writing changes are afoot in the National Parliament (DPR) to revert these elections from direct to indirect through provincial parliaments.

13

26

At the district/city level: District/city governments are led by district/city government leader (bupati/walikota). 14 Health affairs are managed by the District Health Office (DHO) in district/city governments and also by the Family Planning Unit. The legislative instrument is the district/city parliament (members are elected by popular vote). Local government regulations are set by the district/city parliament and the district/city government. Local government plans and executes the annual budget with district/city parliamentary approval. In terms of direct management, the MoH manages Class A hospitals and strategic workforce issues, PHOs manage Class B hospitals and DHOs manage Class C and D hospitals and community health centres (Puskesmas). In summary, the vertical relationships within the health sector in Indonesia are as follows: Figure 3: Outline structure of the SKN

14

Although at the time of writing changes are afoot in the National Parliament (DPR) to revert these elections from direct to indirect through district/city parliaments.

27

1.5

Cross-sector relationships, coordination and accountabilities

Relationships and accountabilities can be characterised by inspecting financing, legislative and political relationships and authorities. In terms of the financing function a number of sources are involved including: • • • • • •

Central Government (Ministry of Finance) Social Security Agency (BPJS) Provincial and district/city governments Community payment Private insurers/employers Out-of-pocket health expenditure

The legislative status of health care is captured by Commission IX in the National Parliament (DPR). Decisions resting at the national level (based on Law No. 12/2011) include the following: • • •

The 1945 Constitution (Undang-Undang Dasar Negara Repulik Indonesia Tahun 1945); Parliamentary Decision (Ketetapan Majelis Permusyawaratan Rakyat); Government Regulation (Undang-Undang/Peraturan Pemerintah Pengganti Undang-Undang).

Government regulation can also act as a replacement of a law (in lieu of law); • •

Government Regulation (Peraturan Pemerintah); and Presidential Regulation (Peraturan Presiden).

Politically, the National Parliament (DPR) together with the President (as executive head) formulate legislation. A draft law can be proposed by members of parliament through a Programme for National Legislation (Prolegnas) or by the executive (Presidential Office). In recent years, many national regulations in the form of laws have been stipulated by Parliament, such as the National Social Security System Law No. 40/2004, the Medical Practice Law in 2004, the Health Law in 2009, the Hospital Law in 2009, and the Social Security Providers Law No. 24/2011, and is currently in the process of developing medical education legislation. Based on these laws the various national health policies are codified as government regulations. The overall structure of the health-care system in a decentralised context at a macro-level is summarised in Figure 1.3 below. It is complex in the sense that many institutions are involved, but also because the relationships between them are ill-defined. Conceptually, the relationships are a natural product of decentralisation, cross-cutting issues and priority programmes. However, in practice, the translation of these concepts into practical working arrangements is problematic in terms of the flows of information, accountability and resources, and this has consequences that impact on clarity, efficiency and coordination. In the case described below (excluding sub-institutional actors), there are 190 possible 1:1 relationships, 6,840 possible 1:1:1 relationships and 116,280 possible 1:1:1:1 relationships. Under such conditions, any lack of clarity would so compromise the capacity of the whole 28

system to function effectively that a priority must be to find ways that the institutions can work better together. Adding to the challenges inherent in this arrangement are factors such as the 5-year plan for the introduction of UHC through BPJS up to 2019, a restructuring of the internal workings of the MoH, a revision of the decentralisation legislation, and the impact of the political turbulence and uncertainty created by elections in 2014. Figure 4: Actors in the health care landscape National Parliament

Central Governme nt

BKKE N

BPJ S

Badan POM

Ministry of Health

Ministry of Educatio n

Ministry of Home Affairs

Provincial Government

Provincial Parliament

Central Governm ent Hospitals

Family Planning Agency

Provincia l Health Office

Provincial Hospitals

District/City Government

Family Planning Agency

District Parliament

District Health Office

District Hospitals

District Communi ty Health Centres

Private Hospitals/ providers

29

There are other players in this landscape, including related government institutions (especially Bappenas, the MPW, the MSW), informal care-givers, practitioners of traditional medicine, professional associations, academic institutions, and health-care providers associated with employment (e.g., for the military) all of whom add to the complexity of the sector. The non-government sector on the demand-side of the health economy includes such agencies as patient/consumer groups (e.g., Yayasan Lembaga Konsumen Indonesia), and development partners (e.g., DFAT, USAID, GIZ, IDB, ADB, the World Bank, WHO, UNICEF, UNDP, etc.). These agencies have the effect of creating demand-side stimulus, including that arising from the private and other non-government sectors. Community empowerment and engagement Poskesdes are community-based health initiatives established under Upaya Kesehatan Bersumberdaya Masyarakat (UKBM). They are village-based and are intended to develop and provide basic health services for rural communities. A village health post (Poskesdes) can be regarded as a health facility that is at the interface between the input of the community and government support. Services in a Poskesdes may include preventive and curative (treatment) delivered by health workers (especially midwives) with the involvement of volunteers. The public health activities conducted within the village health post include the following: • • • • •

Simple epidemiological surveillance (especially infectious diseases), and identification of risk factors (including nutritional status), as well as ante-natal health. Curative activities, especially the management of infectious diseases, as well as reducing risk factors (including nutritional status). Preparedness and emergency response to health disasters. Basic medical services, according to their level of competence. Other activities include health promotion, nutrition awareness, an increase in positive health behaviour, and environmental health.

Poskesdes are organised by health workers (at least a midwife) and assisted by a minimum of two health workers (with different categories). The infrastructure includes buildings, equipment, and medical equipment. Evidence of the commitment to community empowerment is seen in the support for community-based health efforts (Upaya Kesehatan Bersumberdaya Masyarakat, or UKBM). However, improving community access to health care through social mobilisation, community empowerment, and improving the quality of health services, remains a challenge. 1.6

Transactions

A critical issue is the absence of alignment both between systems and also within systems. Flows of accountability, money and information are not aligned, so the effectiveness of planning, regulation and supervision is compromised. Equally, the flows of money, information and accountability are not harmonised through the system, so the mechanisms for planning and regulation lack instruments of implementation (i.e., money). Generally, it 30

would be expected that the three sets of relationships would be aligned such that flows are in parallel, enabling more effective monitoring, planning, supervision and budgeting. Arguably, the RPJMN and the MoH Strategic Plan offer the lubricant required to synchronise the three sets of relationships. At the very least, they do offer a united and shared framework and language. 1.7

Health care delivery

Public health and hospital service provision is directed not only by the MoH. Public health activity is decentralised and delivered at provincial and district/city levels. The PHOs in 34 provinces are responsible for technical oversight of the DHOs. Public hospitals are managed by the MoH, district/city governments, the military and police, and state-owned companies. One of the building blocks of a safe and effective health-care system is the provision of competent care by skilled professionals. In Indonesia, the registration of professional staff is conducted by several institutions: medical doctors and dentists by the Medical and Dental Councils (every five years) and other health professionals by the DHOs. The GoI established the Indonesian Health Workforce Assembly (MTKI) in 2011, which will gradually assume responsibility for the registration of other health professionals (17 professions) from the DHOs. The absence of a clear (or at least explicit) description of how the whole delivery system is envisioned hampers planning and coherent policy frameworks. There is limited clarity concerning the model of service delivery, how the public and non-government sectors complement each other, how comprehensive planning at a regional level can create balanced health economies, and how the financing mechanisms, the workforce planning mechanisms, and investment mechanisms, can capitalise on the opportunity created by the HSR, the RPJMN, and the revised decentralisation laws. 1.7.1

Hospitals

Secondary health care is provided at the district level (kabupaten [rural] and kota [urban], as well as at the provincial level. At the district/city level, secondary health care consists of Class C and D public hospitals, alongside private hospitals. Class D hospitals are headed by a general practitioner (GP) and provide some basic in-patient care (and are one step above Puskesmas). There is very little difference between the care provided at a Class D hospital and a Puskesmas with in-patient beds. Class C district hospitals provide four basic specialties (surgery, internal medicine, paediatrics and Ob/Gyn services), plus three supporting specialties of anaesthesia, radiological imaging and pathology. At the provincial level, Class B hospitals generally provide more specialist services. Specialist hospitals and specialist medical clinics such as pulmonary clinics, eye clinics and medical supporting care are also included amongst secondary health facilities. Tertiary health-care institutions are described variably as “centres of excellence” and/or Class A hospitals, which deploy more sophisticated equipment and facilities that are closer to being “state-of-the-art”. During the past 10 years, the Government has invested heavily in tertiary health-care facilities, such as maternal and child hospitals, oncology hospitals, cardiac care hospitals, and equipped central-referral hospitals (Classes A and B) with more 31

sophisticated devices. Class A hospitals providing “super specialties” have advanced specialist clinical expertise and technology, and are also considered Class A tertiary services. These tertiary facilities are found at the central level (Jakarta), as well in Medan (North Sumatra), Surabaya (east Java), and Makassar (South Sulawesi). Class A hospitals are also designated as teaching hospitals. Ninety-eight provincial hospitals are managed at “arm’s length” by PHOs (mainly Class B). DHOs in over 500 districts are similarly responsible for managing public health services and 54815 hospitals (mainly Class C and D) are similarly managed at “arm’s length” by DHOs.16 The MoH is responsible for around 33 central government hospitals (Class A), mainly the large teaching hospitals.17 Meanwhile, The non-government sector (both for-profit and notfor-profit) provides 1,436 hospital facilities. Other providers include other ministries, stateowned enterprises etc. This arrangement reflects the policy of distinguishing the roles of local government institutions (regulation, financing, and monitoring health service organisations) from direct management of service delivery. Table 2: Hospital stewardship

Category

18

Ownership

General hospitals

Special hospitals

Total

759

92

851

- MoH

14

19

33

Provincial government

53

45

98

- District government

446

10

456

- City government

80

12

92

- Other ministries

3

1

4

120

5

125

43

0

43

Private non-profit

522

202

724

Private

419

226

645

60

7

67

Government

PUBLIC HOSPITALS

- TNI - Police

PRIVATE HOSPITALS

SOE

15

City and District Government Regulation No. 41/2004. 17 MoH (2014), personal contact, RoI. 18 source, MoH, 2014 16

32

TOTAL

1,760

527

2,287

18

Table 3: Hospitals by type

TYPE RS

Hospital Class

PUBLIC HOSPITALS

PRIVATE HOSPITALS

TOTAL:

A

B

C

D

Uncl assifi ed

Total

- MoH

25

6

1

0

1

33

Provincial government

19

44

21

3

11

98

District government

0

68

234

117

37

456

City government

2

47

32

8

3

92

Non-profit organisations

1

56

243

205

219

724

- TNI

1

12

10

14

88

125

- Police

1

2

11

4

25

43

Other ministries

0

1

0

1

2

4

Private / Other

4

46

104

99

111

364

Company

0

13

65

42

92

212

Individual

1

1

17

20

30

69

SOE

2

6

21

13

25

67

56

302

759

526

644

2,287

Within these hospitals there are about 240,000 beds, 9,000 operating theatres, 16,000 maternity suites, and 4,000 isolation beds, representing about 10 beds per 10,000 population—all numbers that are steadily growing. However, despite the steady increases, the resources are unevenly distributed across the archipelago. Achievements in terms of the commitment to health care include a substantial increase in the number of hospital beds (i.e., capacity) across the country, together with a growing number of beds available per population.

33

Figure 5: Beds available

Figure 6: Beds available/100,000

34

Table 4: Human assets by broad category and location

Doctors dentists PUBLIC HOSPITA LS

PRIVATE HOSPITA LS

&

18

nurses

Midwives

Pharmacists

NonOther health specialist workers workers

Total

MoH

3,581

14,399

913

515

3,247

12,671

35,326

Provincial government

4,589

18,411

2,386

1,166

6,091

13,797

46,440

District government

11,156

53,557

11,428

2,964

16,745

3,539

99,389

City government

3,780

15,444

3,053

1,047

4,637

10,614

38,575

Non-profit organisations

17,304

32,812

5,872

3,088

7,829

32,926

99,831

TNI

3,208

8,835

1,220

499

1,408

5,892

21,062

Police

1,320

2,916

502

276

619

2,209

7,842

Other ministries

167

184

7

8

99

142

607

14,078

3,114

1,901

4,142

14,989

48,637

Private Other

/ 10,413

35

TOTAL

Company

5,486

8,950

2,520

1,286

2,256

7,684

28,182

Individual

1,075

1,250

651

288

433

1,541

5,238

SOE

1,634

3,993

443

471

754

3,054

10,349

63,713

174,829

32,109

13,509

48,260

109,058

441,478

36

These data raise a number of issues: • • • • •

It is unclear why the MoH operates Class B and C hospitals. It is unclear why provinces operate Class A, C and D hospitals. It is unclear why districts/cities operate Class A and B hospitals. There is wide variation across the country in the distribution of resources. The private and not-for-profit sectors are major players in the health-care system and cannot be neglected.

The impact of these issues is as follows: • • • •

The referral route of D-C-B-A is confused. If BPJS reimburses for activity alone it will reinforce and even exaggerate the uneven distribution of resources. It is unclear where the source of technical leadership lies given the overlap between hospital classes/types and tiers of governance. The governance role and responsibility of each tier is confusing.

The boundary and role anomalies in the distribution of hospitals need to be resolved by reassigning the stewardship of hospitals to their respective tiers of governance. 1.7.2

Primary health care

Primary health care here is taken to represent both personal curative and preventive services (encounters as defined by the International Classification of Primary Care19) and populationlevel services including public and merit services, such as surveillance, health promotion and education, infectious disease tracking, etc. The WHO20 argues that the development of primary health care is key to promoting good population health and cites four areas for careful attention: • • • •

Universal coverage Service delivery Leadership Public policy

Indonesia created a model of integrated health-service provision through the establishment of primary health centres (Puskesmas) in 1979. Currently, there are 9,655 Puskesmas, of which 3,317 provide in-patient services and 6,338 provide ambulatory services only (MoH, December 2013). The WHO describes the functions of primary health care and summarises them in the form of a diagram (Figure 7). The Puskesmas is the key facility in the provision of both preventive and curative care. Puskesmas represent a comprehensive service model, including the provision of non-clinical services covering public health, health surveillance, and counselling. The design of the Puskesmas model conforms closely to the intentions encapsulated in the WHO model and so the notion should be regarded as a positive aspect of the SKN architecture.

19 20

World Organisation of Family Doctors (Wonca) WHO (2008), Now More Than Ever, World Health Report 2008.

Institutional Analysis 20

Figure 7: The primary health care cube

Achievements in terms of the commitment to health care include a Puskesmas-to-population ratio of 30,000. However, 38 percent of Puskesmas are in a moderately or severely damaged condition,21 while around 2,000 Puskesmas have no physician in place (PODES, op. cit), and 430 sub-districts have no Puskesmas. 18

Table 5: Human assets in Puskesmas Doctors Dentists

22,898

and

Nurses

Midwives

Pharmacis ts

Technicians technicians

84,649

96131

5,264

2,902

&

pharmacy

Total

211,844

There is severe inequity in the distribution and quality of primary care services. The lack of physicians and other health workers mostly occurs in rural Puskesmas. “….there needs to be a focus not only on increasing the breadth and height of coverage but also in ensuring that effective depth of coverage exists, especially in the rural and remote areas of the country.”22 The institutional relationships between Puskesmas and other institutions within the wider health economy are unclear. The degree of autonomy available to Puskesmas is limited. Sub-national public service units (BLUD) have not been applied in Puskesmas. Incentives to encourage

21

PODES, 2011 National Institute of Health Research and Development (2013), Assessing Supply-Side Readiness for Universal Health Coverage in Indonesia, Ministry of Health, Indonesia & World Bank, Jakarta, Indonesia

22

38

Institutional Analysis

efficiency and productivity in service delivery are absent other than reimbursement through BPJS for some curative and rehabilitative services. 1.7.3

Public Health and Merit goods/services

Communicable disease control (CDC) Amongst the main public health functions residing in the MoH is communicable disease control (CDC), including vector transmitted diseases, directly transmitted diseases, neglected and emerging new diseases. The CDC programme covers preventative and curative activities delivered by all lines of health service, both public and private, while disease surveillance falls under the responsibility of the MoH involving PHOs and DHOs. An outbreak-reporting mechanism was established,23 and later upgraded in 2004.24 Subsequent to a report of an increase in the incidence of certain disease cases or symptoms, an investigation team will review the epidemiological characteristics of the outbreak and will produce an action plan to control the outbreak and limit further incidence/prevalence. (Investigations are to be initiated within 24 hours of reporting.) After International Health Regulation (IHR) 2005 came into force, Indonesia has been developing its core capacity as recommended by the WHO. The country has improved its hospitals, laboratories, surveillance, and health workers’ competence. The MoH is strengthening 48 port health authorities and its border surveillance capacity. Furthermore, the MoH needs to continuously introduce global regulations to the public, such as by inviting representatives of related institutions and sectors to seminars on implementation acceleration of IHR. The WHO has supported Indonesia in conducting its two largest simulations. In addition, the WHO has been working together with Indonesia in various aspects of capacity-building, including the coordinating mechanism for a field epidemiology training programme to improve the surveillance competence of health workers, strengthening laboratories to Bio-Safety Level 3, and supporting IHR networks in the MoH, as well as promoting intra-ministries collaboration. The WHO has recommended Indonesia to continue establishing networks in collaboration with other countries. In addition, Indonesia should include a 5-year plan and budget in an annual work plan. Non-communicable disease control NCDs, such as diseases of the cardio-vascular system (CVA and infarcts, in particular), are the leading causes of mortality, replacing lower respiratory tract infections and tuberculosis (TB). The response to this development has been the development of policies around health promotion, nutrition, and a greater investment in curative/rehabilitative services. Health education/promotion These activities are delivered at each level of government, ranging from national campaigns managed from the centre, to locally-specific activities.

23 24

Health Outbreak Law 1969 Ministerial Decree No. 949/2004.

39

Institutional Analysis

Other public health functions Puskesmas include health surveillance officers, as well as sanitarians, and their roles include responsibility for reporting, and environmental health at both individual and community levels. 1.7.4

Private sector/non-Government sector

While both supply and demand for private sector health services are reported25 to have grown in Indonesia in recent years, limited information is available regarding the quality and affordability of private sector health services. In addition, little analysis has been conducted into understanding how the private sector is contributing to public health priorities that have been identified by the Government and development partners. Indonesia’s private health sector is large, diverse, and growing. Unlike many countries around the world, Indonesia has long supported the development of the private health sector, beginning with encouraging private sector participation in the delivery of family planning services. However, the distinction between public and private provision of health-care services and products in Indonesia is not clear. Many publicly employed health personnel have second jobs in their own private practices or other private facilities. Some public facilities deliver private services and some state-owned enterprises are incorporated as private firms even though the sole or majority shareholder is the Government. This lack of clarity may make defining the scope of the private sector more difficult, but it also points to a policy environment that acknowledges the private sector’s contribution to health in Indonesia, and is conducive to private-sector participation in health-care delivery (USAID, op. cit). Although the private sector is included in universal health coverage (UHC), it is unclear what discussions have taken place with the private sector and the degree to which the sector is willing to participate, particularly private hospitals. Equally, it is not clear if reimbursements to the public and private sectors will be the same, or whether these will take into account the different costs incurred. The intention of the Government to continue supply-side funding to public hospitals means that unless the private hospitals are compensated for this discrepancy (public hospitals receiving both supply-side funding and insurance payments), the incentive for private-sector participation is weakened (USAID, 2009). The Government is planning to negotiate tariffs, and therefore the payments, with representatives of health-care providers, and payments are expected to differ according to geographic region (given regional variations in expenses) for both public and private facilities. The private sector on the supply-side is a major player in the health-care landscape. In the urban setting, it is thought to be the major provider of secondary care, and so cannot be neglected in terms of planning and health gains. However, reliable data on capacity, quality and activity are difficult to acquire. In terms of personnel, it is estimated that 60 to 70 percent of all public sector health professionals also provide services in a private capacity (USAID, op. cit). Provider organisations and professional groups/associations include PERSI, ARSADA, ARSVI, ARSANI etc. and professional associations such as IDI, IDAI, PDGI, PPNI, etc., which have roles in the regulation of medical and other health professions.

25

USAID (2009), Private Sector Health In Indonesia: A Desk Review

40

Institutional Analysis

On the demand side, there is a growing market for private health insurance (MoH, 2014, op.cit). Indemnity health insurance has developed, led by large insurers including Prudential, Allianz, ING, etc. In addition to health-care provision and the demand-side entry by private insurers, the private provision of clinical training is a growing phenomenon, particularly in medicine, nursing and midwifery. This creates a significant challenge for a number of reasons. First, the absence of any means of applying leverage over private-sector provision means that the labour economy cannot be balanced to reflect the demand-supply-location balance needed in the country. Second, growth in the number of students challenges the capacity of the SKN to offer good quality clinical placements and supervision. Third, the quality of training cannot be monitored and managed in the same way as public-sector training providers. Whilst the draft Medical Education Law that is currently in process in Parliament may offer some protection in the case of medical education, the same does not apply to nursing and midwifery. Therefore, it should be a priority to devise mechanisms to enable the MoH and the MoE to influence the providers of medical training, in both public- and private-sector settings, to ensure that graduates are produced in the right numbers, to the highest degree of competence, and deployed to the most needy areas of the country. 1.7.5

Vertical programmes

All vertical programmes are financed, planned, delivered and supervised from the centre across the entire country without the engagement of mainstream health-care services. However, a number of programmes do exist that are complementary or associated with the pursuit of health gain (excluding the health gain objectives dividend harvested from infrastructure investment, economic development etc.). These programmes include the National Population and Family Planning Board (BKKBN), which provides services at the national, provincial and district/city levels, and operates in a parallel environment (not competing with the mainstream services for resources), and the National Drug and Food Control Agency (BPOM). Operational Health Assistance (BOK), although delivered locally, represents a centrally-funded hypothecated fund, (that is, earmarked for specific purposes) and so is a centrally governed but locally delivered programme. However, at a local level, although earmarked by central government, it is not ringfenced for disbursement in healthcare) Some programmes that engage international co-operation (e.g., the Global Fund to Fight AIDS, Tuberculosis and Malaria, or GFATM) require MoH-level commitment (in the case of GFATM, a Country Coordinating Mechanism, or CCM), which insists on a centrally-governed partnership agreements. These programmes are slightly unusual in the sense that the discretion/options usually available to local levels of government for general health services are retained by the centre. There is also a conditional cash transfer (CCT) programme known as Program Keluarga Harapan (PKH). PKH provides direct cash benefits conditional upon household participation in locally-provided health and education services. The PKH program, which made its inaugural payments to pilot regions in 2007, is designed to directly reinforce household incentives for investing in health and education. PKH has an immediate impact on household vulnerability while encouraging investment in long-term household productivity that may disrupt the intergenerational cycle of poverty. The quarterly cash transfer component ranges from a minimum of Rp 600,000 to a maximum of Rp 2.2 million per year. Typical of CCTs elsewhere, 41

Institutional Analysis

disbursements are made only after a mother’s verified attendance at pre- and post-natal checkups, a professionally-attended birth, and participation in child health development surveillance activities. PKH gives cash, which can be used to defray the cost of attending conditioned services, while also promoting early investments in health and education that have long-lasting implications for welfare and productivity. These parallel programmes are closely targeted and do not seem to have a significant impact on the functioning of the general architecture of the SKN. However, they will have an impact on the coverage and reach of key services (e.g., MNCH, HIV/AIDS, nutrition). 1.8

Equity and access

During the life of the 2009 RPJMN, the key social indicator (in terms of social solidarity and public commitment to a national scheme of health protection) could be considered to be represented by the achievement of equity. Indonesia has made impressive health gains over the past few decades. Life expectancy at birth has increased from just over 41 years in 1960 to more than 71 years in 2013.26 The infant mortality rate (IMR) dropped from 42 to 27 per 1,000 live births (Index Mundi, op. cit) and the under-five mortality rate dropped from 216 to 31 per 1,000 live births over the same time period (Index Mundi, op. cit). The 1997 economic crisis and the decentralisation of government administration in 2001 do not appear to have had a discernible impact on trends in average life expectancy, infant mortality, and under-five mortality in Indonesia. The country is on track to attain the Millennium Development Goal (MDG) for child mortality. 27 Based on global comparisons, Indonesia’s IMR in 2013 was lower than the average for its income level and its life expectancy was about average for its income. These impressive gains notwithstanding, Indonesia’s performance in some other key health outcomes, such as maternal mortality and child malnutrition, has been relatively poor. Compared with its performance in under-five and infant mortality, Indonesia’s maternal mortality rate (MMR) is amongst the highest in the region, and much higher than one would expect for its income level. Indonesia’s MMR, considered to be one of the best indicators of the performance of a health system, was 220 per 100,000 in 2013 (Index Mundi, op. cit). From a regional perspective, Indonesia lags behind its peers in most health attainment indicators. Its life expectancy, under-five mortality, and infant mortality rates are worse than peer countries such as China, Malaysia, the Philippines, Thailand, and Vietnam. Indonesia introduced a single-payer system in 2014 and has already merged five central schemes. Outputs include a single ID registration system, a unified benefits package, referral criteria, and new payment systems. BPJS now covers 110 million people and aims to cover the entire population by 2019. By then, BPJS will be the largest single-payer system in the world. This would be an admirable achievement, although it has been argued that the “low hanging fruit” has already been picked, and the future extension of the scheme will present far greater challenges. As at 2012, about 151.5 million people were thought to be covered by health insurance, distributed as follows:

26

Index Mundi (2013)

27

UNESCAP et al 2007

42

Institutional Analysis

Table 6: Coverage of Health Insurance, 2012

TYPE OF HEALTH INSURANCE

28

PERSONS

Participants in health insurance for civil servants (Askes PNS) 17,274,520 TNI/Polri (military and police) 2,200,000 Jamkesmas participants* (MoH) (health insurance for the 76,400,000 poor) JPK Jamsostek participants (workers’ social security) 5,600,000 Jamkesda/PJKMU participants (regional governments’ health 31,866,390 insurance) Corporate insurance (self-insured) 15,351,532 Private health insurance participants Total

2,856,539 151,548,981

Jamkesmas, the health insurance programme for the poor and near-poor, has been integrated and merged with other social insurance programmes. Jamkesmas can make the following claims:29 • • • • •

1.9

About 47 percent of poor and near-poor households were covered under the programme. Out-patient and in-patient utilisation rates increased amongst programme cardholders. Levels of catastrophic payments declined. Participation of private providers increased. More than 300 complementary local Jamkesmas-inspired programmes were initiated across the country. Health expenditure

In the health economy, total health expenditure is estimated at 2.5 percent of GDP in 2014. This has risen slowly, from 1.9 percent of GDP in 1996 to 2.2 percent of GDP in 2006, with general government expenditure comprising 50.4 percent of the total, and private spending comprising 49.6 percent (WHO, 2008). The Government expects total health spending to increase to 4 percent of GDP by 2019, following the implementation of national social health insurance (SJSN) in 2014. The GoI notes the experience of other countries in funding UHC, where state expenditure for health is typically 6 to 11 percent of GDP, with tax ratios of over 20 percent. Currently, half of total health expenditure in Indonesia comes from private sources, primarily out-of-pocket payments, including user fees (66.3 percent), with a proportion from private prepaid health care plans, and the rest spent by NGOs and private firms (WHO, 2008). About 5 percent of medical insurance costs are attributable to pharmaceuticals. This represents per capita growth from US$16.4 to US$32.4 in 2011 (public), and US$98.3 to US$126.9 (total).30

28

Source: Republic of Indonesia 2012. Road Map toward National Health Insurance, 2012-2019 World Bank (2014), Indonesia's path to universal health coverage : key lessons from the implementation of Jamkesmas, http://www.worldbank.org/en/news/feature/2014/01/30/improving-jamkesmas-to-achieve-universal-health-care-in-indonesia 30 WHO (2014), http://apps.who.int/gho/data/node.country.country-IDN 29

43

Institutional Analysis

Official Development Assistance for health (in 2010) represented just over US$500 million. The largest disbursements were drawn from multi-lateral funding sources, the most substantial being from GFATM at 38.4 percent of the total.31 1.10

Political commitment

In terms of legislative intervention, a number of key items have been passed into law, or are in the process of passing into law, including the: • • • • •

Health Law No. 36/2009 Social Welfare Law No. 11/2009 Hospital Law No. 44/2009 Social Security Providers Law No. 24/2011 Draft Medical Education Law (still under discussion in Parliament)

These laws have been designed to broaden access, to pave the way towards UHC by 2019, and to increase the supply of skilled professionals in the health care system. They have the effect of stimulating both demand- and supply-side factors in the health economy, producing sustainable growth in access to services. Indonesia has made substantial progress on UHC through the establishment of a clear policy framework. The Government has passed two key laws: (i) the Law on the National Social Security System No. 40/2004; and, after considerable delay, (ii) the Law Establishing the Social Security Agency (BPJS) Law No. 24/2011 in October 2011. Health insurance for the poor and near-poor (Jamkesmas) has been expanded to reach 76.4 million people (32 percent of the population). With the passing of Law No. 24/2011, the Government has developed a guide for the implementation of UHC. This is known as the Road Map towards National Health Insurance - Universal Coverage 2012-2019. There is a tension between the political and social objectives underpinning the decentralisation philosophy on the one hand, and operating efficiencies on the other. Decentralisation encourages social solidarity at a local level, the dispersion of power and influence, and the introduction of checks and balances into governance frameworks. Meanwhile, there are weak mechanisms of regulation, of policy implementation, and of performance management. However, the exercise of direct managerial authority by central institutions is incompatible with the notion of local democracy and accountability to an electorate. The central government’s legitimate interest in implementing policy nationally sits uncomfortably with provincial and district/city government operating privileges. Apart from financial rewards/sanctions, the central government has few levers available to ensure conformity to central strategy and plans. The dual role of provincial governors as both elected representatives and as agents of central government confuses supervision relationships with districts/cities, and further confuses their roles in the direct management of some services. This phenomenon is not unique to Indonesia: all multi-level government models experience this tension to a greater or lesser extent. However, the scale and complexity of Indonesia complicates relationships and makes

31

ODA (2012), ODA Health Disbursement, WHO

44

Institutional Analysis

mitigation difficult. Other countries have opted to extend decentralisation further (effectively creating federal models, e.g., Nigeria) or have created local or regional quasi-non-governmental agencies to manage service delivery with policy implemented through partnership agreements captured through service-level agreements or contracts (e.g., the United Kingdom). One achievement has been movement towards the creation of quasi-non-governmental agencies (e.g. BPJS), but it is unclear how the central government will exercise its authority at provincial and district/city levels through the relationship between the central government (particularly the MoH) and BPJS. 1.11

Health sector achievements

1.11.1 Organisational development Formal organisational relationships are complicated (Figure 4).Not only have new organisations appeared in the landscape, notably BPJS, but the revised decentralisation laws have also altered, and will continue to alter, roles and relationships throughout the health sector and related institutions. The scope of the primary health care reforms include the development of strategic policy and evidence-based planning to ensure the implementation of efficient resource allocation, financial efficiency to obtain the best results at the lowest costs, the establishment of appropriate institutions with proper function and size, and the availability of data that are valid and reliable, as well as governance on the implementation of good health actions. The expected impact of these reforms is the development of inclusive and equitable health status. In terms of organisational development, BPJS has been established, existing national social protection schemes are being consolidated into the BPJS portfolio, and local schemes will be recruited in due course. The MoH is undergoing restructuring partly to reduce payroll overheads and partly to ensure that the structure of the MoH more adequately reflects the life-cycle and building-block approaches to planning and delivery. 1.11.2 Service quality and standards In Indonesia, many policies, organisations, methods and projects contribute to a quality system at national, provincial and district/city levels. Projects and innovations such as the development of Clinical Performance Development and Management Systems (CPDMS), the implementation of ISO 9000 standards in hospitals, the implementation of clinical audits by the national insurance company (PT Askes), the Health Professional Education Quality Project (HPEQ Project) funded by the World Bank, together with many recent projects, such as the preparation of KARS for international accreditation, and the introduction of universal health insurance, are aimed at improving health-care quality. Moreover, quality improvement methods such as accreditation, licensing and certification of health-care organisations have been implemented nationally. In addition, professional organisations have formulated clinical guidelines for each area and some provinces have developed their own quality policies. The reason for introducing minimum service standards (SPM) is to address regional disparities in public service delivery and to improve public services. The standards will ensure that all district/city governments deliver basic levels (milestones) of public services, which are predetermined by the central government. Following the MSS/SPM guidelines published by 45

Institutional Analysis

MoHA, MSS/SPM indicators should follow an output-based approach. This means that local governments must achieve a given level of service output, but they can decide themselves on how best to achieve this level. SPMs for the health sector were issued in 2008 and cover 17 health specific standards covering basic services, hospital services, and epidemiology and community empowerment, and are heavily weighted towards primary care.32 The indicator for the higher coverage of prenatal visits, for example, is defined as the share of pregnant women who “obtained at least four ante-natal standard antenatal examinations”. The specific milestone, which has to be achieved in order to provide the minimum level of services, requires that this share be at least 95 percent. Each standard is legitimised by ministerial decree and supported by technical guidelines and the guideline for planning and budgeting. Table 7: Health Minimum Service Standards

Basic Health Services 1.

By 2015, Antenatal care coverage will achieve 95%.

2.

By 2015, 80% of obstetric complications will be attended.

3.

By 2015, 90% of deliveries will be attended by skilled health providers who have obstetric competencies.

4.

By 2015, post-partum visit coverage will achieve 90%.

5.

By 2010, 80% of neonates with complications will be attended.

6.

By 2010, neonatal visit coverage will achieve 90%.

7.

By 2010, 100% of villages will complete Universal Child Immunization.

8.

By 2010, Under-5 children service coverage will achieve 90%.

9.

By 2010, 100% of poor children aged 6-24 months will receive food supplements.

10.

By 2010, 100% of malnourished Under-5 children will receive healthcare.

11.

By 2010, 100% of elementary school children will receive health examinations.

12.

By 2010, the number of active family planning clients will reach 70%.

13.

By 2010, disease detection and treatment coverage will achieve 100%. [sic!]

14.

By 2015, the coverage of basic healthcare for poor people will achieve 100%.

Referral Medical Services 15.

By 2015, the coverage of referral services for poor patients will achieve 100%.

16.

By 2015, the coverage of level 1 emergency services provided in district hospitals will achieve 100%.

Epidemiological Research and Management of Outbreaks (KLB)

32

MENKES, 2008: Article 2 (2)

46

Institutional Analysis 17.

By 2015, an epidemiological investigation will be conducted in < 24 hours in 100% of villages where an outbreak occurs.

Health Promotion and Community Empowerment 18.

By 2015, 80% of villages will become Active Desa Siaga.

1.11.3 Citizen entitlements The State guarantees a benefits package through Jamkesmas.33 With the exception of private insurance, nearly all of the existing insurance schemes have comprehensive benefits packages that include in-patient and out-patient care. The benefits package supervised by BPJS has to be comprehensive, covering promotional, preventive, curative and rehabilitative services. It will exclude specific conditions, such as those resulting from abuse of addictive substances and cosmetic procedures. There is no expectation of co-payments unless participants choose to pay for an upgrade in services, while additional payments are required for workers who wish to cover more than five family members. Schemes have different levels of coverage based on the benefits package available to different type of facilities and amenities. Some schemes allow beneficiaries access to both public and private providers (e.g., Jamsostek), whereas others require beneficiaries to obtain services from the largely public network of providers (e.g., Askes and Jamkesmas) for most services, with access to private providers for a more limited range of services. Jamkesmas provides coverage for in-patient care in a hospital and Askes covers in-patient care in at least a Class B hospital depending on the position and level of the civil servant. In order to improve maternal and child health in Indonesia, the GoI established a special health insurance scheme in 2010 (Jampersal) that guarantees neonatal and post‐natal care for mothers and their infants. Law on National Social Health Insurance (SJSN) No. 40/2004 states that health services include counselling, immunisation, family planning services, out-patient, in-patient, emergency services and other medical actions, including dialysis and heart surgery. The services should adhere to minimum service standards in order to guarantee the continuity of the programme and the satisfaction of participants. The breadth of health services is adjusted according to the needs of participants who can be charged and the ability of the scheme’s financial resources to pay. BPJS is mandated to implement Law No. 40/2004the law and to accommodate the public need. A draft of the design of the presidential regulation in a social health insurance is currently under review. Jamkesmas offers a comprehensive benefits package, including both in- and out-patient care, as well as maternal and preventive care. In terms of medication, members are only entitled to coverage for drugs from specific formularies and must opt for generic drugs when filling prescriptions. Exclusions from the Jamkesmas benefits package include cosmetic surgery, annual physical check-ups, alternative medicine, dental prosthesis and fertility treatment. Cancer treatment and treatment for heart-related problems are also limited. Jamkesmas beneficiaries are able to seek care at both public and private outlets, although covered ambulatory services are solely public. 33

Adiyas, A, (2014), Expanding Health Coverage: Indonesia, Joint Learning Network for Universal Health Coverage

47

Institutional Analysis

The scheme has contracts with 926 hospitals for service provision, including 220 private hospitals for certain procedures. Overall, free access to many providers, both private and public, and a comprehensive benefits package, make Jamkesmas more attractive to the majority of the population—even those covered under Askes and Jamsostek.34 Almost 80 percent of people who were already enrolled with health insurance schemes, such as Askes and Jamsostek, preferred to be covered by Jamkesmas, as they felt the coverage provided under Jamkesmas was superior to that provided under their existing plans. Jamkesmas has enrolled more beneficiaries than any other Indonesian health insurance plan. 1.12

Headline achievements

The informal culture in the health-care landscape has crystallised in a number of ways. First, at a political level, commitments have been made to expand public health care substantially, apparently with minimal political resistance, as the plan to introduce UHC by 2019 has attracted a high degree of political consensus. Second, at the high level of the executive branch, acknowledgement of the necessity of refocusing the attention of the MoH at a more strategic level has been understood. BPJS has already shown a degree of technocratic professionalism and drive. At the service delivery end of the SKN, attempts to professionalise health-care management are underway, and the private and public sectors are seeking opportunities for collaboration. Significantly, the Road Map targets make explicit reference to enriching community, patient and staff experience of the system. “At least 75 percent of the members shall be satisfied with the services provided at the NHIC and by health-care providers contracted by the NHIC” by 2014 (85% by 2019) and “At least 65 percent of the health workers and health care providers shall be satisfied with and or receive appropriate payment by the NHIC” by 2015 (85% by 2019) and “The NHIC will be managed in a transparent, efficient and accountable manner.” These statements represent a determination to alter the cultural framework within which the SKN operates. Indonesia has achieved substantial progress during the life of the 2009 RPJMN, as seen in the following: • • • •

Growth in the supply of services (primary and secondary). Stimulation and consolidation of demand-side factors (PBJS and UHC). Growth in financial commitment to health. A legislative programme that commits the Government to continue a path of growth and greater inclusion/participation in access to health care.

34

Center for Health Financing Policy and Health Insurance Management (2008), Study on Benefit Package Based on Community’s Preference, University of Gadjah Mada

48

Institutional Analysis

• The roadmaps towards NHIC 2012-2019 containing many targets and milestones; some are in progress, some have not been achieved but many have been achieved,. • Growth in clinical learning opportunities. Table 8: Headline achievements

• • •

• • • • •

The number of citizens insured has increased significantly. The creation of new organisations/systems to stimulate demand (BPJS/UHC). Legislative interventions designed to broaden access, to pave the way towards UHC by 2019 and to increase the supply of skilled professionals in the health care system. These have the effect of stimulating both demand- and supply-side factors in the health economy, producing sustainable growth in access to services. In absolute expenditure terms of per capita growth from US$16.4 to US$32.4 in 2011 (public), and US$98.3 to US$126.9 (total)35. Free access to many providers, both private and public, and a comprehensive benefits package have made Jamkesmas attractive to the majority of the population. At the political level, commitments have been made to expand public health care substantially. , The plan to introduce UHC by 2019 has attracted a high degree of political consensus At the high level of the executive branch, acknowledgement of the necessity of refocusing the attention of the MoH at a more strategic level has been understood. BPJS has already shown a high degree of technocratic professionalism and drive.

35

WHO (2014), http://apps.who.int/gho/data/node.country.country-IDN

49

Institutional Analysis

2.

Remaining challenges

2.1

Introduction

Despite the significant achievements recorded in the 2009 RPJMN, a number of major issues remain outstanding as summarised below. 2.2 • •

• •

• •



2.3

Social determinants of health Increasing life expectancy, and hence aging population (by 2025 the number of elderly people will double to 23 million). An increasing population of the young, producing a population profile that will become bimodal in due course, but with a high dependency ratio until the population of young people become economically active. A population growing at a rate of 5 million per year. Epidemiological transition: A changing pattern of disease from infectious diseases to NCDs which requires ongoing management of chronic diseases, and a tripling of the diseases burden with: (i) the unfinished agenda of communicable diseases, (ii) new-emerging NCDs, and (iii) emerging infectious diseases. A growing middle class of almost 100 million by 2015, with higher expectations for high quality care and more aggressive health seeking behaviours. Increasing medical technologies, services and procedures and new drugs. It has been observed that both patients and providers often demand new treatments even if there is little or no evidence that they are clinically superior to alternative treatments. Increasing JKN coverage. The “insurance effect” due to either the necessity of care or moral hazards from both consumers and the provider. In other words, both supply and demand are growing. The relationship between institutional arrangements and health outcomes

Figure 2 in Section 1.3 refers to the building blocks of a health system. Of these, the sixth refers to “Leadership and Governance”. The MoH has two significant roles in this context: one is to exercise technical leadership in the health sector and the other is to act as the implementing entity on behalf of the President, who is ultimately responsible for public health care. In both of these roles, the MoH faces significant challenges. The first challenge concerns the mechanisms that it has at its disposal to influence and work with partners with whom it has no line relationship (Figure 4 in Section 1.5). The second concerns how to influence health-care delivery in the public sector within the framework of a decentralised statutory structure. The third challenge concerns how to influence and regulate the private sector. Meanwhile, the MoH’s role as a technical leader is not harmonised with flows of money, accountability and information/ reporting. Finally, the structure of the MoH brings with it a number of anomalies. It both directly manages some aspects of service delivery, and also regulates and supervises health strategy and policy. The result is that the management of health outcomes faces considerable structural barriers, both in terms of the exercise of leadership and the efficiency of the institutions in leadership positions.

50

Institutional Analysis

2.4

Health sector oversight and governance

Over the next five years (2015-19), health development is planned to tackle disparities and health inequities. Various programmes and activities that have been effective or are still in progress will continue to be implemented. However, bureaucratic blockages that have hindered the implementation of health programmes still need to be addressed. One of the products of the decentralisation process is regulations that conflict or overlap. The decentralisation of health development over the past 12 years has allegedly contributed to a widening gap in health inequality. Political decentralisation and fiscal decentralisation have not contributed positively to overcoming public health problems at the district/city level. Problems will become more complex with the growing prevalence of NCDs and the major reform of health systems triggered by the institutionalisation of JKN. Others problems to be addressed include changes in disease patterns, changes in values and expectations of the community, high out-of-pocket payments, fragmentation of health services and health programs, the over-emphasis on the curative nature of budget allocations, change in Puskesmas functions to include more curative inputs, economic turbulence, and the revised decentralisation law and other regulations including the revisions of Law No. 32/2004, Law No. 33/2004, Government Decree No. 38/2007, Government Decree No. 41/2007, MoHA Decree No. 59, minimum service standards, etc.). Another problem is the difficulty of paying for primary care services in the same way as paying for secondary and tertiary care, because of the distinctively different characteristics of the services involved. Arrangements for the distribution of power between tiers of government are enshrined in Government Regulation No. 38/2007 (PP No. 38/2007) based on the Law No. 32/2004 on Regional Government. Amongst the main weakness in the PP No. 38/2007 is the unclear distribution of responsibility for health functions. The difference between central and local responsibilities is determined by the word “scale”, meaning the scope of operations based on administrative boundaries. This definition creates contestable interpretations of the “ownership” of roles. Responsibility for health functions becomes the authority of the centre if the issue in question relates to heath matters crossing provincial boundaries. Likewise, health matters that span two or more districts come within the authority of the provincial government and this could mean only within the district/city is it under the authority of the district or city. This unclear distinction of functions has resulted in a lack of clarity about which tier of government is responsible for certain health services. Both local and central governments overlap in managing certain health services and this could mean that neither takes responsibility. Anecdotally, in most cases health is abandoned by local governments, with the expectation that the centre will finance health care through DAK allocations. This is why the MoH is often working on matters that are actually the responsibility of local governments. Under the revision of Law No. 32/2004 on Regional Government, a clear cut distribution of functions between tiers of government is envisaged to address current ambiguities. At present the distribution of functions is arranged under Government Regulation (PP) No. 38/2007, and this creates a problem in that the boundaries of responsibility for functions are not clear, with overlaps occurring in many cases. In the revision, the MoH is expected to have stewardship of Class A hospitals, provinces will have Class B hospital, and districts/cities will control Class C and D hospitals, together with Puskesmas. The revision will clarify the respective roles of the MoH, and the provincial and district/city health Dinas. 51

Institutional Analysis

Put simply, the centre will be responsible for regulation, resourcing, stewardship, standardsetting, planning, strategic capital and advocacy, and for the management and performance of Class A hospitals. Provincial governments will be responsible for governance, strategic planning, advocacy and oversight of province-wide services and for the management of Class B hospitals. District/city governments will be responsible for operational planning, budgeting, advocacy, and the management of Class C and D hospitals, and for all Puskesmas. The continuing discussion concerning the revision of Law No. 32/2004 is expected to produce a clearer distribution of responsibilities amongst the tiers of government. Important issues in the draft revision include: • An emphasis on public health promotion in all tiers of government, since curative activities are reimbursed by BPJS. • The deployment of specialist doctors to remote areas to correct for the shortage of specialist in these areas. • A clarification of responsibilities in each tier of government regarding the stewardship of health facilities and a clarification of the referral mechanisms system between levels of hospital. • A clearer distribution of responsibilities on the supervision of the food, drugs and healthequipment industries, including the distribution of traditional medicine. The revision of the distribution of responsibilities maintains a residual problem in the form of a lack of clarity between the roles of the centre and local governments. It is expected under the new distribution of functions in the revision of Law No. 32/2004 that the responsibilities of the MoH, and those of the PHOs and DHOs, are clarified to eliminate overlapping and the misallocation of funds through various streams deployed (such as Dana Dekon and Dana TP). How these can be reconciled in a decentralised, but nonetheless unitary political economy will continue to create tensions. Finally, the role of provincial governors as both elected representatives and as the agents of central government is anomalous and inconsistent with the principles of representation and appointed agency. In practice, this may seem unimportant, but in principle it is anomalous and emerges as a source of tension between local governments and the centre. This is especially significant when funding is largely on a grace-and-favour basis and is not the product of a process of need-based and policy-driven planning. Under the Constitution, the responsibility for all government functions lies with the President. Since the ultimate responsibilities for governance rest with the President, no matter how broad the autonomy devolved to local governments, there remains a key role for central government in any of those functions. A very minimum involvement of central government is at least needed to support and guide local governments to implement national policy locally, and to support local governments in the management of the devolved functions, particularly around information, financial management and the delivery of targets. The extent of central intervention will depend on the capacity of local governments: the greater their capacity, the less the need for intervention by the centre. This mechanism assumes that the relationship between central and local governments is a command relationship and a product of Indonesia’s unitary model. In Indonesia, local government is a unit of government implemented by the centre (and which could, in theory, be 52

Institutional Analysis

abolished by the central government through the issuance of law by the President and Parliament). Therefore, there remain unclear relationships in key areas; • • •

The unclear relations between the MoH with PHOs and DHOs. The unclear relationship between PHOs and DHOs with local hospitals. The unclear relationship between the MoH, PHOs and DHOs with BPJS.

The President appoints a minister of health to be responsible for health matters nationally. Most of the MoH’s duties on operational matters are devolved to the provinces and the districts/cities. Nevertheless, the final responsibility still lies with the President. In other words, no matter how broad and deep the autonomy on health matters devolved to local governments, final accountability still rests with the MoH. The MoH issues guidance (NSPK) to be implemented by district/city governments (through a Perda) in running their own functions. No matter what degree of autonomy is granted to local governments, they should still conform to the guidance set by the centre. Therefore, the MoH has both a role and responsibility to ensure that PHOs and DHOs discharge the functions devolved to them. Any deviation from national policies should be corrected by the MoH. In practice, it is impossible of the MoH to supervise DHOs effectively and efficiently from the centre. Law on regional Autonomy No. 32/2004 endeavours to mitigate the problem by assigning provincial governors as the representatives of the MoH to supervise and facilitate DHOs (pembinaan). There should be guidance from the MoH to governors on how to exercise pembinaan. However, the MoH has yet to issue any policy to support governors in supervising DHOs. Without such guidance, it is very difficult for governors to implement their roles as central representative. The law states that any cost arising from the roles played by governors as central representative in supervising district/city governments should be borne by the centre. This fund (Dana Dekonsentrasi) is reported to be frequently used to finance other functions (fungibility). The supervision of PHOs should be conducted by the MoH, since the fact that there are only 34 PHOs makes direct supervision and facilitation feasible. The MoH should concentrate on technical supervision as more general supervision will be handled by MoHA. Therefore, there should be a clear-cut division over which areas are covered by the notions of “technical” and “general”. This philosophy is fully understood by neither central nor local elites, resulting in complex and often fractious relationships between them. The centre considers itself to be powerless as local government affairs are already decentralised. The local elites consider there to be no central privilege or formal leverage, since all powers devolved to them are in the local domain and so there is no legitimacy for intervention from the centre. Indonesia is a unitary country but with a federal frame of thinking and de facto federal behaviours (without the structural, constitutional or political legitimacy associated with federalism). The central ministries, instead of exercising pembinaan towards local governments, tend to use financial allocations as a mechanism to manage political relationships. Seventy percent of the national budget is allocated to the centre, leaving only 30 percent to be shared between provinces and districts/cities. When authority decentralisation is not aligned with fiscal 53

Institutional Analysis

decentralisation, a number of undesirable consequences emerge. Patron-beneficiary relationships are formed whereby financial allocations become an instrument of relationship management. Patron-beneficiary relationships are seen in the tendency of many ministries to encourage local governments to establish new organisational forms and elements, even when these do not reflect the need of local governments. One example is the establishment of disaster organisations in regions that rarely experience natural disasters. As a result, local governments are granted a disaster office and equipment financed centrally. The provision of DAK in some cases is not based on financial capacity or need but through lobbying the centre by local governments. These practices are widely known and recognised and have critical responses by many local governments who have less capacity to lobby effectively. The MoH operates outposts throughout the country, but focused on specific activities (e.g., health surveillance, some laboratory services). These are not supervisory operations. The supervision and facilitation of provinces is directly conducted from the centre; technical facilitation by the MoH and general facilitation (pembinaan) by MoHA. This arrangement requires a high degree of cooperation between the MoH and MoHA in supporting provinces. The supervision and facilitation of districts/cities (according to Law No. 32/2004) is delegated to provincial governors as the representatives of central government. It is plainly impractical for central institutions to directly supervise more than 500 districts/cities. There is limited practical empowerment and capacity of governors to exercise this role effectively. The funding for the governors’ supervision and facilitation of district/city health agencies should be borne by central government institutions (the MoH and MoHA) through “Dana Dekonsentrasi”. Currently, governors are not supported by a clear and professional apparatus. In the revision of the decentralization legislation, there will be an agency established and located in the province to support the governor in the conduct of his/her role as central representative supervising and facilitating districts/cities and financed from “Dana Dekonsentrasi”. As representatives of the centre, governors have both the duty and the opportunity to exercise delegated power over districts/cities, reflecting the hierarchical relationship between national and sub-national institutions in the context of a unitary but devolved framework. This mechanism is conceptually clear, but practically complicated. The deployment of officers accountable to the centre, but supporting provincial governors in the new agency will be an asset in strengthening the role of governors as central representatives. What seems to be required is a “light touch” mechanism for the MoH to be able to exercise its legitimate roles through the end-point deliverers in a decentralised framework, based on agreement rather than diktat. Such a mechanism would seem to be that relationships, roles, responsibilities, accountability, reporting/M&E and quality requirements are captured in partnership agreements such that the role of provincial and district/city governments (and specifically Parliaments) is respected but that the MoH retains a role at a meta-level. Partnership agreements could variously be described as service level agreements (SLA), contracts or as performance agreements. The main challenge to using such an option is that the financing element needed in such partnership agreements will be superseded by the reimbursement mechanism administered by BPJS. Therefore, an alternative model would be that various lines of accountability (horizontal,

54

Institutional Analysis

vertical and bi-lateral) are represented by service level agreements, the contents reflecting the roles of the institutions involved. So, there would need to be a contract between the MoH and BPJS to reflect performance requirements (reimbursement intervals, quality delivery, the registration of scheme members, reporting) and MoH imperatives (the pursuit of geographical equity, tariffs that encourage a focus on the developing patterns of disease) and, in return, the contract would deliver stability, certainty and a political interface. There would be similar agreements between the MoH and provinces, and between provinces and districts/cities, reflecting both decentralised roles and privileges, but also providing a mechanism for exercising responsibilities residing “further up” the hierarchical gradient. The notion does not oppose the principles of decentralisation, since performance and activities are the product of agreements, not the exercise of superior power. At a conceptual level, the Constitution perceives that decentralisation delegates the authority of government from the centre to local institutions, but responsibility and accountability remain centrally held. In practice, anecdotally, districts/cities have interpreted decentralisation as surrendering power to local governments, and so assume that they are now autonomous. However, the distinction between autonomy and delegated power is significant. There are anomalous effects of this misunderstanding, including a continuing dependency on the centre by local governments for the direct financing of health care, the anomalous role of provincial governors as both elected representatives and as the agents of the central government in the provinces, and tension between local governments and the centre when funding is largely on a grace-and-favour basis and is not a product of a process of need-based and policy-driven planning. The MoH is treated as a rescuer and colludes in this. Formal performance partnerships rather than the exercise of authority or patronage are a mechanism to reconcile the tensions between the principles of decentralisation and need to deliver services consistent with national policy. Such partnerships would also have the effect of reducing the exposure to fiduciary risks. Within the MoH, the lack of agility referred to earlier creates a heavy bureaucratic burden, long lines of communication, risk aversion, and territorial behaviours. The cumulative impact of these impacts is a MoH that lacks creativity and invention. The reduction in the workforce, merging information resources, withdrawing from the direct management of personal services, and refocusing on its development role, will all contribute to releasing opportunities for innovation and reducing the conservative culture that currently dominates the MoH. 2.5

Cross-sector relationships, coordination and accountabilities

At present, around 60 percent of health expenditure is directly levered by the MoH, and this proportion will fall as BPJS grows its “market share”. Other spending partners include the MoE, the MPW, BKKBN, MoHA. However, as Figure 4 shows, the structural relationships, and relative statuses between them, denies the MoH the direct authority necessary to determine the priorities grasped by partners, other than appealing to the Cabinet level for executive-level intervention. This is an undesirable mechanism for several reasons. First, it creates uncertainty because there is no certainty of successful appeal. Second, it is a clumsy and slow process. Third, it encourages unnecessary tensions between the partners. The notion of performance agreements could apply equally to bilateral relationships (see above), and has been applied in other national settings (e.g., at the federal level in Nigeria and within the Government of Wales). 55

Institutional Analysis

2.6

Transactions

The impact of the lack of clarity between the various tiers of government is aggravated by weak funding and the resulting financial pressures at the local government levels, particularly the district level. This lack of clarity risks causing a deterioration of health service delivery in terms of staffing and service access, producing in turn verifiable undesirable outcomes and reduced health status. Currently, most of the local budget comes from the centre in the form of a block grant. Almost 92 percent of the local budget comes from the centre, and only 8 percent is derived from local revenues. From the entire local budget for running 31 functions, 70 to 80 percent of the money is used for overhead costs, leaving only 30 percent for services. In future, funding for personal services will be disbursed by BPJS, whilst public merit services will continue to be provided by local governments. About 20 percent of the 30 percent for services is absorbed for funding in education as mandated by the Constitution, leaving 10 percent for funding all other services including health. The impact of this is that local governments depend on funding for health services from the centre, and even divert their funding for health with the expectation that the centre will make up for any shortfall (fungibility). Such an opportunity will be curtailed after 2014 and the introduction of UHC via BPJS. Table 9: District/City income source in local budget (2010)

DIRECT SERVICES 21.17% (DAK 6.31%)

%

Disposal Overhead costs, comprising

78.83%



Salaries

45.56%



Project overheads

22.22%



Operating costs

11.5%

Source Central grant: Locally derived revenues DAU from total grant

91.86% 8.14% 70.24%

Under these circumstances, there are limited resources to finance all 31 devolved functions, and so fungible expenditure is encouraged away from the intended area of expenditure. This phenomenon is known to the MoH and MoHA but the devolved status of local government makes it difficult for the centre to intervene to correct the misdirection of funds. Government Regulation No. 38/2007 clearly describes the distribution of health responsibilities between central, province and district/city. All functions assigned to the centre will be delivered by the MoH and funded from the national budget. Likewise, functions of the provinces will be

56

Institutional Analysis

handled by PHOs and funded from provincial budgets, while functions of districts/cities will be handled by DHOs and funded from district/city budgets. The problem arises when there are insufficient funds for the district to deliver its responsibility for its assigned health functions. According to the Law No. 17/2003 on State Finance, local governments are permitted to seek financial assistance either from the province or the centre. The legal mechanism to channeling the money from the MoH to the DHOs is from the Special Allocation Grant (DAK). The DAK should be channeled directly to the local budget and the role of the MoH is to monitor and supervise whether the DAK has been used in accordance with the specific uses proposed by the local government. In reality, instead of channelling it through the DAK mechanism, the MoH tends to channel it through (Dana Tugas Pembantuan (TP). Theoretically, TP is the central fund allocated to local government to run functions that belong to the MoH not to finance functions that are the responsibility of local governments (i.e., delegated expenditure). The main reason of using TP is that the centre can control the application of the funds. TP is not allowed by law to finance local functions, but can only finance central functions delegated to the local government on behalf of the MoH (for reasons of efficiency). This is a fungible use of resources, but is nonetheless tolerated. 2.6.1

The flow of data for effective health management

At present, the management information systems which enable effective planning, effective performance management, and regulation, accurate budgeting and reporting are weak. This applies to private sector activity also. Currently, there are (at least) three data-handling units in the MoH: the Centre for Data and Information (Pusdatin); the Information Centre of Health Services Directorate (Pusat Informasi BUK); and the Department for Health Research and Development (Litbangkes). Pusdatin collects data from Puskesmas through Sistem Pencatatan dan Pelaporan Tingkat Puskesmas/SP2TP (the recording and reporting system at Puskesmas level). The data from Puskesmas are sent to DHOs, and then sent from DHOs to the PHO. These data can also be uploaded onto the Pusdatin website by the DHOs. Few data from private clinics and practices are captured. Whilst some data are available on the epidemiological characteristics of Indonesia, financing (including district health accounts) and some aspects of health sector performance, the flows of data are not parallel to the flows of budgets or of authority. Externally, both Bappenas and BPJS have information and R&D resources that also contribute to the knowledge base, but which are poorly harmonised with MoH systems. This has many effects, including a disconnect between performance and sanctions/rewards, a disconnect between the exercise of governance, supervision and regulation and reporting up to the MoH and Bappenas, and compromising planning reliability. There is an urgent need to unify and align health information systems across the sector. Operationally, BPJS is an information-hungry institution. Data are gathered at an individual level and at that level of granularity are used to manage operational transactions (registration, membership card issue, etc.). The data are then deployed in various levels of aggregation for monitoring, contract compliance and statistical purposes. Statistical and performance reporting are treated as dashboards, monitoring performance against a basket of indicators, in some cases in real time. Amongst the dashboards is one that is supposed to reflect satisfaction with 57

Institutional Analysis

the services provided by contractors (providers) retained by BPJS. There are occasional data audits to add to confidence in reporting and planning. One concern is the quality of coding at the clinical level. The DRG-based classification operates at a very high level of aggregation: for secondary care the locally-derived DRG deploys 800 categories of discrimination (compared with 12,000+ in ICD 11), creating opportunities for contestability and frequent opportunities for the “ad hoc” management of outliers. It is reported that the coding is inconsistent across the country, and that the training of coding clerks is often poorly conducted. This will need to be rectified. 2.6.2

Contracting mechanisms

Few mechanisms are available to the GoI for driving the reallocation of resources to create a more equitable distribution of health-care facilities and hence improved access, other than BOK and ad hoc disbursements. However, contracting agreements/service level agreements/partnership agreements can all be used as a mechanism to resolve the tensions between the concepts of decentralisation and central accountability. Generally, contracts will link payment with performance defined according to the nature of the partnership. This is normally a combination of quality, price and access. Usually, they will also include reporting requirements, and a definition of the sanctions/rewards associated with compliance/non-compliance. For a curative clinical service it may be defined as including such issues as the incidence of hospital acquired infection, waiting times, accreditation status etc. For a payment agency it may be defined according to payment intervals, evidence of good governance, or conformity to the business plan. In primary care it might include such issues as the availability of clinical commodities, availability of clinical staff, waiting times, etc. Care should be taken to avoid sanctions exacerbating an already compromised service by catastrophically reducing resources available for service delivery. BPJS contracts with providers for service delivery, and these contracts include agreements about volumes, quality, reimbursement intervals, etc. The remaining challenges include a series of risks, and perverse incentives that will need to be addressed to ensure prudence, quality, and conformity to policy and equitable access. The prevailing perverse incentives and risks are partly a product of the mixed reimbursement mechanisms; personal services in primary care funded by a form of capitation irrespective of activity, curative and secondary care by a form of fee for services based on a variant of DRG, and merit goods by direct financing to local government. •

An incentive to refer “upwards”. Since primary care is not volume-related, there is no financial loss if patients are referred up the system. Simultaneously, hospitals have an incentive to accept those patients because they are rewarded for activity. This means that patients are cared for at the highest cost level, more often further away from their homes, and exposed to the risk of over-medicalisation and over-treatment. BPJS tracks this tendency by continual monitoring of referrals through a monitoring dashboard. It then uses data from across the system to benchmark referrers and providers, and gives feedback about whether patterns of referral imply inappropriate or unnecessary treatments. BPJS assumes that the feedback will suppress the tendency to gaming behaviours (see Annex 4) The definition of 154 disease groups that should be managed at Puskesmas level is sufficiently imprecise to produce contestability and ambiguity. 58

Institutional Analysis

















DRG-based systems are known to encourage “up-coding”. Up-coding is where the coding of a particular patient is treated at the highest level possible. BPJS undertakes audits to ensure conformity to agreements, and will manage disputes around coding by using the independent Clinical Advisory Board. Risk that the pool will be decapitalised. Given the Government’s commitment to reimburse expenditure, there is a risk that the pool will be decapitalised and care will become unavailable. For 2014, this can be managed by the transfer of assets from existing schemes into the risk pool, but in future years fiscal space will need to be created and maintained. Weaknesses in “exclusion” definitions. The entitlements package includes “exclusion” definitions, mainly around aesthetic treatment and self-induced harm. But if eligibility criteria include the psychological attributes of a condition, then aesthetic care could be treated as clinically significant. Similarly, self-induced harm could also include the clinical consequences of obesity, or of sports injuries etc. These issues are contestable and the reimbursement mechanisms would encourage such behaviours. Cost and volume criteria could be over-emphasised. In the absence of a coherent quality strategy, the contracting mechanisms would require contract criteria to focus on cost and volume, and so neglect the important factors of access and quality. More challenging criteria for private sector access. Access to the market created by BPJS will include the private sector, which will be exposed to a pre-qualification process. It is understood that all public-sector providers will be admitted, although technically they could be rejected. Admission criteria for the public sector are less challenging than those for the private sector. This runs the risk that public sector providers could be protected even though they deliver poorer quality services. Private sector hospitals may not necessarily choose to register as providers with BPJS. At the time of writing, hospitals in Jakarta declined to apply to the scheme, citing the poor quality of the DRG-based (INA-CBG) reimbursement mechanism. This represents about half of all private hospitals, and similar patterns are reported in other parts of Indonesia. In primary health care, it is national policy to ensure that half of all contracts are with the private sector. However, neither BPJS nor the MoH maintains lists of private providers. As an interim measure, they “virtually” contract local private provider associations, which then sub-contract to private sector PHC clinicians and facilities. This raises fiducial risks associated with dual practice. Public sector providers may have an unfair cost advantage. Costs for which the private sector will be liable include the cost of capital, while the public sector is able to treat capital as a free good. In the absence of a capital charging regime significant cost advantage will be enjoyed by the public sector in a competitive environment. Dangers of encouraging a widening of the equity gap. In secondary care, a purely reimbursement-based mechanism will encourage a widening of the equity gap since those providers sufficiently well-resourced to deliver high volumes will be able to capture greater “market share”. Those less well-resourced will not be able to invest in service development compared with the more market-active providers.

The contracting mechanism needs to be developed and adjusted to recognise all of these issues above and mitigate their effects.

59

Institutional Analysis

2.7

Health care delivery

Primary and ambulatory health care still represents a low share of health expenditure in Indonesia. More than 60 percent of all funds are spent at the in-patient level, compared with 30 to 45 percent in most OECD countries, where more care is provided at the primary and outpatient levels. Indonesia’s health system outputs and inputs are relatively low. The DPT3 immunisation rate in Indonesia (64 percent) in 2012 is low compared with the East Asia Pacific region (75 percent).36 A similar pattern is observed in the skilled-birth-attendance rate (83 percent), which is far lower than the average for the region and for lower middle-income countries. In terms of physical health system inputs, Indonesia has a low ratio of doctors (13 compared with 16.3 in the region) and hospital beds (10 per 10,000 population compared with 35.5 per 100,000 for its regional peers). The unclear relationship between the PHOs and DHOs is more acutely felt in the referral mechanism. There should be a clear procedure on when patients will be handled by the DHOs and when they should be referred to PHO facilities. A clear mechanism set under the guidance of the MoH would help to avoid an imbalance of patient referrals between facilities. The structure of medical education (mixed public and private) and the mixed responsibility across ministries mean that it is impossible to design a balanced demand–acquisition–supply framework, and to ensure that skills are dispersed appropriately through the decentralised structures in the health-care economy. This is exacerbated by the role of private medical schools. Systems to assure continuing clinical skills are weak and based on exposure to learning opportunities (often linked to promotional activities by drug and equipment suppliers). There are no effective systems to ensure continuing and developing clinical competence and to sanction incompetence. In terms of the distribution of skills, it remains a challenge to deploy relevant clinical skills in the volumes required in some parts of the country. Incentives associated with opportunities to indulge in private practice, high-quality education for school-age children, housing quality etc., all combine to have the effect of making the more prosperous (and so less needy) communities a more attractive operating base for health-care professionals. This issue needs urgent attention. The management and leadership role of provincial governments could be developed. Provincial governments have the potential to be the pivotal instrument of downwards supervision and leadership and upwards reporting. However, it is reported that their capacity is generally limited in terms of both skills and resources. The revised Decentralisation Law should help to remedy the latter through creating support for provincial governors at a policy, planning and advocacy level, and the introduction of the competence framework will have an impact on personal skills. The finalisation and implementation of this framework for new appointees should be accelerated, and the same framework applied to existing post-holders to guide personal development needs. 36

WHO

60

Institutional Analysis

In terms of absorptive capacity, BPJS is registering, on average, 20,000 members per day. Its priority in the short term is to successfully consolidate existing schemes and to recruit a greater share of established schemes. The remaining challenge to 2019 will be to recruit 100 million members who are not currently enrolled in any scheme, and to ensure a stable and predictable payment of premiums. This is extremely ambitious and may not be achievable in practice. Members become eligible for entitlement as soon as they are registered. Premiums are not risk adjusted, and the risk is that individual registrations will be made by patients with pre-existing conditions, members will access the care required, and will then cease their contributions. It is within the basket of powers available to BPJS to make compliance compulsory, but how this will be enacted in practice, particularly for the informal economy, remains unclear. The GoI established the Indonesian Health Workforce Assembly (MTKI) in 2011, which will assume responsibility for the registration of other health professionals (17 professions) from the DHOs. The new institution has experienced some challenges in implementing the registration program. The absorption of all the professions into the framework is a work in progress. The preparation of a competence framework, jointly specified by MoHA (management and governance) and the MoH (technical) is underway. It is not clear whether this framework will be applied only to new recruits, or whether it will also be applied to existing post-holders, so that personal development programmes can be agreed and implemented. The latter option is preferred. Some of the DHOs simply lack the capacity to absorb the management, governance and planning demands placed upon them. The concept of “capacity” here represents skills, systems and sheer numbers. In terms of service delivery, there is an argument in favour of local shared services where scarce skills and capital resources might be shared amongst neighbouring Puskesmas at the sub-district level. This could help to achieve a critical mass of service volumes that would make a service viable, for example speech therapy, laboratory services, clinical psychology etc. 2.8

Equity and access

The scale and diversity of Indonesia, economically, socially, geographically, politically and in health status, means that it is very difficult for the central government to target resource allocation in a way that secures increased equity and access in the more needy parts of the country, and then to assure delivery of strategies to increase equity. The national averages for health indicators mask significant geographic and income-related inequalities within the country. Indonesia is a large, diverse, and geographically-dispersed country. In poorer provinces, such as Gorontalo and West Nusa Tenggara, infant and child mortality rates are four to five times higher than those in more prosperous provinces such as Bali and Yogyakarta. In addition, health indicators for the poor are far worse than those for the prosperous: child mortality rates amongst the poorest quintile in 2013 were three times the rate amongst the richest quintile. In terms of inequity amongst and between income groups, the disparities are even more marked. The Gini coefficient for Indonesia37 was 35.6 and 38.1 in 2010 and 2011, respectively, showing a widening of the wealth gap. International economic crises have had an impact in 37

World Bank, (2012), Databank, http://data.worldbank.org/indicator/SI.POV.GINI

61

Institutional Analysis

Indonesia, leading to an increase in the number of unemployed. This will have an impact on health status, where the ability to maintain the health of households is reduced so that the poor depend more on the Government to meet their health needs, as demonstrated by the utilisation of Jamkesmas. Health insurance coverage rates remain fairly low in Indonesia. About 63 percent of the total population in Indonesia is covered by some form of insurance (as of 2012), which is up from 43 percent in 2010. The major insurers that provide this coverage include: Askes (civil servants and pensioners), Jamkesmas (poor and near-poor), Jamsostek (formal sector workers), Jamkesda (district-level schemes), and private health insurance (most urban coverage). • • • • • • •

Nearly 40 percent of the population remains without any coverage, including millions of Indonesians working in the informal sector. Out-of-pocket spending remains high even amongst those with coverage. Lack of equitable access to quality health services in rural, remote areas. Evidence of considerable mis-targeting and leakages to non-poor families. Low levels of awareness of benefits. Low utilisation of health services. Regional inconsistencies in the availability of the basic benefits package, and poor accountability and feedback mechanisms.

In order to resolve these issues it will be necessary to:29 • • •





• •

Improve targeting for the poor and near-poor, as more than half of Jamkesmas beneficiaries are not from these groups. Improve socialisation (social marketing) to increase public awareness. Ensure supply-side availability and readiness. The experience of Jamkesmas highlights the significant disconnect between entitlements on paper versus what the system will deliver. Ensure sustainability through improving efficiency and effectiveness of implementation. Jamkesmas is entirely financed through central government taxes and premiums are not based on sound actuarial calculations. Supply-side constraints and supply-side subsidies have given a false impression that the financing of Jamkesmas is sufficient. In reality, the programme fails to reimburse the full cost of services and relies heavily on supply-side subsidies. Make provider-payment mechanisms more results-focused. Under Jamkesmas, payments to providers are fees-for-service (including diagnosis-related groups for hospital-based care). At present, there are no additional incentives to improve quality and provider performance. Providers are not given incentives to achieve targets. The BPJS regime will not resolve this issue. Establish a robust and reliable information system to support monitoring and evaluation, and continuously update the calculation of the programme’s costs. Make use of learn lessons from selected provinces and districts/cities that have attained virtual universal health coverage (UHC), such as Bali, Aceh and Jakarta. The Government can learn from these regions’ experiences and estimate costs from existing examples.

62

Institutional Analysis

2.9

Health expenditure

Whilst an element of the capitation formula used to allocate resources to districts was designed to encourage equalisation, the value is marginal. Also, it was not hypothecated (that is, it is not earmarked for application in health care at district level) and therefore application in the health sector was not guaranteed, so governance mechanisms were largely ineffectual. A larger proportion of budgeted public expenditure now appears in district budgets (see Section 0). In part, this merely reflected the transfer of responsibility for meeting salaries of civil servants from central to regional governments. A large proportion of programme operating expenses continued to be provided in a tightly earmarked fashion to regional governments from the decentralised component of the national-level budget. The increase in the proportion of public services previously financed under Dana Tugas Pembantuan (TP) which was controlled by the centre, was converted into DAK and introduced into local budgets. The application of DAK still should be compatible with national priorities similar to the BOK for district/city health and the BOS for education. The greatest share of local budgets is still represented by an inflexible overhead of around 70 to 80 percent, leaving just 20 to 30 percent for public service delivery. This national budget has risen significantly in recent years (albeit from a very low base), largely reflecting additional spending from the decentralised component of allocations and the commitment to provide insurance coverage for the poor. In 2006, Indonesia’s health sector was not heavily dependent on external inflows, which accounted for less than 2 percent of total expenditure. The inflows constituted a large share of public financing (6 percent), and a larger share still of public financing at the central level (16 percent). However, 2002 probably marked a low point in external inflows, which had risen markedly during the economic crisis years of 1997 to 2000 before declining. The inflows rose again in subsequent years, with the inception of new sources of funding, such as the Global Alliance for Vaccines and Immunisation (GAVI) and Global Fund to fight HIV/AIDS, Tuberculosis and Malaria (GFATM). Health-care markets are notoriously inelastic, where need does not necessarily translate into demand, demand does not necessarily translate into supply, and supply does not necessarily translate into price and cost control levers. Therefore, markets may not balance supply, demand and price effectively. The immediate impact of this is an increase in demand-side forces, expressed as growth in the use of private-sector services, together with the increased registration of beneficiaries in the various social welfare programmes, is creating expanding health-care demand. Consequently, the health care system must be able to absorb a greater volume and be able to co-ordinate and manage a developing mixed-economy of care. 2.10

Political commitment

The Preamble of the 1945 Constitution clearly mandates that the purpose of the establishment of the Republic of Indonesia is to protect all people and the entire country of Indonesia. Article 28H (1) Amendment IV of the Constitution of 1945 states that every person has the right to live prosperous physically and spiritually, have a proper home, a good and healthy living environment, and receive adequate health services.

63

Institutional Analysis

The WHO Regional Meeting on Revitalising Primary Health Care held in Jakarta in August 2008 resulted in the articulation on the need for Primary Health Care Reform including Universal Health Coverage, Health Care Reform, Public Policy Reform and Leadership Reform. The current conditions of increasing demand (created by UHC, the epidemiological shift, and other factors), increased supply (created by strong private-sector growth, capital investment in secondary care, and the development of Puskesmas) will lead to increased global costs. Whether Indonesia has sufficient fiscal space to absorb this growth within the planned increase in expenditure remains to be seen. If not, then the responses available to Government all have political ramifications. These ramifications include the following: •



• •



Increase in tax revenues to finance greater health expenditure. This would be politically unpopular and difficult in the formal sector, unless applied through increasing consumption taxes, which has a regressive effect. Increase in the contribution made by individuals and/or by employers. Once again, this would be disproportionately borne by the formal sector, and a disincentive to employment. Increase revenues through increased user fees/co-payments. This would reduce access to those in greatest need. Increase the share of public expenditure allocated to health care, therefore reducing the share allocated to other programmes. This would create government-level turbulence and would be managerially challenging given that only 60 percent of health expenditure is spent directly through the MoH/BPJS. Create new disposable resources through securing greater efficiencies. Efficiencies could be secured through reducing overhead costs, encouraging changing patterns of care via a contracting regime, and managing contract-pricing based on achievable performance rather than expenditure-reimbursement.

Any one of these, or any combination of them, would require political drive and courage to implement. In addition, it is possible for provinces and districts/cities to deploy internally generated revenues in order to subsidise care for their citizens. However, there is little political incentive to do so, as it could have the effect of aggravating inequities and the least needy communities would likely experience the greatest benefits. 2.11

Organisational development

The MoH is a large organisation that combines many roles and types of role. It is simultaneously regulatory, executive, supervisory, strategic, operational, technical and supporting. Its advocacy role is confused by its supervisory role. But the MoH is not alone in this: there is some evidence that ministries of health worldwide have also been unable to cleanly distinguish the policy/strategic role from the executive role.38 The MoH has hierarchical legitimacy (over sub-national institutions) but very limited means to exercise authority. Sanctions (in the form of withholding of support) are the principal means of influence. Initiatives such as service standards, clinical guidelines, quality initiatives etc., no matter how respectable and rigorous, are not directly enforceable at sub-national levels of government (or service delivery), or within private/non-government institutions. Furthermore,

38

Ham, C, (2000), The Politics of NHS Reform 1988-97: Metaphor or Reality?, King's Fund, London

64

Institutional Analysis

the separation of supervisory functions between different ministries based on technical, legislative and financial will create confusion (and possibly conflicts) over the priorities at the sub-national level. In a turbulent environment (UHC, recent elections and a new incoming government) where the tensions between policy, politics and management of the SKN could mean further reorganisation in the future, the MoH will also be less able to steer the system than before. In addition to the more limited role and capacities of the MoH, the empowerment of agencies, partner agencies and regulators such as BPJS, creates potential for incoherence, duplication, and territoriality at the centre. All these challenges come without taking in account the vertical relationships with sub-national government institutions. What is presented as the decentralisation of power away from the centre in effect amounts to a form of centralisation, partly through the powers of patronage exercised around the MoH’s financial support for under-resourced districts/provinces. Within the principal organisations there are further complexities. The MoH is reported to have 51,986 persons on its payroll.39 This represents a heavy superstructure and is a product of the wide range of roles of the MoH, its complex relationships and overlapping internal functions. Management costs are notoriously difficult to calculate, depending on contestable definitions of management, manager and the share of time devoted to management/technical duties for those who have both. In Indonesia, a relatively clear calculation is that represented by “administration”.40 The notion of “administration” is not without anomaly, but the reported 26 percent of expenditure going to administration is a high proportion by any standard, even taking into account anomalies of where specific expenditure lines are coded. The MoH recognises this and is taking steps to reduce the headcount through a process of natural wastage (not replacing leavers and retirees). However, this means that it will take a very long time to make substantial inroads into the administrative burden on the SKN. It also means that the ultimate shape and skill mix within the MoH will be a product of the demography of the MoH. Without destabilising large-scale re-skilling and redeployment, there is a risk that the MoH will become an organisation misaligned with its purpose, challenging its “fitness for purpose”. However, it would be useful to agree on a definition of what constitutes “management” expenditure so that it can be measured and managed in the future. A description of the existing structure of the MoH, together with a description of the planned new structure, appears as an Annex. The main conclusion is that the new structure is more streamlined, is more closely aligned with the “Life-Cycle Approach” and the “Building-Blocks Approach”. However, it seems to pay less regard to the realities of current, known factors including how it will work with BPJS, the impact of the revision of the decentralisation laws and its own developing role. It is anomalous that the MoH directly manages some direct personal care services (Class A and some Class B hospitals), for a number of reasons: the skills required are not those typically found in a civil service environment; there are conflicts of interest in the sense that there are conflicting roles as regulator and provider; and the role of the MoH as an advocate and 39 40

MoH (2014) MoH (2013), National Health Accounts.

65

Institutional Analysis

ministerial adviser does not sit comfortably with the role as service provider. For these reasons, the MoH should surrender its role as a direct provider of personal health services to an alternative agency. In management terms, the MoH suffers from long lines of communication and so a lack of agility. It also suffers from an absence of professional leadership across the professions, country-wide. It would therefore be a step forward to appoint a Chief Medical Officer, a Chief Nursing Officer, a Chief Pharmaceutical Officer and a Chief Scientific Officer, whose roles would be around the development of their professions rather than as ministry functionaries. Their roles would be to lead the development of guidelines (clinical, quality, HTA. etc.) to lead the regulation of the professions and to advocate on behalf of those professions (in a technical sense rather than in the sense of pay and terms of employment negotiators). BPJS represents a different proposition. BPJS is an executive agency of government and managed at “arm’s length” from the machinery of the civil service. In legislative status terms it is the equal of the MoH. Its role in the short term will be to consolidate and manage the existing schemes operating throughout the country at the sub-national level over the period to 2019. This will provide a platform for the implementation of UHC, including administrative, data management, recruitment and quality management. All of these roles are largely apolitical. Outstanding challenges for BPJS include covering the remaining informal sector of over 70 million people, and unifying over 360 Jamkesda schemes that vary in benefits, eligibility criteria, management and payment systems. BJPS has deployed and expanded the systems developed as PT Askes (particularly registration and IT systems), and so has a platform from which to build further capacity and volume. It remains to be seen whether these systems are robust enough to be able to absorb the additional schemes and the many millions of individual subscribers expected to be recruited. 2.11.1 Efficiency and planning Two issues affect efficiency: first whether what is being done should be done and, second, whether what needs to be done is done well. Several issues relate to inefficiency in the MoH. First, it is inappropriate that the MoH is a direct provider of personal services. This should be delegated to an agency. Second, the “ad hoc” movement of money (both revenue support and capital) is conducted on a “grace-and-favour” basis, distorting priorities, encouraging fungibility at the district level and consuming energy and effort. Third, the annual budget-setting process is a negotiation process at the Echelon 2 level. The linkage between budgets and policies, priorities and plans, is at best opaque and at worst weak. The product of the second and third issues is that allocations are inconsistent, undermine the search for equity, and reward the inefficient. These two issues could be regularised by formalising the process in a consistent and transparent way, by making allocations to provinces/districts (which are irregular or “ad hoc”) and internally within the MoH subject to a robust business case-making process. In terms of internal efficiencies, there is evidence of duplication, especially around the information/research/M&E functions, and around the various aspects of the finance function. These could, and should, be streamlined. In terms of operating efficiencies, assuming a more rigorous contracting process is introduced the PHOs should be required to produce and submit business plans every three years, leading to annual operational plans. In planning terms, these would become the currency of 66

Institutional Analysis

supervision, in the same way that contract performance is the currency of management. However, this would require support in terms of developing capacity/skills in business planning. Similarly, districts should have the same relationships, but with PHOs. The superstructure of the MoH has been referred to earlier. Whilst it is difficult to determine management overhead precisely, the reported figure of 26 percent representing “administration” is, by any measure, very high. Internationally, management overheads in health care systems of 10 percent are considered high, and a more typically level is 7 to 8 percent. Therefore, efforts should be made to reduce this high administrative burden. However, using only a process of “natural wastage” will take a very long time to have any significant impact on management overheads. In terms of health-care practice, there is evidence of inefficient operating practices, including, for example, a very high dependency on in-patient care, a very low level of use of 24-hour surgery, high drug costs, and a reimbursement regime that is likely to aggravate this. Three mechanisms would help manage these inefficiencies: first, a contracting framework encouraging and rewarding efficient clinical practices, second, a set of credible and agreed clinical guidelines for the management of the major causes of morbidity and, third, unifying the various drug lists into a common approved formulary would help to manage drug costs. Budgets have sometimes been released and implemented as late as 8 months into the financial year, creating significant problems in financing care in the early part of the year, and leaving limited time for rational spending in the final few months of the year. This phenomenon will need to be managed until such time as BPJS is the sole payer. In terms of planning, as mentioned, BPJS and the MoH have similar statuses in the hierarchical firmament of central government. Mechanisms to harmonise high-level MoH policy with service delivery determined by BPJS, such as provider contracts, are unclear. Likewise, how the Government can discriminate in favour of particular regions, or populations or socio-economic groups, is unclear. Obvious mechanisms include, on the supply-side, higher tariffs for certain regions, specific health conditions or certain citizens. On the demand-side, the obvious mechanism would be through a premium subsidy. In both cases, it is unclear how the practical implementation of such strategies would operate. The scope of the Primary Health Care Reform includes the development of strategic policy and evidence-based planning to ensure the implementation of efficient resource allocation, financial efficiency to obtain the best results at the lowest costs, the establishment of appropriate institutions with proper function and scale, and the availability of data that are valid and reliable, as well as the governance and supervision of reliable health plans and investments. These reforms would help to achieve a more inclusive and equitable health status. In addition, the revitalisation of Puskesmas as part of the Health Systems Strengthening based on Primary Health Care Reform is a necessity focusing on institutional, management and provision of health resources.

67

Institutional Analysis

2.12

Service quality and standards

Although quality policies, methods, organisations and projects have been initiated, they are not widely recognised or integrated in a national quality strategy. As a result, they create little impact in improving quality of health care in the country. At present, the PHOs and DHOs licence health providers, but the licensing process appears to be inconsistent and of limited scope. It is also inconsistent with their roles as service providers. There is no precondition that providers are accredited before they are allowed to provide services. However, there is progress in accrediting PHC facilities, although little progress so far in accrediting hospitals and the private sector. There is no effective mechanism to enforce private-sector conformity to licensing requirements. However, BPJS applies a checklist of characteristics before admitting providers onto its list of approved providers (described as “credentialling”). The checklist is based on the minimum service standards (SPM) and is mainly focused around inputs such as space, equipment, etc. Although the concept of MSS/SPM is taken seriously41 (DSF, 2011), a recent study42 concluded that SPM appear to have had little effect so far. With regard to nationwide effects of SPM, the 2010 World Bank Report on Maternal Health 43 summarises that “the MSS/SPM remain illdefined, complex to measure and few districts actually apply them”. Our own investigation in 2011 also suggests that both at the central level and at lower government tiers many ambiguities concerning SPM remain. Compared with private competitive markets, the problem of public-service delivery is that public-service providers are not directly accountable to consumers. This problem is known as the “long route of accountability.” Voters (as well as the central government) have to hold local politicians accountable for the allocation of the local budgets, and in turn politicians have to hold local public-service providers accountable for delivering the services. 44 If one of these relationships does not work properly, public-service provision may become dysfunctional. Theoretically, the concept of SPM might be a solution for these problems, because SPM indicators should focus on the output of public-service delivery. The performance risk of local public-service providers will shift to local policymakers, because now politicians are not only responsible for the allocation of the budget, but also for the output of public services. Since local policy-makers’ performance evaluation will be based on MSS/SPM, they have to hold service providers accountable and the “long route of accountability” shortens to a voterpolitician relationship (or to a central government-local politician relationship). One phenomenon is that line ministries tend to set “over-ambitious” minimum service standards (DSF op. cit). Development agencies and academics criticise this practice, because it is unclear how these SPMs should be financed, and because once minimum standards have been achieved there is no incentive remaining to further develop services. It is therefore suggested that providing incentives for SPM implementation for local governments would improve

41

Decentralisation Support Facility Dixon, G & Hakim, D (2009), Making Indonesia's Budget Decentralization Work: The Challenge of Linking Planning and Budgeting at the Local Level. International Public Management Review (10/1): 119-169. 43 World Bank (2010): Indonesia Maternal Health Assessment, February 2010, Jakarta 44 Ahmed, J (2005), Decentralization and Service Delivery, World Bank Policy Research Working Paper (3603), Washington, DC: World Bank. 42

68

Institutional Analysis

implementation. Therefore, central government should provide financial incentives that encourage competition. Moreover, it should publish MSPM achievement data in order to increase voters’ knowledge on local government performance. 2.13

Summary

Remaining challenges largely concern: equity (geographical, health status, economic and access), the capacity to absorb growth (both supply and demand), the clarity and focus of the roles of institutions, technical capacity (information systems, managerial and clinical skills), accessing “hard-to-reach” groups, co-ordination between institutions and engaging with the private sector. The suggestions made here will not eliminate the remaining challenges, but will go some way to helping to manage their impacts.

69

Institutional Analysis

3.

New and emerging challenges

3.1

Introduction

The health-care system is now entering a period of turbulence that may not have been fully predicted at the time of the last HSR. This turbulence includes (i) the rapidly shifting epidemiological balance across the country, leading to the need to manage a higher prevalence of chronic and continuing conditions, and (ii) the roll-out of UHC to reach a further 100 million people within the next five years. Those 100 million include a large number of “hard-to-reach” groups, particularly those working in the informal sector of the economy. The shift to a reimbursement mechanism for financing health-care costs, political turbulence and uncertainty created by the legislative and presidential elections in 2014, the new decentralisation framework, and the shift in the role of the MoH, all create a hard-to-manage environment. Finally, the expectations of large-scale growth in health expenditure (growth of over 25 percent over the next five years), together with the factors above, will challenge the SKN’s capacity to absorb and manage the operating changes required if the national strategy is to be delivered, particularly in the context of relatively inflexible government structure. Across all areas of the landscape is the pressing issue of creating an effective performance management process in conditions of a complex mixed-economy of health. This complex environment involves, both conceptually and practically, inherent tensions in many institutional relationships, risks associated with gaming and fungibility, unclear roles [for example, (i) how do MoH strategies and plans get translated into BPJS actions? (ii) how is new capital investment developed? (iii) how are new or innovative services established before reimbursement can begin to flow?], and questions over how the system will deal with the challenges created by UHC, growing supply, and growing demand. An evaluation of fitness-for-purpose requires unequivocal clarity in our understanding of “purpose”. Performance management needs to address the following: • • • • • • • • • •

Clinical efficiency. Clinical safety. Clinical effectiveness. Productivity. Procurement. Managerial competence. Leadership competence. Financial management and prudence. Staff productivity. Community engagement (including the patient/service user experience).

These are system-wide challenges and need to be faced with system-wide strategies, requiring at a minimum a definition of what “performance” is understood to represent at different points in the SKN and beyond. Some interventions are already in place or en-route. For example the forthcoming decentralisation law will clarify the roles of different government tiers in the health care system. 70

Institutional Analysis

Any model of pay-for-performance, or payment-by-results, or performance-based-financing requires several characteristic features to make it viable, as listed below: • •





3.2

Performance must be defined, measurable and unambiguous. Performance delivery must be in the hands of the person or institution being rewarded or sanctioned, otherwise rewards or sanctions are the products of events beyond the control of the person/institution, and so becomes a random and demotivating organisational feature. The organisation or individual must be equipped to deliver to high-performing levels, including the availability of skills, competence, resources, information systems, authority and opportunity to fail safely. The value of the sanction or reward must not compromise the functioning of the organisation. Financial sanctions that are too high will withdraw resources from the service, while sanctions that are too high will be demotivating. In addition, rewards that are too high will create financial instability and lead to gaming behaviours aimed at maximising those rewards.

Social determinants of health

The burden of disease is changing in volume, type and social/macro-economic characteristics. The “demographic dividend” (the point at which the benefits form the economically active members of the community overtake the cost burdens represented by the economically inactive) will be realised if that population is able to secure economically gainful employment. The impact of this is likely to be an increase in demand for privately provided curative heath care. If proposals are introduced requiring employers in the formal sector to contribute to health insurance, this could encourage employers to “casualise” their workers, possibly by using zerohour contracts. The propensity of employers to “casualise” the workforce to avoid their contributions has the potential to increase the size of the informal sector, thereby reducing the size of the formal sector and increasing the number of citizens in the “hard-to-reach” group, thereby making the rate of replenishment of the risk pool less predictable. As a consequence, health-care costs would fall more heavily on those in greatest need, and would increase the risk of exposure to catastrophic health spending. The Government has recognised this risk and has reserved the right to alter the balance of personal/corporate/government contributions to the risk pool. The Government will need to remain vigilant in order to ensure that this balance is managed to provide access to quality health care on the one hand, and fiscal and market prudence on the other hand. 3.3

The relationship between institutional arrangements and health outcomes

From early 2014, curative activities or personal health treatment will be paid for through BJPS. Since the costs of health treatment will be borne by BPJS, primary, preventative and promotive functions will be more central to the main role of all tiers of government. Empirical studies show that health promotion activities are more susceptible to unclear distinctions between providers, and attract a lower priority at all levels than policy would suggest. There is a need, therefore, to clearly distinguish between tiers of government and their respective roles in health promotion, and to establish SPM standards that encourage 71

Institutional Analysis

appropriate levels of expenditure. The MoH should design a comprehensive policy and strategy for health promotion based on its responsibility for the provision of public merit services. For example, this could include a national policy on tobacco use and on safe sexual practices. Health promotion should also be conducted locally for locally-specific action, such as generating local demand for services such as immunisation, ante-natal care and childhood surveillance. The MoH should also give sufficient opportunities to provinces and districts/cities to provide health promotion services without overlap. Currently, expenditure is heavily biased towards curative services and capital acquisition. In seven sample provinces, 45 total health expenses in 2010 were disposed of as follows: Table 10: Range of expenditure in 7 sample Provinces by disposal (2010)

Health Promotion;

3% - 12%

Curative

30% - 66%

Salary and Infrastructure

30% - 60%

In the revision of Law No. 32/2004 on Regional Government, responsibility for health promotion activities will be distributed across central and district/city governments using “scale” as the determining feature of responsibility. Health promotion in provinces and districts/cities will be implemented by the health Dinas and also by Puskesmas. However, the word “scale” is imprecise and could create opportunities for contestability, overlap and gaps. Since the revision of the law will be followed by a government regulation to replace PP No. 38/2007, the roles of each tier in health promotion should be clarified to avoid confusion and overlaps. This clarification remains “work in progress” at the MoH. 3.4

Health sector oversight and governance

Since 2001, Indonesia has gone from being one of the most centralised countries in the world in administrative, fiscal, and political terms to one of the most decentralised. Under this process, local governments have assumed new responsibilities that were previously covered by the central government, as well as managing new financial resources that have been transferred from the central government or raised within their own localities. Local governments have therefore experienced substantial increases in financial resources, mostly through increased transfers from the centre for the delivery of education, health, and basic infrastructure services. Indeed, local governments now manage 38 percent of total public expenditure and carry out more than half of all public investment. However, public expenditure reviews show that fiscal decentralisation in Indonesia still faces many challenges. These include improving and speeding up spending, and reducing local government dependency on central government transfers.

45

Ascobat, 2010

72

Institutional Analysis

3.4.1

Decentralisation and UHC

The general decentralisation process implemented in 2001 has had many impacts on the health system, even though it was not designed specifically with the health sector in mind. In particular, health financing, health information systems, human resources for health and service provision have been affected. Under decentralisation, responsibility for health-care provision is largely in the hands of sub-national government institutions. Indonesia has adopted broad autonomy as stipulated by Article 18 of the Constitution. The meaning of broad autonomy is a blurred concept. In practice all government responsibilities except defence, police, fiscal policy, foreign affairs, justice and religious affairs are devolved to sub-national governments, both provincial and district/city. In all, 31 functions have been devolved, giving sub-national governments one of the broadest ranges of autonomous power in the world (the so-called “big-bang” policy). Indonesia is amongst several low- to middle-income countries aiming to improve their health financing systems and implement universal health coverage (UHC) so that all citizens can access quality health services without exposure to catastrophic health expenditure (CHE). The 2005 World Health Assembly Resolution (WHA, 2005) and the 2010 World Health Report (WHO, 2010) provided momentum towards this goal, with a focus on social health protection and equity in access to health care (WHO, 2010). Momentum increased with the release of the 2010 World Health Report on universal coverage and the approval of the United Nations General Assembly Resolution on UHC in December 2012. The movement towards UHC is becoming a key focus of the post-Millennium Development Goals (MDGs) development agenda (Latko, Temporao et al, 2011; Vega, 2013). Challenges to UHC in Indonesia include: a fragmented health financing system; political turbulence and uncertainty; decentralisation; the demographic transition; high out-of-pocket spending; and low levels of government spending on health. Indonesia also fares relatively poorly in achieving the health MDG, with the least progress being made on MDG 5 to reduce maternal mortality by three-quarters (Rokx, Schieber et al, 2009). 3.5

Cross-sector relationships, coordination and accountabilities

Several government agencies have a material influence on the relationships with health development and operation, including BPJS, the MoF, MoHA, Parliament, the MPW, Bappenas, the Vice-President Office’s National Team for the Acceleration of Poverty Reduction (TNP2K), and the National Population and Family Planning Board (BKKBN). How population-level health gain will be secured through harmonising and targeting health-giving activities across all sectors of the economy falls within the remit of Bappenas. There is a well-established and understood set of complex relationships between health outcomes, health-care systems and the wider interconnectivity of elements of the political economy of countries. Health outcomes are a product not only of the architecture of health-care services, but also of issues around education, economic security, and economic parity/social solidarity, all of which combine in a complex adaptive system.46 One significant impact of this

46

Marchal, B et al (2014), Complexity In Health: Consequences For Research & Evaluation, European Union Seventh Framework Programme (FP7/2007-13), Grant Agreement no 261449, FEMHealth project.

73

Institutional Analysis

for the HSR is the reliability of interventions not only in institutional design, but in all aspects of health-care intervention. Health is a product not only of health care, but also of other systems that affect population health. For example, very substantial health gain has been achieved through the impact of the fluoridation of drinking water, through making the wearing of car safety-belts compulsory, through legal intervention to prevent using mobile telephones whilst driving, and through tax disincentives to reduce smoking, etc. The obverse is similarly an issue. The role of the health-care system in supporting the interests of ministries and agencies, for example in the disposal of hazardous waste, the impact of health-care capital developments on traffic volumes, the impact of health-care capital developments on utilities (water and utilities) and, most significantly, the relationship between education and health and the supply of trained personnel, and the provision of training placements for students and trainees (not just in the clinical professions). If epidemiological research undertaken by the MoH has provided evidence that there are costeffective non-clinical solutions to medical problems, then it is not clear how the MoH can lever action and expenditure from partner ministries/agencies, such as the MPW, Parliament, or the tax authorities. If the issue were one of sufficient political profile, it could be resolved/co-ordinated at cabinet or presidential level, but this seems an uncertain “backstop” to depend on. An alternative strategy could be to invoke the involvement of the Co-ordinating Ministry for Social Welfare but, again, it is unclear what leverage might be applied. Restructuring of the MoH and general bureaucratic reforms across government could provide opportunities by creating a planning mechanism that requires co-ordination. Alternatively, the BOK (or a BOK-style fund) could be modified/extended to create a planning mechanism to undertake joint activity. If health experience and health care are considered to be significant elements of human development, then a wider network of inter-connected factors that deliver health benefits or health hazards needs to be recognised and considered. Some governments (admittedly none of the scale or complexity of Indonesia) have engineered planning mechanisms designed to reflect and capitalise on the interconnectivity. For example, the Governments of Kazakhstan and of Wales have designed cross-sector planning mechanisms that require ministries to identify ways in which their activities can contribute to the achievement of the objectives of other ministries. This could be a model adopted by Bappenas to promote “joined-up” national development. However, whilst Bappenas has the capacity and competence to undertake highquality R&D and planning, it is less able to intervene in the mixed economy of care to lever adjustments to the ways in which that economy works. The private sector is a major player in some parts of the country (especially the major conurbations) and is virtually absent in other parts of the country (mainly the economically vulnerable rural communities). Single-provider practices are widespread, and publicly-employed clinicians often also operate in a private setting. It is MoH/GoI policy to encourage the development of a mixed economy of care to enhance quality and price suppression through the exercise of a quasi-competitive market, and to give access to private capital. As the health market develops, it will be necessary to enable and encourage private-sector provision by 74

Institutional Analysis

neutralising competitive disadvantages experienced through differential criteria for admission to providers’ lists and reducing the economic burden on private-sector providers of capital financing. There are three ways of achieving this. First, tax benefits could be provided to the private sector to compensate for the cost of capital. Second, a capital charging overhead could be applied on public-sector providers, reflecting the public dividend capital invested in public assets. Third, a differential tariff could be levies to reflect the lower costs experienced by the public sector. These three strategies would have a beneficial effect on the vibrancy of a well-mixed local economy of health. However, in those areas where either the public or private sectors are virtual monopolies, the strategies would have disadvantages. They could either increase the costs of the public sector with no productivity gains, or create risk-free publicly-subsidised businesses in the private sector. Finally, co-ordinating the life-cycle approach with the cross-cutting ambitions of Bappenas, together with the need to translate both of these into short-term expenditure programmes, will be challenging. Not only will there be challenges in inter-organisational relationships, but also challenges in the technical capacity needed to model the financial flows couple with programme objectives. 3.6

Transactions

Given that the national budget (APBD) will continue to be the main source of funding for most health-care costs, budget from the centre may not be translated into direct expenditure once the funds have been processed through provincial and district/city treasuries. The MoH has no mechanism to intervene strategically and correct for the chronic imbalances in health expenditure locally, especially in those areas where fiscal space for local correction is limited. As things stand, BPJS also has no mechanism to correct for inequities, since its role is currently limited to reimbursing variable health costs (5 to 10 percent). The funding process currently delivers funds for health care from the centre via provincial or district/city treasuries. Health-related income is considered part of the general revenue stream flowing into provinces or districts/cities, especially when such small shares of expenditure come from locally generated income. There is no mechanism to ensure that the income stream associated with health care is actually used for health care. Therefore, the theoretical funding of health care may not be actually deployed at the point of delivery. Financial information systems do not provide a good understanding (at the centre) of how resources are deployed in practice, and what health benefits or activities derive from them. However, it is widely believed that health expenditure at the local level is less than the expected allocation. So, there is no mechanism to pledge the health-related share of central funding to the local level (in other words, an earmarked allocation from central government to local government does not necessarily translate into a ring-fenced budget commitment in local government expenditure outturns), and there is no mechanism to link allocation with performance, or to make the release of funding conditional upon performance, efficiency or allocation. This has the perverse effect of rewarding inefficiency and penalising efficiency. However, if central government pledges funding to provincial or district//city governments, or BPJS disburses funds directly to health-care providers, then this undermines the principle of decentralisation. This is a 75

Institutional Analysis

dilemma that needs to be resolved by matching delegated power with accountability, captured through a mechanism such as service-level agreements (SLAs) or performance contracts in which all parties share accountability through the agreement or contract conditions. In primary care, the revitalisation of Puskesmas is an integral part of the reform of health care, i.e., the strategy, the level of care, and the package of activities. While simple reforms at the Puskesmas level will have no impact on health development, a major transformational revitalisation of DHOs/PHOs (even the MoH), referral systems to secondary and tertiary hospitals, inter-sectoral collaboration and networking, community participation and provision of the building blocks of the health systems could have substantial impacts. Such a systemic structural reform should reflect the following principles: •



• •

Universal health coverage (health equity, solidarity, social inclusion): everyone has the same right to health, including health services, in the sense of equal rights in accordance with their needs. Service delivery (people-centred care): everyone should be viewed as a whole, including the family and the environment, and services provided in a comprehensive manner— promotive, preventive, curative and rehabilitative—to ensure that the referral system is working. Public policy in the sense of health development is promoted not driven by the protection of existing patterns of care delivery. Effective leadership towards reliable health-care delivery.

The nature of the diagnosis-related group (DRG-based) system for reimbursing secondary-care activity, together with a population capitation-based mechanism for Puskesmas and BOK for health promotion, means that there will be an incentive for upward referral of patients. This will likely lead to upward cost pressures, patients being cared for further from home than necessary, and “gaming” (referred to earlier). 3.7

Health care delivery

As primary care is developed, and direct financing is implemented, there is a risk that “weaker” Puskesmas will have no mechanism to rectify their weaknesses. Equally, BPJS will have no mechanism to judge quality. Therefore, it is in the interests of central government, sub-national governments, commissioners (that is, those individuals and institutions – generally fundholders – who commission providers to provide specified services) - and providers to be able to judge the vulnerability of Puskesmas that do not have sufficient scale to be autonomous (unlike hospitals) so that they can assume the expanded and developed role that national policy envisages. One way of categorising the strength of Puskesmas is to make judgements on the basis of seven areas of interest. Many metrics can be devised to reflect strength/weakness in these areas, and an illustration is summarised in an Annex 5.

76

Institutional Analysis

3.8

Equity and access

3.8.1

Economic equity

Kristiansen, S. et al47 state, “The main conclusions are three-fold: the local administration of health-care services is without transparency and accountability, health centres are turned into profit centres, and the increasing roles of private actors tend to reduce concerns over preventive health care and the conditions for poor people”. The authors suggest that increased public spending would redress inequities arising from regionalisation. The impact of BPJS reimbursement of funds for marginal expenditure is that it might reinforce or even exaggerate inequity across Indonesia. If the APBD continues to be the main source of central funding, and local governments are able to allocate funds as they judge appropriate to local needs, BPJS will have no mechanism to correct inequities for the bulk of the expenditure (salaries and wages). 3.8.2

Service access

As insurance coverage increases, so the “insurance effect” is likely to increase demand for services (ambulatory, in-patient, pharmaceuticals). Given that service distribution is already inequitably distributed across the country and that areas of poorer public provision are generally (but not entirely) those that also experience poorer levels of private provision,48 there is a risk of a widening gap between supply and demand. The impact of this is twofold. First, those who are most in need will have a relatively poorer experience of access to care than expected. Second, insurance premiums will produce fewer benefits/less value for those in the less well-provided areas. Solutions to this could include increasing capital investment (referred to earlier), increasing the deployment of human assets, and enabling/encouraging the private sector to develop services in under-served areas. 3.9

Health expenditure

The flow of financial resources between elements of the health-care sector and the framework determining the value of the resources is opaque. Whilst theoretically there is a populationbased allocation system, in reality a large share of the allocation is a product of historical and/or arbitrary agreements. This situation will become less predictable in the future as an activitybased reimbursement mechanism is introduced from 2014, implying post-hoc expenditure (at least for marginal costs), together with a population-based capitation system for primary care. Furthermore, there seems to be no mechanism to reflect referrals between parts of the system. For example, a referral out of primary care into secondary care would not bring with it a reimbursement from primary care reflecting the avoidance of expenditure. So, there is an incentive to refer, and no financial incentive to retain, an episode of care at the lowest cost and most geographically desirable level. The General Purpose Grant (DAU) for non-personal services allocations is processed through provincial or district/city treasuries and is counted as a part of the general income stream into a

47

Kristiansen, S et al (2005), Surviving decentralisation? Impacts of regional autonomy on health service provision in Indonesia, Health Policy 77 (2006) 247–259. 48 National Institute of Health Research and Development, Ministry of Health, Indonesia & World Bank (2013), Assessing Supply-Side Readiness for Universal Health Coverage in Indonesia, Jakarta, Indonesia

77

Institutional Analysis

province or district/city. There seems to be no mechanism to ensure that these funds go into health care, apart from BOK. 3.9.1

Reimbursement and disbursement of funds

A significant challenge for all partners in the health-care landscape is the management of cash flow following the introduction and growth of BPJS coverage. How exactly UHC will work in practice remains unclear at the time of writing. The expectation is that BPJS will reimburse secondary-care providers for marginal costs based on a DRG-type tariff. Thus, the bulk of expenditure will continue to be made through ABDP. It is also expected that Puskesmas will continue to be funded through central allocations disbursed via BPJS, associated with population characteristics rather than through service outputs. There are two issues here. First is the actual cash flow into curative services from BPJS to the end-user, and whether this will be timely and accurate. Second is the risk of fungibility in the deployment of funds from local governments to Puskesmas, both in terms of volume and timing. One significant effect of these issues could be an incentive to refer “up the system” to avoid expenditure locally. This has two undesirable consequences: first, it runs counter to MoH policy to build primary-care-driven SKN, and second, it pushes care to the more expensive levels of delivery. Associated with DRG-based systems are opportunities and incentives for “gaming” behaviours, such as up-scaling of codes, incentives to “over-treat”, treating at the highest level of cost within the SKN, etc. Districts/cities and provinces will need to be vigilant in auditing these processes, and will need a mechanism to intervene in order to mitigate the negative effects of DRG-based reimbursement mechanisms. BPJS will need to reach agreement on adjustments to the payments for each region. The MoH and BPJS will need to generate policies to guide payments, including determining the upper limits for the various regions. The process for these negotiations has yet to be devised, including who will be the final decision-maker in the event of an impasse. The process for integrating regional funds under BPJS management remains unclear. Jamkesda funds currently cover an estimated 31.6 million people from around 350 districts/cities. All provinces have schemes, with the exception of Gorontalo, Papua and West Papua (ILO, 2012). Given that many local governments already contract PT Askes to manage their health insurance schemes, there is an assumption that many regions are already prepared for integration into BPJS. However, many regions may be reluctant to surrender their current schemes. Potentially, the recommendation for local governments to reinvest the funds previously allocated to local insurance into their health facilities and personnel to improve quality, coupled with the expectation that the national insurance scheme should bring better coverage than Jamkesda, may encourage local governments to embrace the scheme. The extent to which regional governments have been consulted during the development and planning of the scheme is also unclear. Anecdotally, concerns among sub-national stakeholders over the centralised management of the scheme have been reported.

78

Institutional Analysis

The Government plans to negotiate the payments with provider representatives (currently drawn exclusively from the public sector), and these payments are expected to differ according to geographical region (given regional variations in expenses) for both public and private facilities. However, it is not known if the differential payments will also include an element for reducing inequity. Private-sector representatives should be included in the negotiations. Figure 8 shows how the share of the health budget is allocated to MoH and non-MoH public sector bodies. It excludes out-of-pocket payments to public and non-public providers.

Figure 8: Allocation of public health spending

Clearly, the relationship with other government agencies is exercised through financial allocations, but which areas and in which ways expenditure occurs will require a degree of sophistication in the negotiations. For example, this could involve coordinating MPW expenditure with DAK expenditure, or managing APBD and MoF linkages, etc. In practice, budget releases down to districts/cities are sometimes implemented as late as 8 months into the financial year, creating significant problems in financing care in the early part of the year, and leaving limited time for rational spending in the later months of the year. These relationships have the potential to create tensions at political and operational levels. 3.9.2

Specific Purpose Grants (DAK)

Most intra-governmental transfers to local governments are made through the General Purpose Grant (Dana Alokasi Umum, or DAU) and the Specific Purpose Grant (Dana Alokasi Khusus, or DAK). DAU is mainly used to fund salaries and other administrative costs, but will be reduced by the development of the BPJS scheme, which will be responsible for disbursing INR26 trillion. DAK is allocated to finance specific investment expenditures that are aligned with national priorities and carried out under the jurisdiction of local governments. In 2010, the DAU share of total intergovernmental transfers was 63 percent, while the DAK share was around 7 percent. There are currently 14 sectors eligible for DAK funds, including four infrastructure sectors— irrigation, roads, sanitation, and water supply—which represent about 21 percent of total DAK allocations. 79

Institutional Analysis

The amount of the DAK allocation is determined on an annual basis within the national budget. The MoF allocates DAK to each sector and within those sectors local governments have the flexibility to determine which activities (sub-projects) they invest in. The MPW provides technical guidelines on the implementation, monitoring, and evaluation of DAK for each infrastructure sector. The MPW’s technical guidelines clearly specify the outputs on which the DAK grants can be spent, and these are mostly limited to maintenance, rehabilitation, and the upgrading or improvement of existing infrastructure.49 In April 2010, the Government created the Bantuan Operasional Kesehatan (BOK) grant to fund preventive health services in Puskesmas, such as maternal and child health, immunisation, nutrition, disease control, and environmental health. The goals of the BOK grant are to ensure that the minimum health service standards (SPM) are met at the district/city level in order to achieve the targets of the RPJMN 2010-2014, and to meet the MDG targets by 2015. Unlike DAU, the BOK grant comes with restrictions. The grant cannot be used for curative services, salaries, medicine, vaccines, or health technology. Instead, Puskesmas should use the money for materials for health education within the community, food for meetings, and transportation fees for health volunteers. The BOK grant acts as supplementary funding for preventative health activities at Puskesmas, but the operations of the health clinics and the health budget in general are still controlled by the local district/city legislature. The BOK grants are equivalent to a stimulus package designed to prioritise specific health activities that were previously neglected, while still ensuring the autonomy of districts in public health decision-making.50 However, it could be argued that BOK is incompatible with the principles of decentralisation. 3.10

Political commitment

The provisions in the law regarding contributions to BPJS specify that if there are sufficient funds to cover five years, contributions can be reduced. They further state that should contributions be inadequate, participants and employers will need to increase their contributions. The law does not provide a definition of ‘adequate’ funds, nor offer guidance on how this should be determined, or the process for increasing funds, including communication with the public and stakeholders. These factors will be critical to maintaining the public trust needed for the success of the UHC scheme. Article 56, Paragraph 3, of the Law on the BJPS No. 24/2011 allows for adjusting benefits and contributions as a last resort in order to maintain operations. An analysis51 of the amount of government contributions to cover costs of the poor was undertaken. Subsequently, the Government agreed that these contributions would range between Rp 22,000 and Rp 27,000 per person per month. However, following protracted inter-ministerial negotiations on the government contributions, a revised contribution level of Rp 15,500 per person per month was agreed upon (Sutriyanto, 2013; Wicaksano, 2013).

49

Ellis, P et al, (2011), Strengthening Fiscal Transfers in Indonesia: an Output-based Approach, World Bank. Ryan, M, (2011), The BOK Grant: a reaction to the failure of preventive health in decentralized Indonesia?, University of Michigan. 51 This was based on a study of contribution adequacy undertaken by the National Social Security Council (DJSN), University of Indonesia, together with other universities, the World Bank, a team from PT Askes and JPK Jamsostek and the National Team for Poverty Reduction Acceleration. 50

80

Institutional Analysis

However, Presidential Regulation on Health Insurance No. 12/2013 states that contributions from the informal workforce will not be covered by the Government, but instead individuals will need to pay their own contributions. The Government has acknowledged the difficulties involved in covering the informal sector in a contributions scheme, including unstable incomes of non-waged labourers that compromise their ability to make regular contributions. While the decision to exclude this group from government assistance creates fiscal space (and some certainty), the complexities involved in attempting to collect contributions from this sector, such as the likely inefficiencies and administrative costs, remain unresolved. The outcome of this situation is that the most economically vulnerable workers may be unable to pay regular contributions and be subject to user fees to gain access to services. This increases the risk that the near-poor will descend into poverty through exposure to CHE. For those workers currently covered by Jamkesda schemes, the Government plans to dismantle these schemes and integrate them into BPJS, possibly rendering this vulnerable group even more vulnerable. There needs to be a social marketing mechanism to both recruit citizens into Human Capital Information System (HCIS) and inform those eligible of their eligibility status and expected benefits. 3.10

Organisational development

3.10.1 Structural alignment An annex 1 summarises the flow of funds, the flow of authority/governance, and the flow of information/reporting between the various entities involved in the planning and delivery of health-care services. There is clearly misalignment between them. As a result, regulation, planning, budgeting, financial management and performance management cannot be treated as elements in a comprehensive system of management and control. In order to harmonise management processes, this requires that these systems and sub-systems are also harmonised. The role of the MoH vis-à-vis BPJS will represent a significant conceptual and practical issue. In particular, there are unanswered questions around the mechanisms that will exist: (i) to enable the MoH to exercise its policy and planning duties, (and to ensure implementation), (ii) for the MoH to encourage BPJS to reflect national policy and plans through its contracting and re-imbursement processes, (iii) to capture policies around equity and reaching poor, vulnerable and indigent communities, (iv) to ensure that the outcome of MoH (and local) licensing of facilities is reflected in BPJS contracting agreements, and (v) to align the quality assurance duties of the MoH and the quality requirements established by BPJS through its service agreements. When the Constitution stipulates the adoption of broad autonomy, the question is whose power is devolved to local governments? There are three branches of power, namely: legislative, executive and judicial. Since Indonesia is a unitary country, there is also unification of the law. This means that there is no judicative power in local governments. All justice powers in the locality are an extension of justice powers at the centre. Legislative power in the form of lawmaking privilege lies with the centre. Consequently, the power devolved broadly to local governments is executive power. Executive power rests with the President as stipulated in Article 4 of the Constitution. 81

Institutional Analysis

The President according to the Constitution is the head of government authority. Government authority was designed to reflect authority to deal with issues from “the cradle to the grave”—a life-cycle model. This government authority is then used to form cabinet ministries by the President. The power of each ministry is then devolved in the broadest sense down to local governments, but the final responsibility of all devolved power still rests with the President as a characteristic of being a unitary country. When the cost of treatment is borne by BPJS, Puskesmas will become health-care providers. There will be a form of capitation based on the size and characteristics of the catchment population served by a Puskesmas. The main difficulty for Puskesmas is that they cannot use the revenues associated with capitation directly. Currently, the budget is deposited with local government treasuries and regarded as general local government revenues, not pledged to Puskesmas. Puskesmas at best experience cash flow problems, and at worst underfunding, unless there is a mechanism devised to disburse and/or reimburse budgeted costs securely, quickly and accurately. The new financial instrument, sub-national public service units, or BLUD (Badan Layanan Umum Daerah), may represent a solution to mitigate challenges in financing Puskesmas. In future, it is expected that income for Puskesmas will be deposited directly via BPJS to Puskesmas, and not through provincial/district/city treasuries. This will require greater management competence and effort at the provider level. There is a risk that those districts/cities whose health allocations exceed the minimum expected by national policy will withdraw their “generosity”, because it may no longer be seen as a district/city issue. There are few mechanisms for the GoI to drive the reallocation of resources to create a more equitable distribution of health-care facilities and thereby access. There is no requirement for provinces and districts/cites to produce costed, explicit business plans linking health-needs analysis, investment, workforce and development intentions. The absence of such plans prevents the MoH from regulating and judging conformity to standards. It also adds to financial opacity. Whilst it is expected that 10 percent of district/city expenditure should be allocated to health care, how actual expenditure is coded and assigned to budgets creates difficulties in the interpretation of actual expenditure (e.g., general purpose vehicles treated as ambulances, road improvements attributed as investments to improve access to Puskesmas). There is no effective mechanism for the MoH/provinces to ensure that districts/cities disburse funds in line with the RPJMN A significant anomaly arises when a public-sector political/policy entity (the MoH, province or district/city) is also a provider of services under contract to BPJS. At the MoH level, this should be resolved by removing the management role of the MoH over Class A hospitals and delegating this role to an executive agency. However, in the case of provinces and districts/cities a similar action is less practical due to the balance of scale and strategic interest. Some countries have managed this issue by creating a “National Hospitals Authority”—an arm’s-length agency of government managing all hospital operations across a country (e.g., Ghana). However, in Indonesia the scale and complexity of the country, and the already complex governance arrangements created by decentralisation, make this an unattractive proposition. Instead, the most practical proposition is to create an internal firewall within PHOs 82

Institutional Analysis

and DHOs that will explicitly distinguish the supervisory, legislative and regulatory functions from the management functions (Class B, C and D hospitals and Puskesmas). 3.10.2 The changing role of the MoH As the roles of the MoH change and become more focused on policy-making, M&E, regulation, facilitation (pembinaan), governance, managing strategic workforce planning (seven basic medical specialists), and maybe managing Class A hospitals, if this continues, the MoH should have the capacity to regulate and oversee health-service delivery at a national level. There should be awareness that the ultimate responsibility and accountability for health matters lies with the central government through the MoH. In the context of a unified state, the relationship between the MoH and local government health agencies is hierarchical. Therefore, it remains the responsibility of the MoH to supervise and to facilitate (pembinaan) local government health agencies. However, it is anomalous that the MoH should have a direct management role of health care provision, whilst simultaneously being responsible for regulation and supervision. Therefore, after the introduction of social health insurance through BJPS in 2014, the MoH role should surrender some functions, whilst others will need to be enhanced. In terms of direct “health-giving” activity, the MoH will need to develop its capacity and its energy in public health functions, such as the following: • • • • •

Health surveillance; Health status surveillance; Water, sanitation and hygiene; Epidemic surveillance; Health promotion/behaviour change communication (road safety, healthy life-style promotion etc); and • The provision of public “merit” goods.

In terms of indirect health system oversight and development the MoH will need to focus on developing the following areas: • • • •

Regulatory systems; Governance standards; Planning and policy-making; Harmonising medical and technical education production and curriculum development with health system needs; • Nationwide licensing regulation (especially for health professionals); • Advocacy (with BPJS, the MoF, MoHA, Parliament, the MPW, PHOs, DHOs, Bappenas, etc.); and • Health Management Information System (HMIS)

In terms of system leadership, the MoH should assume responsibility for the development of the following: • • • •

Clinical protocols/care pathways; Health technology assessment; Emergency/ disaster/epidemic management; The development of clinical standards; 83

Institutional Analysis

• The drug regulatory framework (pharmaco-vigilance, DRA, product licensing, QC, GMP, GDP etc.); • (Possibly) management of the central purchasing of clinical materials and commodities; and • Supervision and co-ordination of development-partner relationships (e.g., GAVI, GFATM initiatives). The role of the MoH in the direct management of Class A hospitals will need to be considered. Either an external/independent regulatory body will need to be established if the MoH continues to be responsible for their management, or a management agency for Class A hospitals will need to be created if the MoH decides to retain its regulatory role. Similarly, it is anomalous that the same Directorate of Medical Services will be responsible for both supervising, leading and managing (some) services, and also responsible for quality improvement and accreditation. Within the MoH, a number of unclear relationships and roles exist, including the following: • • • • •



Different data-handling units; Incompatible roles as both regulator and manager of some services (Class A hospitals in particular); Unclear relationship between the MoH Bureau of Planning and the Directorate of Health and Nutrition of Bappenas; Unclear relationship between the value of block allocations by MoHA and MoH advocacy; How the MoH will exercise influence, at a policy and service-delivery level at both primary and curative service levels. For example, it is GoI policy to maximise the number of deliveries supervised by skilled attendants. How can this be implemented without the means to influence DHOs directly, given that direct intervention through recruiting and deploying of staff is incompatible with the principle of decentralisation?;and How the MoH will operate with BPJS, in the context that BPJS is a payment/reimbursement agency, not a strategic instrument (for the time being at least)? How will greater equity be secured, and how will the acceleration of inequity be reduced?

It is not clear that the MoH is structurally designed to be able to absorb the requirements of the new landscape. The appearance of BPJS, the planned introduction of UHC, and the impact of the new decentralisation legislation, will all present new challenges to the MoH. Furthermore, the likelihood of political turbulence may exercise the MoH in terms of policy leadership and priorities. The risk is that the MoH is not be sufficiently agile to surrender existing practices, processes and behaviours, and redirect its attention towards the new challenges, or to acquire the agility and capacity (both organisationally and in its competence) to be able to supervise systems rather than to manage them. In particular, developing a new focus on public heath practice and policy development will be challenging.

84

Institutional Analysis

One proposition is that a new policy development/advisory service is created, whose role would be to provide high-level analysis and advice at the ministerial level. However, this has been previously attempted, but proven to be unsustainable. The reason is not associated with the absence of skill or objectives, but more to do with the distortion of internal structural relationships and the notion that such a service/unit should have a relationship with the minister, which would compromise the roles of Bappenas and the MoH’s Bureau of Planning. In terms of scale, the structure of the newly designed MoH is no leaner than its predecessor. Consequently, the MoH will continue to experience similar problems of scale, namely sluggishness, long lines of communication, a heavy superstructure and challenging coordination. Meanwhile, anomalies will persist, for example between several data management centres, and several points of expertise for some services, such as pharmacy, technology, and finance/budget. 3.11.3 The developing role of BPJS Typically (but not everywhere) a health insurance fund will develop and become responsible for the following areas: • Collection of funds (often in conjunction with other agents, for example the MoF). • Pooling, stewardship, actuarial analysis and the management of funds. • Development and definition of a basic/essential benefits package, including assuring access for the poor and chronically ill, and other vulnerable groups. • Tariff-setting and contracting, with both public- and private-sector providers. • Managing reimbursement systems and reimbursement, including recurrent/marginal costs, and possibly capital if a capital-charging element of tariffs is included. • Quality monitoring and assurance of service delivery. • M&E of the service provision programmes. It is unclear (i) what mechanism will be invoked to ensure the pursuit of equity (as opposed to aggravating/securing inequity), (ii) how capital funds will be managed, (iii) what will be the instruments for providing strategic leadership, (iv) how contract relationships with private providers can be engineered in such a way as to offer best value in a mixed economy of health, (v) how referral patterns between primary, secondary and tertiary care can be managed to ensure appropriate and best-value treatment regimes, (vi) how drug costs can be managed, (vii) how clinical developments can be financed, (viii) how the mix of service provision can be encouraged to respond to the burden of disease, (ix) how the outcomes of quality assurance and M&E regimes can be applied, and (x) how the central reimbursement regime can be reconciled with sub-national privileges and responsibilities. Whilst detailed guidance on the unification of systems and databases for establishing BPJS has been published, less consideration has been given to potential issues relating to the different contribution and benefit packages of the different systems. Currently, the JPK Jamsostek contribution for formal sector workers is paid only by the employer. However, Law No. 40/2004 stipulates that contributions are to be “jointly borne” by employers and workers. The proposed changes to standardise the schemes imply that employees under JPK Jamsostek will be expected to make payments. There are also potential increases for employer contributions. The Government has set the expected contributions from employers and employees at 5 to 6 85

Institutional Analysis

percent of monthly salaries (2 to 3 percent for workers and 3 to 4 percent for employers), which represents an increase on current contributions. While government-owned health facilities will be included, BPJS will select the clinicians and private-health facilities to be contracted under the scheme, and it will pay health facilities in primary care prospectively through capitation and retrospectively in secondary care for service outputs (diagnosis-related groups, DRGs) using the Indonesian definitions (INA-CBG). Public trust in the system will be facilitated through auditing, monitoring and supervision by the supervisory board and the National Social Security Council (DJSN). The accommodation of complaints by participants, doctors, and health facilities will be conducted through various community channels. The criteria applied in the selection process are blunt instruments, based around price and quality judged mainly through input and resources rather than outcomes or processes. Table 11: Consolidation of funds into BBJS

Resource contributions

Government

Extent of pooling

Purchasing / provision

Existing funds to be pooled by Subsidy of INR 2014: - Jamkesmas 22,000- INR 27,000 per person per - TNI/Polri (military and police) Hospital month for those - Askes PNS (civil servants) categorised as poor.

DRG payments based on INACBG. Amounts to be negotiated with hospital associations and to vary according to region.

- JPK Jamsostek 5-6% of wages, 2019: total population, including shared between Formal sector remainder of Jamkesda schemes PHC employer and total: 257.5 million employee Self-funded BPJS as single institution contributions 5-6% managing pooled funds to be Benefit Informal sector of monthly income (+ formed by conversion of PT package some government Askes contribution)

Monthly capitation based on registered users for public and private clinics Comprehensive. Initially public ward for govt contributor and 2nd class ward for self-funded. Shift to 2nd class for all by 2019

The BPJS-MoH relationship will mature and develop with time, but the lack of clarity about what should and should not appear in the basic benefits package (through UHC), about whether tariffs will be based on expenditure or diagnosis, and how the pursuit of equity will be promoted, will need to be resolved. The organisational relationship between BPJS can be captured in a number of ways: at a strategic and governance level by ensuring that the MoH is well represented at Board-level within BPJS, that the MoH scrutinises BPJS business plans, and that the operational relationship is captured in a Service Level Agreement (SLA) specifying performance in terms of payment intervals, prudent financial management, budget conformity, positive discrimination to develop the weaker districts/services, and the management of “gaming behaviours” by providers. The relationship should also include a degree of influence by the MoH concerning which providers (public and private) could/should be admitted to an approved list of providers based on some judgement of quality (accreditation, for example). 86

Institutional Analysis

At the point of full implementation of UHC, the role of BPJS may develop, by default, into a more strategic role (for example, such as Monitor in the UK), with direct connection to neither the policy structures of government, nor with the communities it serves. Mechanisms to accommodate this anomaly will need to be devised. There needs to be development of standard procedures of the National Health Information Centre (NHIC) (known as BPJS Kesehatan) to be used starting on January 1, 2014, and to be deployed initially in the consolidation of the existing schemes. This is still in progress, but the risk is that to include all of the absorbed schemes will require that the minimal standard procedures be adopted as a lowest common denominator across the whole of the BPJS operation. Finally, BPJS needs to be accountable and responsive to the community it serves and which finances it either through political leadership and engagement, or through an operating relationship with central and local governments. The relationship between BPJS and sub-national governments is problematic. There seems to be no obvious mechanism to harmonise reimbursement and the wider health economy, or to deploy BPJS engagement in the development of local strategies to deliver community-wide health gain. This issue needs to be resolved. In addition, there seems to be no planned mechanism to mitigate the risk of the compression of fiscal space or an excess of funds in the BPJS risk pool. If the BPJS risk pool becomes decapitalised part way through a financial year, there seems to be no mechanism to correct the deficiency. Equally, if the mechanism raises more assets in the pool than can be usefully disbursed, there is no obvious mechanism to adjust premiums downwards. The former eventuality is far more likely. Given that the premium will be a fixed rate amount, not adjusted for risk, and which can be made compulsory, then it is tantamount to a tax. There is a well-known “insurance effect” in which subscribers, namely whose risk is not assessed, will seek to gain benefit from their expenditure and so will seek a return on their investment, especially those who perceive that they are low risk and are not getting value for money from their payments. The capacity to absorb growth in both demand and supply has not been risk-assessed, so it is not known whether systems, competence, willingness, or capability exists to absorb the impact of UHC within a decentralised framework. Specifically, it is not known whether: • • • • •

Financial systems exist to manage reimbursement (both billing and payment) once the existing five schemes have been consolidated; Capacity exists to manage DRG coding accurately and reliably; There is willingness to release health-related income (at district/city and provincial level) to health-related services; There is local willingness to increase health expenditure towards Abuja agreement52 levels; There is technical, managerial and governance competence to manage the central financing of local services; or

52

In April 2001, heads of state of African Union countries met and pledged to set a target of allocating at least 15% of their annual budget to improve the health sector. At the same time, they urged donor countries to "fulfil the yet to be met target of 0.7% of their GNP as official Development Assistance (ODA) to developing countries 87

Institutional Analysis



There is sufficient risk-taking capacity for the MoH and BPJS to withdraw from direct healthcare control and to focus on regulation, governance, advocacy and standard-setting.

A risk assessment should to be undertaken and contingency plans developed to confront the predictable conditions in which absorptive capacity (fiscal space, skills, information systems, inability to recruit Jamkesda, etc.) jeopardises the implementation of UHC.

Figure 9: Road map: implementation of UHC

53

It can be seen that there are many points at which, if assumptions are unsound or do not conform to expectations, the planned progress to full UHC could be jeopardised. There is a strong risk that 100 percent coverage may not be delivered by 2019. 3.11.4 The role of sub-national health offices The provincial level has the potential to develop as the key link in the chain of command, as it is both sufficiently local to ensure sensitivity to local needs and priorities, and of sufficient scale to interact meaningfully with central government. A formal constitutional role for provincial governors, who are the representatives of the centre, also gives them an opportunity to exercise central prerogative at the local level.

53

MoH (2012), The National Health Insurance of Indonesia (Ina Medicare) 2012-2019

88

Institutional Analysis

If the licensing role of local offices was standardised and strengthened, and supported by a robust quality assurance framework, then the licensing (and possibly accrediting role) of provinces could be deployed by BPJS in the appointment of providers in the UHC scheme. Furthermore, the relationship between PHOs and regional BPJS branches offers the potential for mutually supportive management of health economies. The role of the PHOs/DHOs ranges across direct management of the Class B hospitals, the Class C and D hospitals and Puskesmas services, quality assurance, management of APBD, DAU and DAK incomes, licensing of facilities, deployment of human resources and advocacy with the MoH/PHOs. After the social insurance regime is expanded beyond 2014, it is unclear how a fee-for-service model of reimbursement can be reconciled with the managerial role of PHOs/DHOs, unless the contractors are the PHOs/DHOs rather than the patient-facing service providers. If this is the model adopted, then the benefits would be more stable cash-flow management, but the disadvantages would include the risk of fungibility/re-allocation and the de-linking of the contract relationship between payment and quality. The risk is that an executive agency (BPJS) could subvert the role of the publicly-accountable instruments of local government. Puskesmas income is related to catchment population characteristics and not to activity, except marginally, when there is an incentive for upward referral, thereby protecting unconditional DHO income (through Puskesmas) and taking advantage of the conditional income earned through BJPS activity-based contracts. PHOs/DHOs have no economic incentive to suppress this possibility. Anecdotally, it is reported that there is limited capacity within PHOs and DHOs to exercise an assertive and leadership role in an environment characterised by a complex mix of centrallymanaged social health insurance, coupled with the underlying and non-negotiable features of a decentralised governance and government framework. Capacity here is taken to include competence in all three areas (technical, managerial and governance), the presence of systems (financial, M&E, MIS, etc), and the distribution of assets and resources. Once again, the disconnect between the principles of decentralisation as envisaged by Law No. 22/1999, the policy, planning, regulatory and oversight functions assigned to sub-national institutions, and the practical day-to-day management of service delivery is clear, and is challenging to reconcile, particularly in a low-capacity setting. Following the implementation of the revised decentralisation legislation, districts/cities will assign “business managers” to Puskesmas and work in partnership with villages. Their role will be to support the administrative functions (finance, HR, etc.) at the local level, and to bridge the relationship between villages and districts. The risk is that this resource may be very thinly spread across partners, and there is uncertainty about the availability of professional capacity. Regarding competence, in theory the Competency Test Development Agency (LPUK) has a role to play in defining and assessing competence. In practice there is no binding regulation to require local governments to reflect competence in the appointment of heads of Dinas including health Dinas. Consequently, the appointment of a health Dinas head is at the discretion of the governor, or bupati/mayor. Anecdotally, it is reported that there are many examples where the head of the health Dinas is appointed without sufficient knowledge on health matters (technical, managerial or governance). In order to avoid such practices, in the revision of Law No. 32/2004 89

Institutional Analysis

there will be a requirement for officers appointed as heads of Dinas to be able to exhibit the three competences: technical, managerial and governance. In the case of the health Dinas, technical competence will be set by the MoH; managerial competence by the KemPan (Ministry for Administrative and Bureaucratic Reform) and; governance competence by MoHA. In the draft Regional Government Law technical competence will be defined by the technical ministries, governance competence by MoHA, and managerial competence by the Ministry for Administrative and Bureaucratic Reform. However, no clear competence specification defined for the MoH, provincial or district/city-level, either in terms of technical, governance or leadership roles, has been defined. Therefore, no reliable or consistent judgements of capacity can be made. However, anecdotally, the capacity to absorb the challenges created by UHC within a decentralised framework at the sub-national level is considered to be low. 3.12

Service quality and standards

The previous Sub-section 0 refers to the development of standards (both service through MSS/SPM, and through operating procedures of the NSPK. A well-established phenomenon is that MSS become targets and so therefore become maximum aspirations. A number of studies have shown that MSS applied to care settings have a discernible effect only on those providers whose service quality falls well short of the minimum standards set, but have little overall effect. 54 , 55 Furthermore, there is some evidence that providers whose services exceed the MSS may well reduce investment to match minimum standards, so reducing overall care quality. 56 It seems that “value-added standards” have a greater effect on overall system performance.57,58In such a framework, standards are expressed as rates of improvement. In order to improve service standards and quality, the MoH should modify some of its MSS/SPM standards to encourage continuous improvement beyond a threshold of minimum standards, rather than simply meeting a minimum standard and then accepting no further improvement. 3.13

Mitigation of new and emerging challenges

Throughout this paper, reference has been made to new and emerging challenges, and possible responses to them. This section focuses on the priority areas in terms of the institutional impact of addressing some of these issues. 3.13.1 Mitigation of new and emerging challenges for the Government Central government should promote the quality of health workforce education across both public and private sectors, through the rigorous regulation of the learning experience (including clinical placements). This should be in collaboration with the MoE and the MoH, and actively manage the health labour economy and the recruitment and deployment of specific categories 54

Zhang X, Grabowski DC, (2004), Nursing home staffing and quality under the nursing home reform act, Gerontologist. Feb; 44(1):13-23. 55 Park, J and Sally C Stearns, S., (2009). Effects of State Minimum Staffing Standards on Nursing Home Staffing and Quality of Care, Health Serv Res. Feb 2009; 44(1): 56–78. 56 WHO (2002), A Rapid Review Of Health System Organisation And Performance, With A Focus On Improving Health Outcomes For The Poor, An Aide Memoire, based on the WHO Health System Performance Assessment Framework. 57 Friedman NL, Kokia E, Shemer J (2003), Health Value Added (HVA): linking strategy, performance, and measurement in healthcare organizations, Isr Med Assoc J. 58 Manley MJ. (1996), The promise of integrated health care; simple strategies for transforming organizations, Med Group Manage J.

90

Institutional Analysis

of the health workforce that is not possible at the local government level (e.g., clinical specialists and pharmacists). Health workforce categories and establishments should be determined based on the type of services that Puskesmas deliver, not on universal norms. For areas with specific characteristics, such as isolated islands, remote and border areas, the skill mix of the workforce should be adjusted to reflect specific local needs. The pattern of minimal workforce in Puskesmas can be adjusted based on the type of Puskesmas, such as Puskesmas with beds (in-patient Puskesmas) or Puskesmas in urban or rural areas, instead of requiring conformity to a national, generic model. 3.13.2 Mitigation of new and emerging challenges for the MoH A number of areas will exercise the MoH over the coming years and these have already been considered in this report, including structural, relationship and contractual issues. Concentrating on the MoH role as policy-maker and technical leader, its role as the champion of primary care provides an opportunity to assert its policy and technical leadership, and to exercise its commitments as a signatory to a number of global and regional health initiatives. Redefining Puskesmas Based on Ministerial Health Decree No. 128/2004, Puskesmas are the a Technical Implementation Units (UPT) of the DHOs, and responsible for organising health actions in specific work areas (sub-districts). The functions include roles as a centre of public health policy development, a centre of community empowerment, and a centre of health services delivery. In reality, the functions are skewed towards curative rather than public health activity. Hence, the existing regulatory frameworks59 need to be applied more rigorously to assure the shifting of services away from curative care. However, Puskesmas also need to be strengthened to encourage preventive and promotive activities, i.e., towards encouraging behavioural change through all community-based health actions, such as Posyandu, Desa Siaga, Pos Obat Desa, Pos Malaria, etc. The contract with BPJS offers the mechanism to implement this objective. The health activities required of Puskesmas should be grouped into mandatory actions such as health promotion, maternal and child health, communicable disease and NCD prevention, nutrition, environmental health, emergency and essential health services, and discretionary/desirable health actions accounting for locally-specific health problems in the area (and accounting for the availability of resources). Again, the contract with BPJS offers the mechanism to implement this objective. Heads of Puskesmas Based on previous regulations, the position of head of a Puskesmas is Echelon IV/a, parallel to a head of a sub-district (camat) echelon. After decentralisation, based on Government Decree No. 41/2007, it remains Echelon IV, similar to echelon of head of village, while the camat head

59

Ministerial Decree No. 118/2004, and the strategic plan 1994 are designed to strengthen and support promotive and preventive care.

91

Institutional Analysis

became Echelon III. This situation has triggered anomalies in coordination and management of technical and operational matters. To resolve this issue the status of head of Puskesmas should be recalibrated not as a structural position, but as a functional position. This means that the role will not be confined to those qualified as a particular echelon or a civil servant position. Instead, a PTT doctor60 will be allowed to be the head of a Puskesmas. Many Puskesmas have no medical doctors. Whilst it is not universally accepted that medical doctors are always necessary in all primary care settings, it is nonetheless policy in Indonesia that Puskesmas requires the presence of a medically qualified person(s) as the leader of the health team. In order to run well, the role of the Puskesmas head should be able to be held by a public health worker with a background in medical education or a bachelor of public health, a bachelor of nursing, dentist or pharmacist. This requires that a MoHA decree stipulate that in the pursuit of operational efficiency and effectiveness a Puskesmas head can be held by a functional health worker, rather than uniquely a bureaucrat. Skill mix It should be a local government responsibility to calculate the quantity and skill mix of the health workforce required to equip Puskesmas with the human assets to deliver their duties, other than for highly specialised, scarce skills that need to be more regionally or nationally managed. 3.13.3 Mitigation of new and emerging challenges for BPJS There are five areas that are emerging as significant challenges for BPJS. Each is discussed below: Recruitment: BPJS expects to be able to absorb up to 300+ existing schemes and to recruit many millions of people from the informal sector of the economy by 2019. These seem very ambitious expectations. It may be that more sophisticated targets for BPJS are required, combining the consolidation and absorption of existing schemes, and the recruitment of workers in the informal sector. The availability of primary-care providers: Assuming a planned coverage rate of one health centre per 50,000 citizens, Indonesia will need around 5,000 Puskesmas. The public sector provides over 9,000 but a very high proportion of these fail to meet standards in terms of staffing structure, the basic conditions of the premises, and the equity of distribution. The solution—and one that avoids large-scale, short-term capital investment from the public purse—would be to encourage the independent sector to contract with BPJS. Contracting: If contracts are uniquely fee-for-service contracts, using INA-CBG, then the administrative effort (for both BPJS and providers) involved in reimbursing every individual episode of care and dealing with every contestable submission would be overwhelming. However, if every contract was a block contract, there would be limited opportunity for BPJS to reimburse for service developments, and to assure quality and conformity to contract conditions.

60

Pegawai Tidak Tetap (PTT), or contract physician scheme 92

Institutional Analysis

BPJS needs to engineer a contract portfolio that is both manageable and that enables the proper supervision of the performance of many thousands of providers through a blend of contract types (fee-for-service, block, and cost & volume) reflecting local specificities, and to have close relationships with local institutions (e.g., DHOs) to whom contract conformity monitoring and reporting can be delegated. Quality assurance: The contract mechanism offers the opportunity to boost quality standards beyond the minimum service standards that currently prevail and that are captured in the “credentialling” criteria used in the process of admission to the approved providers list. Contracts should include a series of quality standards associated with access, clinical quality, efficiency, reporting requirements and coding accuracy. The absorption and consolidation of existing schemes: Existing schemes have different benefit packages associated with them. For example, the PT Askes scheme has a benefit representing one month of care for chronic conditions, whilst the successor scheme offers one week of care. For such subscribers, the options include continuing to pay the same for fewer benefits (creating dissatisfaction and reputational damage reducing the probability of being able to absorb other schemes), or reducing the subscription for a reduced range of benefits (and so reducing planned income to the risk pool), or consolidating benefits upwards, increasing the cost commitment associated with pay-out for the less generous schemes and/or increasing subscription costs for subscribers. The least unacceptable solution would be to consolidate schemes upwards, and for the Government to use a proportion of the planned increase in health-care expenditure to subsidise the risk pool until such time as they are naturally equalised. 3.13.4 Mitigation of new and emerging challenges for provinces The effective implementation of Puskesmas programmes needs good management. Guidelines on technical assistance and supervision from DHOs/PHOs/the MoH are needed, supported by a good information system to ensure the quality of Puskesmas services. Standard operating procedures/protocols (SPOs) for required primary care activities should be developed. Referral systems should be strengthened, and driven by a protocol/pathway determined by the MoH exercising its leadership and guidance role. This is not only due to the limited resources at Puskesmas, but also to ensure that patients are assured the highest quality of care at the lowest cost and with greatest access appropriate to their needs. A mechanism for managing referrals should be developed and sustained. The revision and consolidation of the Regional Government Law will reinforce the hierarchical relationships from the centre down to villages. The BPJS reimbursement regime will reduce the capacity of provincial structures to intervene in order to remedy inequity. A uniform system within a unified country using a uniform legislative framework, but applied to 34 provinces with immensely different characteristics (socially, economically, geographically, epidemiologically, politically) commends the principles of decentralisation, which enable and encourage local institutions to operate in a locally-specific way. However, this also confounds the mechanisms supporting central oversight, and the ultimate responsibility for services held by the President. This dilemma has been referred to throughout this report, and the impact of the BPJS mechanism further complicates already complicated relationships. This represents both a conceptual and practical difficulty, and can be summarised as a quandary of whether the 93

Institutional Analysis

principle of decentralisation has greater value than the principle of efficiency. Districts/cities are in a similar predicament. The obvious solution has already been proposed, namely that the MoH exercises its supervisory and guidance roles by focusing on desired outcomes—conformity to standards, etc.—through receiving business plans created at the local level, which then need to be negotiated and agreed upon at the provincial level. The outcome of the agreement is then captured in a service level agreement (SLA), which identifies costs, quality, access and reporting requirements. Community involvement in the preparation of the business plan would be a requirement. The business plan would also include provincial concerns. This solution protects the interests of the principles of decentralisation, because it represents an agreement made between the province and the centre, not accession to a greater authority. It involves the community because it is transparent. It would be an evidence-informed plan and the MoH would have the means to exercise its responsibilities through the SLA. There is no correlation between the design of PHOs and DHOs and local necessities. A small local government and a substantial local government (in terms of the size of population served, or its geographical characteristics, or its economic condition) will have the same structure of PHO and DHO. Therefore, PHOs and DHOs need to be restructured by giving consideration to the size of the population that should be served. There will be Class A, B and C structures of PHOs and DHOs depending on the size of the population served. The classes represent capacity of the health offices to exercise its duties in the context of local needs. No matter how small the structure, it should be able to manage its responsibilities entirely. There is a non-linear relationship between the population served and the scale of a PHO/DHO. Some functions are population-related, whilst others are not. The role of the MoH is to assign the health functions required of all sizes of PHO and DHO. In order to maintain the line of coordination and monitoring between the structure in the MoH with its counterpart in PHOs and DHOs, a common nomenclature/taxonomy should be devised to enable better and more accurate communication between divisions or departments in the MoH with their local counterparts. This would enable the recording and reporting of activities to be synchronised, as opposed to synchronising structures. There is no good reason why every PHO and DHO should be a microcosm of the MoH. However, bearing in mind the distinctively different nature of the provinces, they should be free to design their own organisational form so long as the agreements reached with the MoH about outputs and outcomes are observed. The revision of the decentralisation framework and associated legislation will see the introduction of new mechanisms of support at the provincial level. This support is intended to help provincial governors exercise their responsibilities both as central representatives and as supervisors of district/city-level performance and activities, in addition to their role as publicly accountable elected representatives. This new mechanism has the potential to significantly improve the capacity of PHOs in terms of leadership and professional management, providing that the mechanism attracts high quality staff, can establish an assertive presence and legitimacy, and that provincial governors are equipped with a clear mandate and have sufficient resources at their disposal including research and information capacity. 94

Institutional Analysis

3.13.5 Mitigation of new and emerging challenges for districts/cities In accordance with the principles of decentralisation, the cost of providing adequate services at the Puskesmas is the responsibility of district/city governments based on minimum service standards (SPM). The concept of SPM is based on the minimum standard of health services that should be provided by local governments (as opposed to provincial or central) to their communities. SPM are based on standards agreed by the MoH with the PHOs and DHOs by considering their financial capacity and also the availability of medical personnel in local governments throughout the country. A study61 on district health accounts showed that the allocation of budget was skewed towards curative rather to public health activities. Moreover, based on current regulations, the financial management of Puskesmas refers to MoHA Decree No. 59/2007, in which all of the budget available for Puskesmas should be allocated through the district/city treasuries, giving no autonomy to Puskesmas to manage their own budgets. Thus, the capitation fee from national health insurance (JKN) would go to the Puskesmas. For non-personal services, if local governments do not allocate their budgets on the expectation that any shortfall in health expenditure will be borne by the MoH, then the MoH could ask MoHA not to approve the budget of the province. In turn, MoHA could request the governor—as central representative—not to approve the district/city budget. These powers already exist and could be deployed, but in practice rarely are. Deployment of this mechanism would encourage district/city governments to apply their resources to achieve the SPM as a right of citizens. Future funding for personal services will be through BPJS and will go directly to Puskesmas, including a capitated element. This means that the role of districts/cities will be focused on support, technical leadership and co-ordination and for the provision of public merit services (e.g., mosquito spraying, fogging for dengue fever, etc.). The role of the new “business managers” will be crucial in this new scenario. As a consequence, there is a need to reform the regulations governing financial management and governance to provide more flexibility to manage budgets to meet the needs of Puskesmas. Health financing needs to be more predictable and needs to be defined as the responsibility and authority of both government and society. Although health services are a ”right” of communities, the community should be encouraged to participate in the financing of health (both personal and public health). Whilst co-payment is an established principle for personal services, it is less so for public merit services. A mechanism to secure shared responsibility for the health status of communities should ensure that the planning and budgeting is governed by technocratic, district/city governments and communities through community empowerment and using single-member districts (SMD) and multi-member districts (MMD)62 as the instruments of decision-making. This will require substantial skills and capacity development at the district/city level to support the management and governance of primary care, both directly through business managers, and through district/city-wide activities (health promotion, community engagement, etc.).

61

Ascobat op.cit A multi-member electoral district (MMD) is an electoral district electing more than one representative to office. A single member electoral district (SMD) is an electoral district electing only one representative to office.

62

95

Institutional Analysis

The relationship between districts/cities and provinces should be captured and formalised through a business plan/SLA arrangement. This has the advantage of creating opportunities for community empowerment mechanisms to operate at a service planning and delivery level. Local governments should include the achievement of SPM in their five-year development plans and be implemented under the annual plan and budget. The MoH should monitor and supervise the achievement of the SPMs. The outcome of the monitoring may conform to the annual plan and budget, or it may be that SPMs are not achieved for a variety of reasons, for example: low fiscal capacity (the MoH could help local governments by disbursing a specific grant); a deficiency of medical personnel (the MoH could request approval for additional medical personnel from the Ministry of Administrative and Bureaucratic Reform); or skill shortages (the MoH could conduct training for medical personnel in local government). 3.13.6 Mitigation of new and emerging challenges for the private/non-government sector The private sector (including the entire non-government sector, including faith-based organisations, non-profits, etc.), including health providers and health educators, is a major player in the Indonesian health economy and cannot be neglected in any analysis of the healthcare landscape. The main issue is how the private sector can be engaged in the planning and governance of the health sector. The engagement of private-health providers (ranging from large-scale hospitals to single-handed practitioners) requires that both payers/commissioners (mainly BPJS) facilitate partnership opportunities for private-sector providers. Several obstacles seem to be dampening the enthusiasm for engagement, including the impact of the tariffs on the market place, the validity/reliability of INA-CBG as a contract currency, the disclosure of commercially-sensitive information, and the competitive commercial environment related to the treatment of capital and tax liability for non-profits (capital is free for the public sector). These are all issues that should be addressed by BPJS, the MoH and the Government, and will require close attention over the next five years. On the other side of the equation, the non-government sector must also be prepared to expose itself to conformity to quality standards (including clinical quality), to sharing activity and reach data, and to sharing clinical information at an aggregate level, if it wishes to benefit from publically subsidised income streams (that is, stable and predictable revenue streams). In terms of private-sector educators (see Section 0), a similar proposition exists—namely, that in return for exposure to supervision and regulation of quality conformity, and the regulation of supply—the private sector will be given access to public clinical environments for clinical placements. 3.13.7 Mitigation of new and emerging challenges for partners Partners include MoHA, the MPW, BKKBN, and the MoE. This report makes numerous references to how relationships leading to co-ordinated action might be captured given the dispersed structure of government (both laterally and vertically) and the complexity of financial flows (the MoH accounting for 60 percent of health expenditure). Joint action is a pre-requisite for delivering health gain (e.g., infrastructure development, water/sanitation, nutrition, supervision of provincial and district/city activity), but also a pre-requisite for achieving a 96

Institutional Analysis

balanced health-care sector (e.g., balancing supply and demand for well-trained health professionals). The role of co-ordinating ministers is critical in this process. Given that coordinating ministers have access to very limited resources to offer incentives for adjusting priorities and programmes, their role becomes more critical as mediators in managing the negotiation of agreements between ministries. 3.13.8 Mitigation of new and emerging challenges at the system level Managing these issues whilst still preserving the principles underlying decentralisation create dilemmas at both conceptual and practical levels. Mechanisms to resolve them may require legislative intervention and/or intervention through the Medium-Term Development Plan (RPJMN). Seven mechanisms can be suggested, as follows: 1. 2.

3.

4. 5.

6.

7.

Aligning reporting/information flows with accountability and financing flows. PHOs and DHOs to produce locally-specific costed business plans. These will reflect both RPJMN priorities and local specificities, so that funding can be aligned with health gain objectives, and in which allocations, releases, and budgets are transparent (as would connections/disconnections between them). This will require support in terms of developing capacity/skills in business planning. Engineer contract or service-based agreements between district/city governments and provinces, and between provinces and the MoH. These agreements would capture the intentions set out in the business plans, would require that the agreements are two-way (funds one way; performance the other), and that they would capture agreements around quality, access, equity and efficiency. Alter BPJS’s share of allocations. This would then include semi-variable cost elements in the tariff values, as well as the planned variable costs. Extend the concept of matched funding to create incentives for districts/cities to conform to central priorities. This would also have the effect of providing an instrument for the MoH to encourage greater equity. Re-visit the structure and roles of units within the MoH, provinces and districts/cities to ensure that no conflicts of interest exist. An alternative strategy would be to consider creating a regulatory authority that would operate vertically and horizontally throughout the system. This would require that the authority had the instruments and leverage to actively manage the outcome of its regulatory role, and may require legislative intervention to create an entity empowered to intervene. Whilst in theory a regulatory body exists (Badan Pengawas Rumah Sakit, or BPRS) and should be operated at the provincial level, it is not compulsory. At the national level only a coordinating mechanism exists. However, BPRS has little power of enforcement and operates more as an arbitration instrument. The DHOs and PHOs have the power of hospital licensing and recommendation for admission to the list of approved providers, but there are structural role conflicts and anomalies in these structures and so enforcement is weak. All demand-side institutions need to develop systems, processes and competences to recognise and influence the impact of market weaknesses inherent in the health-care context. These institutions include the MoH, BPHS, and the PHOs/DHOs. 97

Institutional Analysis

4. Targets to be achieved by 2019 Targets are organised into four areas, all of which are associated with institutional form, as follows: • • • •

Governance and system-level Input-level Output-level Outcome-level

Governance and system-level: The governance and system-level targets focus on securing the major policy objectives of the MoH: namely confronting NCDs and infectious diseases, achieving equity, and providing accessible services to all citizens. The targets require that the GoI has in place policies and plans to provide clear guidance on the achievement of these objectives, and a focus on the creation or updating of plans and policies. These targets also require the creation, updating, or clarification of relationships, information sources and guidance. Input-level: The input-level targets focus on the resourcing, capacity, and structure of the SKN. These targets are concerned with human resources, the stability of the risk pool, and the existence of guidelines. These are all targets that reflect stability and institutional responses to prevailing conditions. Output-level: The output-level targets reflect issues concerning access/coverage, efficiency and resource quality and distribution. They assume that outcomes are at least partly a product of outputs of the system, and that they are in turn partly a product of input guidelines. Outcome-level: If the logic chain above is reliable, then the outcome-level targets can be presented as targets that secure access to high-quality services by all citizens. If these targets are achieved, then it will be possible to confidently assert that the mechanisms implemented are producing the desired results. The main targets are: • • • • • • • • •

Full implementation of UHC. Growth in public health expenditure to 4 percent of GNP. Growth in primary care and public health expenditure to levels typical of the Asia Pacific Region. Extension of BLUD to Puskesmas. Investment in Puskesmas to reduce the proportion of 38 percent that are either moderately or severely damaged. Deploy physicians in 2,000 Puskesmas that do not have any physicians in place, and create Puskesmas in the 430 sub-districts that as yet have none. Strive to ensure that 50 percent of primary care can be delivered by private-sector providers. All PHOs and DHOs to prepare business plans. Service-level/performance agreements in place between the main actors in the healthcare landscape.

98

Institutional Analysis

4.1

Main issues

Major policy questions: The major questions concerning policy can be summarised as follows: • How should health care be paid for? • How to attract professions committed to new models of provision (UHC)? • How to manage government costs in an age of commitment to funding secondary care reimbursement (BPJS, Pharma costs, DRG-based tariffs)? • How to secure quality services in a decentralised framework? • How to manage the relationship between the MoH and BPJS and other bi-lateral partners? • How to enable the policy role of the MoH in a decentralised context? • How to clarify the executive roles retained by the MoH? • How to reflect the needs of those who are currently being underserved in the tariffs for care in order to increase equity? • How to regain some central influence in the context of a unitary country, especially over policy implementation, quality, distribution and equity? Critical policy areas: At a systems level, critical policy areas will need to be developed include the following: • • • • • •

Distribution of health functions and NSPK. Minimum service standards for health. Institutional and inter-sectoral arrangements at the central and sub-national levels. Relationship of the centre with sub-national governments in relation to health. Improving the relationships between the MoH and BKKBN, BPOM, BPJS and K/L in health. Developing a guidance and oversight mechanism over sub-national health agencies.

Strategic issues: In order to address these policy areas, strategies focused on the following issues will be required: • • • • •

• • • •

Reconciling UHC with local, devolved privileges Aligning the flows of money, accountability and information. Referral systems to secondary and tertiary hospitals. Defining and supporting the development of managerial/leadership competence at local level. Promoting Health Systems Strengthening by building on Primary Health Care (PHC). However, confining reform to the Puskesmas level will not have substantial impact on the improvement of the whole health system unless accompanied by reform of DHOs, PHOs and the MoH that focuses on the WHO Building Blocks of Health Systems, where the objectives of whole system reform are to increase the equity, quality and efficiency of the SKN. Capacity-building of district/city governments. Development and implementation of the SPMs. Effective Binwas (oversight mechanism). Mechanism to develop community participation. 99

Institutional Analysis

• Develop an information system that can distinguish between district/city and Puskesmas-level activities. • Develop or revise regulations captured within the legislative programme specifically on the following: Fund channelling (DAK for the region) Performance management (reward and sanction) at the district/city government level. Strengthening institutional capacity and HRH. Improving the quality of hospitals and Puskesmas towards the achievement of UHC. Implementing Presidential Decree No. 72/2012 concerning SKN. Guidance for Health Systems Strengthening based on PHC and Puskesmas Revitalisation. Strategies for reform: Specifically, strategies to be implemented to secure the reform objectives should be developed as follows: • Strengthening the DHOs, PHOs and the MoH. • Strengthening regional general hospitals (RSUD) at the district/city level, and central general hospitals (RSUP) at the provincial level. • Growth and development of community participation, as well as private roles. • Strengthening cross-cultural collaboration. • Meeting the needs of a competent, committed and equitable health workforce. • Health financing reform. • Provision of other elements of health system building blocks. • Bureaucratic reforms at every level. Policies to improve the decentralisation framework: In the development and refinement of the decentralisation framework, a number of policies should be adopted by the MoH: • The clarification and redistribution of responsibilities. When responsibilities are clearly set out in the revised Law No. 32/2004 on Regional Government, the duty of the MoH will be to issue guidance (NSPK) to be used by the PHOs/DHOs in implementing their responsibilities. • The development and issuance of technical guidance for governors to implement their function as central representatives to supervise and facilitate the DHOs. • The development, clarification and issuance of MoH regulations setting out the referral procedures between health facilities managed by all tiers of government, reflecting the mitigation of tendencies to refer cases upwards in the system.

100

Institutional Analysis

4.2

Strategies

4.2.1

Recurrent themes

A recurring theme of this analysis is the obstacles to implementing high-level policy in practice, in the field, through the absence or weakness of systems, data and capacity. The policy framework is clear, but the collateral impact of larger structural characteristics affects the mechanisms available for the SKN to implement policies. This refers particularly to decentralisation and its effect on the policy objectives of achieving equity. Furthermore, the relative status of the various institutions involved in the protection of health and the delivery of health care, limits the ability of the lead political institution—the Ministry of Health—to directly determine how providers are reimbursed, accredited, supervised and governed. Illustrations of these obstacles include the following, and are paraphrased from a variety of sources63: •

“In the case of the MoH, first because it has very few levers of influence other than regulatory instruments, and financial sanction, and second because the decentralised structure militates against direct intervention in operational affairs.”



“BPJS is tasked with delivering UHC by 2019. This seems an ambitious target, based on the scale of work required to register an additional 100 million people, of whom the majority operate in the informal sector.”



“There are repeated references to a lack of clarity, and both overlap and conflict of roles between tiers both at a governance level and at a service delivery level.”



“Puskesmas are expected to achieve 100 percent accreditation by 2019, and secure recognition of quality-assured service provision. The only sanction for non-compliance to quality standards would be withdrawal of services from a Puskesmas. If this sanction were exercised, many communities would be left without access to care of any kind. Since the accrediting agency is also the body responsible for Puskesmas, it is unlikely that this sanction would be implemented in reality.”

63

Policy Brief Series, AIPHSS; Trisantoro, L (2009), Indonesian Health Reform in a decentralized system; World Bank (2013), Indonesia’s Health Sector Review; Simmonds, A et al (2013), Institutional analysis of Indonesia’s Universal Health Coverage policy 101

Institutional Analysis

5. Policy options and strategies 5.1

Recurrent strategic remedies

Many suggestions and recommendations have appeared throughout this paper, including issues around leadership development, procurement practices, information management and institutional assessment. However, two themes appear frequently. The first is how the MoH exercises a leadership role in the context of a decentralised constitutional structure, and how bilateral relationships can be managed in a transparent way (with the MoE, BPJS, the MPW, etc.). The second is how governance is conveyed down through a hierarchical structure from the centre to the village-level, whilst respecting the legitimacy of provincial and district/city parliaments. The remedy suggested for the first issue is through a network of contract and service-level agreements (contracts with providers, SLAs with other government entities), on the grounds that government cannot contract with itself. These would be implemented via a joint ministerial decree, and negotiated through co-ordinating ministers. The benefits of this would be as follows: • •



Such contracts would help to align the three misaligned systems (information, money and authority flows). Such agreements would be consistent with the principles of decentralisation, a unified constitutional form and with the structure of the government executive, since contracts/SLAs are the products of negotiated agreements. They are not the blunt instrument of the exercise of authority, which would compromise the role of sub-national representative Governments. Such agreements would be politically acceptable at the sub-national and central level.

The remedy suggested for the second issue is for the requirement of autonomous/semiautonomous government entities (hospitals, DHOs, PHOs) to prepare business plans every three years and annual operational plans for submission up to the next hierarchical tier (Figure 5.1). The benefits would be as follows: (i) the hierarchical principle would be retained; (ii) it would create an opportunity to promote quality and equity; and (iii) it would promote continuing dialogue with a manageable span of control. This principle is illustrated below. Figure 10: Contracted relationships

Service level agreements

Contracts

Business plans

102

5.2

Policy options, strategy and indicators

Table 5.1 summarises the main policy options and strategies called for, together with the key indicators associated with their achievement/ implementation. The table is organised by main policy area. Table 12: Policy options and strategies

Institutional arrangements under decentralisation

POLICY OPTION

STRATEGY

INDICATOR/TARGET

Formulate regulations on the clarity of distribution of health functions and authority between levels of government as referred to in the Regional Autonomy Law

Revise regulations on the distribution of functions and authority

Revision of regulation on distribution of function and authority to sub-nationals (completed in 2014)

Identify the overall NSPK; update, revise those that are obsolete, and develop those that are not established Socialise NSPK and facilitate to the subnationals Formulate a policy that expressly regulates relationships between DHOs and hospitals or Puskesmas Strengthen policy on relationship between the MoH with PHOs/DHOs Develop a policy on the level of government administration that should be

NSPK inventory: update, revise those that are obsolete, and develop those that are not established Disseminate implementation

information

and

M&E

of

Revise PP No. 41/2007 to set out clear and unequivocal relationship between DHOs with district/city hospitals or Puskesmas Develop a Permenkes referring to revised PP No. 41/2007 that regulates relationship between the MoH and PHOs/DHOs for clarity of coordination facilitation.

the the the and

Develop Permenkes that regulates the level

Inventory of NSPK conducted by 2015

will

have

been

Establishment of NSPK by 2015-2017 Socialisation (2015) and M&E conducted every year Publication of the revised PP as a replacement of PP No. 41/2007 which sets out clear and unequivocal relationship between DHOs and district/city hospitals or Puskesmas, by 2015 Permenkes on the relationship between the MoH and DHOs will have been developed, referring to the revised PP No. 41/2007, by

Institutional Analysis

responsible for public hospitals Develop a policy on financial autonomy for province/district hospitals and Puskesmas Require autonomous/ semi-autonomous government entities (hospitals, DHOs, PHOs) to prepare 3-yearly business plans and annual operational plans for submission up to the next hierarchical tier Adjustments of the relationship SOP between the MoH and DHOs in the context of the unitary country. 10.

of government responsible for the appropriate hospital based on class Develop a joint ministerial decree between MoHA and the MoH on financial autonomy for province/district hospitals and Puskesmas Develop a Permenkes to require autonomous/semi-autonomous government entities (hospitals, DHOs, PHOs) to prepare 3yearly business plans and annual operational plans for submission up to the next hierarchical tier

2015 Permenkes regulating the level of government that is responsible for classes of hospital will have been developed, by 2015 A joint ministerial decree between MoHA and the MoH on financial autonomy for province/ district hospitals and Puskesmas will have been developed by 2015 Permenkes on SOP of relationship between the MoH and DHOs will have been developed by 2015:

Use Binwas to support business planning: A programme evaluation that is in line with 10. A road map for Health Systems MoH function, to avoid the duplication of duties10. To formulate Permenkes on SOP of Strengthening based on PHC will have been that belong to the local health authorities relationship between the MoH and DHOs developed by 2016

11.

Develop a Permenkes that regulates health11. To formulate the operationalisation of11. An assessment of MoH programmes that information systems throughout the country Health Systems Strengthening based on PHC overlap with DHO functions will have been conducted (annually) 12. Harmonise MoH programmes with DHOs,12. To conduct an assessment of MoH especially in achieving MSS programmes, to identify those that belong to12. A Permenkes on revision of a Health sub-national function Information System will have been developed 13. Strengthening coordination between the by 2015 MoH and MoHA in imposing sanctions when13. To formulate a Permenkes on a health local authorities fail to comply with national information system 13. DHO Strategic Planning will have been health policy developed referring to the MoH’s Strategic 14. To harmonise MoH programmes with DHO Planning programmes, especially in achieving the MSS targets 14. A joint ministerial decree between MoHA 104

Institutional Analysis

15.

To formulate a joint ministerial decree between MoHA and MoH for imposing sanctions on sub-national governments that fail to comply with national health policy

and the MoH for imposing sanctions on subnational governments that fail to comply with national health policy will have been developed by 2015

16.

Development of road map of JKN 2012-2019 and introduction of UHC POLICY OPTION

STRATEGY

INDICATOR/TARGET

Risk assessment of system shocks, including; a. Inability to induct 100% of population by the end of 2019.

the

b. Financial over-commitment c. Financial underspend Create resilience in the road map of JKN.

d. Inability to engage the private sector e. Provider failure

Risk assessments completed by end-2015 Quality strategy completed, reflecting valueadded improvements by end-2014. Contingency plans completed by mid-2016.

Develop a quality assurance strategy and contract model to ensure that service providers deliver to minimum standards. Develop contingency plans to mitigate the impact of system shocks.

105

Institutional Analysis

Growth of BPJS (and relationship with other partner agencies) POLICY OPTION

STRATEGY

INDICATOR/TARGET

Increase the synergies between the MoH, BKKBN the DG Kemdagri in implementation of FP-Health programmes

To formulate a joint ministerial decree amongst MHA, the MoH and BKKBN to strengthen the FP-Health programme

A joint ministerial decree among MoHA, MoH and BKKBN to strengthen the FP-Health programme has been developed, by 2015

There needs to be a policy dialogue forum between the MoH and related health sectors/agencies, especially with BPJS, and between PHO / DHO with health related institutions and BPJS at subnational

To develop a high level policy dialogue forum between the MoH and health-related sectors/agencies either at national or subnational level to support the healthy public policy

A high-level policy dialogue forum between the MoH and health-related sectors/agencies either at national or subnational level will have been conducted at least twice a year since 2015

Need strengthening of system thinking in each institution, particularly the MoH, and PHOs DHOs. By using the principles of Complex Adaptive Systems in management and planning, more locally-specific processes can be designed.

To develop joint ministerial decree (Presidential Decree) for an integrated oversight mechanism for implementation of JKN (MoHA, the MoH, the MoL & Trans, and BPJS)

A joint ministerial decree for integrated oversight mechanism for implementation of JKN has been developed, by 2015

Establish a network of contract and servicelevel agreements (contracts with providers, SLAs with other government entities.

Strengthen the promotion of health public policy. Senior MoH representation on the Board of BPJS.

Establishment and dissemination of health public policy on each of the health-related sectors, by 2015-19. Board membership on BPJS by 2015 SLA/Performance agreements negotiated and implemented.

Enact a joint ministerial decree to establish a network of contract and service level agreements (contracts with providers, SLAs with other government entities, on the grounds that government cannot contract with itself). 106

Institutional Analysis

Capture agreements through SLAs with other government entities, on the grounds that government cannot contract with itself. These would be implemented via a joint ministerial decree, and negotiated through co-ordinating ministers. Financial commitment to health POLICY OPTION

STRATEGY

INDICATOR/TARGET

Assure availability of services for all members of Maintain planned growth of public expenditure on At least 4% of national health devoted to health by the population without exposure to catastrophic health. 2019. health expenditure. Create tax incentives for non-government Year-on-year increase of private sector providers provision of health services. engaging with JKN. Create a system of differential tariffs to encourage Year-on-year decrease in service differentials service provision in underserved areas of the between the least and most generously-served country. districts/cities. Supply-side readiness (including the development of Minimum Service Standard (MSS/SPM) on health) POLICY OPTION

STRATEGY

INDICATOR/TARGET

There is a need for a regulation to replace the existing Permenkes741/2008 on MSS

Develop Permenkes on MSS based on the life-cycle approach, including technical guidance and costing for implementation

Establishment of Permenkes on MSS based on a life-cycle approach including technical guidelines and costing by 2015

Develop a joint team with MoHA for

A joint ministerial decree between MoHA

Facilitating sub-national organisations that

107

Institutional Analysis

have not yet achieved MSS Improve coordination with funding the MSS through APBD

MoHA

for

Strengthen coordination with MoHA for M&E and oversight mechanisms Implement a regulation on sanctions to mitigate non-conformity to the national policy Implement a joint ministerial decree between the MoH and MoHA to undertake joint oversight of district/city authorities Develop an instrument for governors to perform oversight mechanisms with districts/cities in the field of health Promote community participation in improving social responses on health matters

socialisation, facilitation, M&E and oversight mechanism

and the MoH for implementation of MSS will have been developed, by 2015

Implement the M&E and supervision to local government

The MSS will have been socialised and facilitated to district/city governments by MoHA by 2015-16

Follow up on the results of M&E Formulate a Permenkes on an incentive and disincentive policy in overcoming nonconformity to national policy by district/city governments Develop a joint ministerial decree between the MoH and MoHA to conduct an integrated effective oversight mechanism for districts/cities Develop instruments for governors to conduct effective Binwas on districts/cities in the field of health Conduct M&E and supervision on implementation of the obligatory health functions by the MoH

M&E on the implementation will have been carried out (2015-19) M&E will have been followed up throughout the year by: a. Underfunding assisted by the DAK b. Low skills improved by training c. Apply sanctions for deliberate negligence in coordination with MoHA A Permenkes on incentive and disincentive policy will have been developed by 2015 A joint ministerial decree between the MoH and MoHA to conduct an integrated effective oversight mechanism on districts/cities will have been developed by 2015 Instruments for governors to conduct effective Binwas on district/cities in the field of health will have been developed by 2015 108

Institutional Analysis

M&E and supervision on implementation of the obligatory health functions by the MoH will have been conducted throughout the year (2015-19)

109

6. Programmes Programmes are taken to represent the actions and interventions that will be determined within the RPJMN 2014-2019. They refer to some of the specific activities required to implement the aims elaborated in paragraph 5.2. Other more detailed specific actions appear in Annex 7 of the main report. Table 13 identifies priority programme areas and associated activities in the implementation of the main aspects of the institutional analysis, led by the MoH. Areas that are related to the institutional analysis are included even though they appear in other aspects of the Health Sector Review, but this is because they require institutional intervention. Table 13: Selected priority programme areas for the MoH

NO PROGRAMME AREA 1

MAIN ACTIVITIES

Strengthening management 1. Planning and budgeting and health information system

OUTPUT INDICATOR Improving quality of planning and budgeting of health development

(Secretariat General) 2. Data and information

Improving health information system

3. Personnel

Improving personnel administration services

4. Law and organisation

Increasing legal products that support the implementation of health development

5. Health promotion community empowerment 2

Strengthening public health

and Improving health promotion and community empowerment

1. Nutrition

Improving community nutrition status and prevention of nutrition problems

2. Family health

Increasing access to quality of family health services

(DG of Public Health)

Institutional Analysis

NO PROGRAMME AREA

3

MAIN ACTIVITIES

OUTPUT INDICATOR

3. Community health

Increasing access to quality community health services

4. Environmental health

Improving the quality of environmental health and risk factors control

5. Occupational and sport health

Improving access and quality of occupational health and sport health

Prevention and control of 1. Epidemiology and quarantine disease

Improving epidemiological surveillance and disease risk factors

(DG of Disease Prevention and Control) 2. Communicable disease

Strengthening communicable disease prevention and control

3. Zoonotic and vector borne

Improving prevention and control of zoonotic and vector borne diseases

4. Non-communicable (NCD) 5. Mental health 4

Medical services (DG of Medical Services)

1.Primary service 2.Secondary service

3.Traditional,

disease Strengthening non-communicable disease prevention and control

Improving access and quality of mental health

(essential)

(referral)

alternative

medical Increasing access to quality primary medical services

medical Availability of qualified referrals health facilities that are accessible to the public

and Improving access and quality of traditional, alternative and 111

Institutional Analysis

NO PROGRAMME AREA

MAIN ACTIVITIES

OUTPUT INDICATOR

complementary health

complementary health programme

4.Support, infrastructure and health Increasing equitable access to quality of medical support services, equipment facilities and medical devices

5.Quality and accreditation 5

Increasing quality and accreditation

Improving pharmacy 1. Public medicine and health Ensuring the availability, distribution and affordability of drugs and services and quality of supplies medical devices pharmaceutical and medical devices (DG of Pharmaceutical and Medical Devices) 2. Pharmaceutical services

Improving quality of pharmaceutical services and rational use of medicines

3. Medical devices, and household Improving pre-market and post-market surveillance of medical health supplies assessment supplies and household health supplies 4. Standardisation and supervision Improving the competitiveness of the pharmaceutical industry and of medical devices and household medical supplies in the country by fulfilling the standards and health supplies requirements as well as the independence efforts in the field of drug raw materials 5. Pharmaceutical production and Improving

management

of

pharmaceutical

production

and 112

Institutional Analysis

NO PROGRAMME AREA

6

Research and development (DG of NIHRD)

MAIN ACTIVITIES

OUTPUT INDICATOR

distribution

distribution through good governance

1. Biomedical epidemiology R&D

and

clinical Improving Biomedical and clinical epidemiology R&D

2. Health services pharmaceutical R&D

and Improving health services and pharmaceutical R&D

3. Community health R&D 4. Health policy technology R&D 7

Strengthening resources for development empowerment

and

Improving community health R&D health Improving health policy and health technology R&D

human 1. HRH planning health and

Improving quality HRH planning

(DG of HRH Development and Empowerment) 2. HRH education and training

Improving education and training

3. HRH utilisation

Improving HRH utilisation

4. Development and control of Improving development and control of HRH quality HRH quality 8

Increasing oversight & 1. Review accountability of state

Improving planning and financial report

113

Institutional Analysis

NO PROGRAMME AREA

MAIN ACTIVITIES

OUTPUT INDICATOR

2. Audit

Increasing operational performance

3. Corruption prevention

Improving anti-corruption culture

4. Evaluation

Improving policy and performance

apparatus/civil servants (Inspectorate General)

Table 14 refers to priority programme activities that require co-ordinated action across and between agencies. Table 14: Selected priority programme areas for multiple agencies

Programme area

Action

Led by and involving

Legislative programme

Requirement to submit business plans.

MoH, MoHA, PHOs, DHOs

(“Legislation” is taken to include primary legislation, decrees and PerPres) Mechanisms to lubricate inter-ministerial Co-ordinating ministers, MoH, MoHA, BPJS, collaboration using inter-ministry agreements. BPPKN Systems development:

Development of a contracting framework (and MoH BPJS including service level agreements) including the creation of standard contracts for different service types Development and implementation of a business MoH, MoHA, PHOs, DHOs, providers 114

Institutional Analysis

Programme area

Action

Led by and involving

planning framework. Development and implementation of a quality MoH, BPJS management system and leading towards the accreditation of all health facilities, public and private. BPJS to develop contracts with existing Jamkesda BPJS, DHOs, providers schemes. Aligning financial, information governance pathways.

system

and MoH, BPJS, MoHA, PHOs, DHOs, providers

Financial mechanisms which encourage private MoF, MoH, BPJS sector entry into the market place Capacity development:

Improving the quality of clinical coding.

BPJS, providers

Capacity development for contract and service level MoH, BPJS, MoHA, PHOs, DHOs agreement development, quality management. Leadership development, particularly at sub-national MoHA, MoH, PHOs, DHOs level. Improvement Puskesmas. Infrastructure development

of

physician

deployment

to MoH, MoHA, PHOs DHOs

Capital investment in Puskesmas to improve the MoH, PHOs, DHOs, MoPW, MoHA, MoF fabric of existing assets and to create new centres in underserved areas.

115

Institutional Analysis

Programme area

Action

Led by and involving

Support for developing the role of UKBM in advocacy MoH, DHOs, MoHA and planning.

Table 15: Other programme actions

Action

Responsibility

Capacity development at all levels of the SKN to ensure that the practice of business planning and contracting as a means of MoH doing business and managing complex relationships is embedded in systems, capability and culture. The boundary and role anomalies in the distribution of hospitals need to be resolved by re-assigning the stewardship of hospitals MoH, to their respective tiers of governance. DHOs Implement WHO recommendation to establish networks in collaboration with other countries. (Communicable Disease Control).

MoH

Create a 5-year plan and budget for Communicable Disease Control.

MoH

PHOs,

Devise mechanisms to enable the MoH and the MoE to influence the providers of training in both public and private sector MoH, MoE, PHOs, settings equally to ensure that that graduates are produced in the right numbers, with the highest degree of competence and DHOs, training deployed in the most needy parts of the country. providers Agree a definition of what constitutes “management” expenditure so that it can be measured and managed in the future.

MoHA, MoH, MoF

The MoH should surrender its role as direct provider of personal services to an alternative agency.

MoH, MoHA

Appoint a Chief Medical Officer, a Chief Nursing Officer, a Chief Pharmaceutical Officer and a Chief Scientific Officer whose MoH roles would be around the development of their professions rather than as ministry functionaries. Various lines of accountability (horizontal, vertical and bi-lateral) captured in service level agreements, the contents reflecting the MoH, BPJS, roles of the institutions involved. MoHA, MoF, MoE, 116

Institutional Analysis

Action

Responsibility PHOs, DHOs etc.

Improve targeting for poor and near-poor, as more than half of Jamkesmas beneficiaries were not from these groups. Improve socialisation (social marketing) to increase public awareness of entitlement.

MoH, MoHA

BPJS,

Ensure sustainability through improving efficiency and effectiveness of implementation. Make provider payment mechanisms more results-focused. Establish a robust and reliable information system to support M&E, and continuously update the calculation of the programme’s costs. Learn lessons from selected provinces and districts/cities that have attained virtual universal health coverage, such as Bali, Aceh, and Jakarta. Unify and align health information systems across the sector.

MoH, BPJS, PHOs, DHOs, MoF, providers

The contracting mechanism needs to be developed and adjusted to recognise issues of gaming, quality assurance and policy BPJS alignment and to mitigate their effects. Improve the efficiency of operating practices first through a contracting framework which encourages and rewards efficient MoH, BPJS clinical practices, and second a set of credible and agreed clinical guidelines for the management of the major causes of morbidity and, third, unifying the various drug lists into a common approved formulary would help to manage drug costs. PHOs should be required to produce and submit business plans every 3 years, leading to annual operational plans. In planning MoH, PHOs terms, these would become the currency of supervision, in the same way that contract performance is the currency of management. This will require support in terms of developing capacity/skills in business planning. Formalise the allocation process in a consistent and transparent way by making allocations to provinces/districts/cities (that are MoHA,

MoH, 117

Institutional Analysis

Action

Responsibility

irregular or “ad hoc”), and internally within the MoH subject to a robust business case making process.

PHOs, DHOs

Revitalise Puskesmas as part of the Health Systems Strengthening based on Primary Health Care Reform.

MoH, DHOs

Efforts should be made to reduce the administrative burden in the SKN.

MoH. MoHA. PHOs, DHOs

Provide incentives for MSS/SPM implementation for local governments.

MoHA, MoH

The finalisation and implementation of a competence framework for new appointees should be accelerated, and the same MoHA, MoH framework applied to existing post-holders to guide personal development needs. The management and leadership role of provincial governments could be developed as it has the potential to be the pivotal MoHA instrument of downwards supervision and leadership and upwards reporting. Across all areas of the landscape is the pressing issue of creating an effective performance management process

MoHA, MoH, PHO, DHOs

There needs to be a social marketing mechanism to both recruit citizens into HCIS and inform those eligible of their eligibility BPJS status and expected benefits. In terms of direct “health giving” activity, the MoH will need to develop its capacity and its energy in public health functions such MoH, cabinet as: Health surveillance Health status surveillance Water, sanitation and hygiene Epidemic surveillance

118

Institutional Analysis

Action

Responsibility

Health promotion/behaviour change communication (road safety, healthy life style promotion etc) The provision of public “merit” goods

In terms of indirect health system oversight and development, the MoH will need to focus on:

MoH, Cabinet

Regulatory systems. Governance standards. Planning and policymaking. Harmonising medical and technical education production and curriculum development with health system needs. Nationwide licensing regulation (esp. for health professionals). Advocacy (with BPJS, the MoF, MoHA, Parliament, the MPW, PHOs, DHOs, Bappenas etc). HMIS

In terms of system leadership, the MoH should assume responsibility for the development of:

MoH, Cabinet

Clinical protocols/care pathways. Health technology assessment. Emergency/ disaster/epidemic management.

119

Institutional Analysis

Action

Responsibility

The development of clinical standards. The drug regulatory framework (pharmaco-vigilance, DRA, product licensing, QC, GMP, GDP, etc). (Possibly) management of the central purchasing of clinical materials and commodities). Supervision and co-ordination of donor relationships (e.g. GAVI, GFATM initiatives). Ensure that the MoH is well represented at Board-level within BPJS, that the MoH scrutinises BPJS business plans, and that the GoI, MoH, BPJS operational relationship is captured in a Service Level Agreement (SLA) Agree a mechanism whereby providers (public and private) can be admitted to an approved list of providers based on some MoH, BPJS judgement of quality (accreditation, for example). If the licensing role of local health offices were standardised and strengthened and supported by a robust quality assurance BPJS, framework, then the licensing (and possibly accrediting role) of provinces could be deployed by BPJS in the appointment of DHOs providers in the UHC scheme

PHOs,

Enable and encourage non-government provision by neutralising competitive disadvantage experienced through differential MoF, MoH, MoHA criteria for admission to providers’ lists and through minimising the effect of the economic burden of capital financing. Private sector representatives should be included in tariff negotiations.

BPJS

A risk assessment needs to be undertaken, and contingency plans developed to confront the predictable conditions in which BPJS absorptive capacity (fiscal space, skills, information systems, inability to recruit Jamkesda) jeopardises the implementation of UHC. MoHA decree to stipulate that, in the pursuit of operational efficiency and effectiveness, the head of a Puskesmas can be a MoHA functional health worker, rather than uniquely by a structural, bureaucratic occupant. Health activity required of Puskesmas should be grouped into mandatory actions such as health promotion, maternal and child MoH, DOH health, communicable disease and NCD prevention, nutrition, environmental health, emergency and essential health services, 120

Institutional Analysis

Action

Responsibility

and discretionary/desirable health actions accounting for locally-specific health problems in the area (and accounting for the availability of resources). Existing regulatory frameworks64 need to be applied more rigorously to ensure the shifting of services away from curative care.

MoHA, MoH, DHO

Puskesmas need to be strengthened to encourage preventive and promotive activities. District/city governments should be responsibility for calculating the quantity, and skill mix health of the workforce required to PHOs, DHOs equip Puskesmas with the human assets needed to deliver their duties. The status of the head of Puskesmas should be recalibrated not as a structural position but as a functional position.

MoHA

BPJS needs to engineer a contract portfolio that is both manageable and that enables the proper supervision of the performance BPJS of many thousands of providers through a blend of contract types (fee-for-service, block and cost & volume) reflecting local specificities, and to have relationships with local institutions (e.g., DHOs) to whom contract conformity monitoring and reporting can be delegated. Contracts should include a series of quality standards associated with access, clinical quality, efficiency, reporting requirements BPJS and coding accuracy. More sophisticated targets for BPJS are required that combine, consolidation, absorption of existing schemes and the GoI BPJS MoH recruitment of citizens operating in the informal sector. Referral systems should be strengthened, and driven by a protocol/pathway determined by the MoH exercising its leadership MoH and guidance role. Standard operating procedures/protocols (SPOs) for required primary care activities should be developed.

64

MoH, DHOs

Ministerial Decree No. 118/2004, and the strategic plan 1994 are designed to strengthen and support promotive and preventive care.

121

Institutional Analysis

Action

Responsibility

The community should be encouraged to participate in the financing of health (both personal and public health). There is a need to reform the regulation to provide more flexibility to manage the budget to meet Puskesmas needs.

MoHA

Substantial skills and capacity development at district/city-level to support the management and governance of primary care, MoHA, both directly through the business managers, and through district/city-wide activities (health promotion, community engagement PHOs etc).

MoH,

The relationship between districts/cities and provinces is captured and formalised through a Business Plan/SLA arrangement. MoH, This has the added advantage of creating an opportunity for the community empowerment mechanisms to operate at a service DHOs planning and delivery level.

PHOs,

District/city governments should include achievement of the SPM in the 5-year development plan and implemented under the PHOs, DHOs annual plan and budget. The MoH should monitor and supervise the achievement of the SPM. The non-government sector to strive to conform to quality standards (including clinical quality), to share activity and reach data, Providers and to share clinical information at an aggregate level, in order to benefit from publically subsidised income streams. Private sector educators (see Section 6.6) to expose themselves to supervision and regulation of quality conformity, and the regulation of supply in order to allowed access to public clinical environments for clinical placements.” Align reporting/information flows with accountability and financing flows.

MoH, MoF, MoHA

All institutions on the demand side will need to develop systems, processes and competences to recognise and influence the MoH, BPHS, impact of market weaknesses inherent in the health care context. PHOs and DHOs Alter the BPJS share of allocation to include semi-variable cost elements in the tariff values as well as the planned variable BPJS costs. Engineer contract or service-based agreements between local governments and provinces, and between provinces and the MoH, MoH. The agreements would capture the intentions set out in the business plans, would require that the agreements are two- DHOs,

PHOs, BPJS, 122

Institutional Analysis

Action

Responsibility

way (funds one way; performance the other), and would capture agreements around quality, access, equity and efficiency

MoHA

Extend the concept of matched funding to create incentives for districts/cities to conform to central priorities, and to provide an MoHA, MoH instrument for the MoH to encourage greater equity. Require provincial health agencies and district/city government health agencies to produce locally-specific costed business MoH, plans that reflect both RPJMN priorities and local specificities, so that funding can be aligned with health gain objectives, and in DHOs which allocations, releases and budgets are transparent (as would connections/disconnections between them).

PHOs.

Re-visit the structure and roles of units within the MoH, provinces and districts/cities to ensure that no conflicts of interest exist MoH, (e.g., between regulatory functions and management functions). DHOs, MoHA

PHOs, BPJS,

123

Institutional Analysis

7. Indicators In addition to the specific indicators that appear in Section 0, other systems-performance indicators are suggested. Two types of indicator have been proposed for measuring governance: rules-based and outcome-based. Rules-based indicators measure whether appropriate policies, strategies and codified approaches for health system governance exist. In the health systems context, these indicators include the existence, for example, of a national essential medicines list or a national policy on malaria control. They are part of a larger class of indicators called governance determinants. In addition to the existence of rules (called “formal procedures”), the determinants of health-careprovision governance include four other broad categories: ownership arrangements, decentralisation, stakeholder participation, and contextual factors. In this framework, determinants of governance are contrasted with governance performance. Outcome-based indicators measure whether rules and procedures are being effectively implemented or enforced, based on the experience of relevant stakeholders. For health systems, examples may include the availability of essential medicines in health facilities or the absenteeism of health workers. Since the outcome-based indicators relate directly to the functioning of other health system “building blocks”, only the rules-based indicators for measuring health system governance are presented. Using the WHO Building Blocks Approach, 65 a review of national health policies in respective domains (such as essential medicines and pharmaceutical, TB, malaria, HIV/AIDS, maternal health, child health/immunisation) is indicated. This represents a composite “basket” of measures that the WHO considers to efficiently represent the effectiveness of a health system. In the case of the HSR, several parallel work-streams are likely to propose similar indicators, especially finance and quality. In Table 16 a selected shortlist of indicators is identified. Table 16: Summary of proposed indicators for health systems governance

Indicator

Score (1/0)

Existence of an up-to-date national health strategy linked to national needs and priorities Existence and year of last update of a published national medicines policy Existence of policies on medicines procurement that specify the most costeffective medicines in the right quantities; open, competitive bidding of suppliers

65

WHO (2010), Monitoring the Building Blocks of Health Systems, Geneva

124

Institutional Analysis

of quality products Tuberculosis: existence of a national strategic plan for tuberculosis that reflects the six principal components of the Stop-TB strategy as outlined in the Global Plan to Stop TB 2006–2015 Malaria: existence of a national malaria strategy or policy that includes drug efficacy monitoring, vector control and insecticide resistance monitoring HIV/AIDS: completion of the UNGASS National Composite Policy Index questionnaire for HIV/AIDS Maternal health: existence of a comprehensive reproductive health policy consistent with the ICPD action plan Child health: existence of an updated comprehensive, multi-year plan for childhood immunisation Existence of key health sector documents that are disseminated regularly (such as budget documents, annual performance reviews and health indicators) 10. Existence of mechanisms, such as surveys, for obtaining opportune client input on appropriate, timely and effective access to health services Policy index: sum of the scores of 10 indicators; maximum score is 10 Scoring: If adequate policy does not exist or cannot be assessed: 0; If adequate policy is available: 1 Table 17: Indicator 1, Governance:

2015 1

Policy index.

2016

2017

2018

2019 10

125

Institutional Analysis

Table 18: Matrix of indicators for RPJMN 2015-2019; Institutional Analysis Area of HSR

Indicators

2015

2016

2017

2018

2019

Notes and sources

Share of APBD going into health

10%

Target is at least 10%.

Reduction in scale of the MoH

13%

System level

Data source is DAU and DHAs

Development of a citizen-based health information system.

Year-on-year in terms of administration costs of the SKN from 26% to 13% Implemented by 2019

Input level Management minimised

costs

Rate of replenishment as a % of decapitalisation

Public management costs 0

>0

>0

>0

A measure of the sustainability of BPJS. Must be a number greater than 0

Regulation/policy on financial management of hospitals and Puskesmas

Implemented by 2016.

Establishment of a Medical Advisory Board (as mandated by Law No. 36/2009 on Health)

Implemented by 2016.

126

Institutional Analysis

Establishment of guidelines for community empowerment Integration with related sectors (under the coordination of BPMD)

Implemented by 2017.

Establishment of Complaint Resolution Unit at the health centre and hospital)

Implemented by 2015.

The number of health workers in remote and difficult locations.

Year-on-year growth.

Increased availability of vaccines and essential drugs at Puskesmas

Stock-out rates reduced to 0% by 2019

Output level % the population covered under a single scheme managed by BPJS

100%

Number of insured, 47% and who clearly know that they are insured.

100%

2019 – 257 million

The needs of the underserved population reflected in tariffs.

Number of Jamkesda incorporated into BPJS; number

2014 – 151 million

Tariffs adjusted to enable more vulnerable areas to spend more – so a loading in the tariff to reflect health needs. 100%

A measure of reputation and trust of BPJS. Supposed to be

127

Institutional Analysis

remaining independent

100% by 2019

Number of private sector providers conforming to agreed quality standards

A measure of potential capacity when coupled with the public sector provision. BPJS needs to make this target.

Contracts in place that reward efficiency.

Yes/no – contract construction by BPJS

Number of private sector providers conforming to agreed quality standards.

BPJS needs to make this target

Appointments/clinics cancelled due to the unavailability of personnel.

Year-on-year reduction by 10%

Percentage of health care facilities that meet quality standards

Year-on-year of 5%

Development and implementation of an accreditation system for primary health care facilities

Yes/no by 2018

increase

Outcome level Number of poor people utilising health services facilities Proportion of pregnant women making at least 4 ante-natal

A measure of the reach of the JKN.

100 %

128

Institutional Analysis

attendances Immunisation coverage

% second dose of measles immunisation to under fives: supposed to be 90% flat from 2015-19

Achievement of MDG 4

Theoretically by end2015

Achievement of MDG 5

Theoretically by end2015

Reduction in inequity (a basket of measures; Rp per capita, physicians per capita, nurses per capita, referrals per capita)

The MoH to devise a measure of equity comprising measures of access. Measure to be devised by end-2015

The indicators above, when taken together, will give a good indication of the capacity and efficiency of the institutions involved in the SKN to deliver their objectives efficiently, effectively and consistently. However, some indicators tend to reflect the preoccupations of parallel aspects of the HSR, and so a selection has been highlighted that 6is thought to conform to the requirements of good indictors. These requirements are as follows •

Sensitive: they vary in a way which reflects changes in performance.



Specific: they measure what they purport to measure and are relatively uncontaminated by externalities.



Significant: what they measure is important.



Schematic: actions to manage the trajectory of the indicator are clear.

129

Institutional Analysis

ANNEXES

130

Institutional Analysis

Annex 1: Landscape by flows Figure 11: Institutional relationships

National Parliamen t

Central Governm ent

BKKEN

B P J S

Bad an PO M

Ministry of Health

Ministry of Home Affairs

Provinci al Parliam ent

Provincial Governm ent Central Governm ent Hospitals

Family Planning Agency

Provincia l Health Office

Provincial Hospitals

District/Ci ty Governm ent

District Parliam ent

131

Institutional Analysis

Family Planning Agency

authority

regul ation

District Health Office

District Hospitals

District Commun ity Health Centres

Private Hospitals/provi ders

finance

A critical issue is the absence of alignment both between some systems and within some systems which would normally be expected to be parallel. Flows of accountability, money and information are not aligned, so the effectiveness of planning, regulation and supervision is compromised; and equally the flows of money, information and accountability are not harmonised through the system, and so the mechanisms for planning and regulation lack instruments of implementation (i.e. money). Generally it would be expected that the three sets of relationships above would be aligned such that the flows are in parallel, enabling more effective monitoring, planning, supervision and budgeting. It could be argued that the RPJMN and the MoH Strategic Plan offer the lubricant required, and that is to some extent true, in that they offer a united and shared framework and language. The significance of this figure is that the various flows of information, finance and regulation are not aligned, and they need to be made so.

132

Institutional Analysis

Annex 2: Stakeholders Description of the Ministry of Health The MoH is the lead agency in the health care landscape. It is currently (March 2014) undergoing a restructuring which is designed: • • • • •

To re-align the Ministry with the Life-Cycle” approach. Reduce the scale of the MoH. Create a more symmetrical structure. Simplify internal communications and reduce administrative blockages. Create a structure more closely aligned with the new architecture of SKN (BPJS, UHC, decentralisation revision). The context in which the MoH operates is summarised below. Figure 12: Ministry of Health context

133

Institutional Analysis

Figure 13: Ministry of Health structure to 2014

STRUKTUR ORGANISASI KEMENTERIAN KESEHATAN 1.Staf Ahli Bid. Teknologi Kesehatan dan Globalisasi; 2. Staf Ahli Bid. Pembiayaan & Pemberdayaan Masyarakat; 3. Staf Ahli Bid. Perlindungan Faktor Risiko Kesehatan; 4. Staf Ahli Bid Peningkatan Kapasitas Kelembagaan & Desentralisasi; 5. Staf Ahli Bid. Mediko Legal.

(Up to 2014)

LAMPIRAN PERATURAN MENTERI KESEHATAN NOMOR 35 TAHUN 2013 TENTANG PERUBAHAN ATAS PERATURAN ME 1144/MENKES/PER/VIII/2010 TENTAN KERJA KEMENTERIAN KESEHATAN MENTERI KESEHATAN STAF AHLI WAKIL MENTERI KESEHATAN

SEKRETARIAT JENDERAL INSPEKTORAT JENDERAL SEKRETARIAT ITJEN BIRO PERENCANAAN DAN ANGGARAN

INSPEKTORAT

DIREKTORAT JENDERAL PENGENDALIAN PENYAKIT DAN PENYEHATAN LINGKUNGAN

DIREKTORAT JENDERAL BINA UPAYA KESEHATAN

BIRO KEPEGAWAIAN

DIREKTORAT JENDERAL BINA GIZI DAN KESEHATAN IBU DAN ANAK

DIREKTORAT JENDE BINA KEFARMASIAN ALAT KESEHATA

DIT DIREKT

DIT DIREKTORAT

DIT DIREKTORAT

DIT DIREKTORAT

BADAN PENGEMBANGAN DAN PEMBERDAYAAN SDM KESEHATAN

BADAN PENELITIAN DAN PENGEMBANGAN KESEHATAN

SEKRETARIAT BADAN

SEKRETARIAT BADAN PUSAT

PUSAT DATA DAN INFORMASI

PUSAT KERJASAMA LUAR NEGERI

BIRO HUKUM DAN ORGANISASI

SEKRETARIAT DITJEN

SEKRETARIAT DITJEN

SEKRETARIAT DITJEN

BIRO KEUANGAN DAN BARANG MILIK NEGARA

PUSAT

PUSAT PENANGGULANGAN KRISIS KESEHATAN

PUSAT PEMBIAYAAN DAN JAMINAN KESEHATAN

PUSAT KOMUNIKASI PUBLIK

PUSAT PROMOSI KESEHATAN

PUSAT INTELIGENSIA KESEHATAN

Directorate General of Public Health DIREKTORAT JENDERAL KESEHATAN MASYARAKAT

SEKRETARIAT DIREKTORAT JENDERAL

DIREKTORAT GIZI MASYARAKAT

DIREKTORATK ESEHATAN KELUARGA

DIREKTORAT KESEHATAN KOMUNITAS

DIREKTORAT KESEHATAN LINGKUNGAN

DIREKTORAT KESEHATAN KERJA DAN OLAHRAGA

Directorate General of Medical Services

DIREKTORAT JENDERAL PELAYANAN MEDIK

SEKRETARIAT DIREKTORAT JENDERAL

DIREKTORAT PELAYANAN MEDIK DASAR

DIREKTORAT PELAYANAN MEDIK RUJUKAN

DIREKTORAT KESEHATAN TRADISIONAL, ALTERNATIF DAN KOMPLEMENTER

DIREKTORAT PENUNJANG, SARANA, PRASARANA, DAN

PERALATAN KESEHATAN

DIREKTORAT MUTU DAN AKREDITASI FASYANKES

National Institution of Health Research and Development

BADAN LITBANG KESEHATAN

SEKRETARIAT BADAN

PUSLITBANG BIOMEDIS DAN EPIDEMIOLOGI KLINIK

PUSLITBANG PELAYANAN KESEHATAN DAN FARMASI

PUSLITBANG KESEHATAN MASYARAKAT

PUSAT KAJIAN KEBIJAKAN DAN TEKNOLOGI KESEHATAN

Secretariat General SEKRETARIS JENDERAL

BIRO PERENCANAAN DAN ANGGARAN

PUSAT DATA DAN INFORMASI

BIRO KEPEGAWAIAN

PUSAT KERJASAMA LUAR NEGERI

BIRO KEUANGAN DAN BARANG MILIK NEGARA

PUSAT PENANGGULANGAN KRISIS KESEHATAN

PUSAT PEMBIAYAAN KESEHATAN

BIRO HUKUM DAN ORGANISASI

PUSAT KOMUNI KASI PUBLIK

BIRO UMUM

PUSAT PROMOSI KESEHATAN

PUSAT KESEHATAN HAJI

Institutional Analysis

Table 19: Characteristics of the previous and new MoH structures

PAST STRUCTURE (based on Ministerial FUTURE STRUCTURE: Regulation no 35/2013): Minister of Health

Minister of Health

Vice Minister

8 Echelon I:

13 Echelon I:



Secretariat General



Secretary General



Inspectorate General



Inspector General



Directorate General of Public Health



Directorate General of Health Services





Directorate General of Disease Prevention and Environmental Health

Directorate General of Disease Prevention and Control, and Environmental Health



Directorate General of Nutrition and Maternal and Child Health



Directorate General of Health Services



Directorate General of Pharmaceutical and Medical Devices

Directorate General of Pharmaceutical and Medical Devices



National Institute of Research and Development

National Institute of Research and Development



National Board of Human Resources for Health Development and Empowerment

National Board of Human Resources for Health Development and Empowerment



5 Senior Adviser to Minister of Health









5 Senior Adviser to Minister of Health

The new structure is: •

More focused on primary care.



Better aligned with the MoH landscape (see Figure 12).



Less top-heavy. The MoH will comprise a number of Directorates, Centres and Bureaux. The Directorate General has the task of formulating and implementing policy and technical standards in the fields for which each Directorate is responsible. The Directorate is the implementing arm of the Ministry. Planned Directorates are:



Public Health including responsibility for; 140

Institutional Analysis









Family Health



Nutrition



Community Health



Environmental Health



Occupational Health

Prevention and Disease Control including responsibility for; •

Infectious Diseases



Zoonoses and animal-borne infectious diseases



Infectious Diseases



Non-communicable diseases



Mental Health

Medical Services including responsibility for; •

Basic Medical Services (inc Primary Care)



Medical Referral Services (secondary and tertiary care, including non-Government hospitals).



Traditional Medicine.



Clinical support (infrastructure and equipment)



Quality and Accreditation

Drugs and Health Equipment including responsibility for; •

Directorate of Medicines and Health Supplies



Directorate of Pharmaceutical Services



Technology Assessment Directorate



Equipment Standardisation



Directorate of pharmaceutical production and distribution

In addition there is to be a matrix of supporting agencies, centres and bureaux, notably; •

Agency for Health Care Research and Development



Agency for Human Resources for Health The Secretariat General is an administrative/bureaucratic element of the MoH and is responsible for; 141

Institutional Analysis



Planning and budget.



Staffing.



Finance and BMN.



Law and organisation.



Common services. The Minister has a panel of expert divisions including those which advise on;



Global Health



Health Economics



Risk and health technology



Decentralisation and bureaucratic reform



Medico Legal Finally there is a range of Centres, including those responsible for;



Data and information



Foreign Cooperation



Emergency and Crisis Management



Health Financing



Public Communication



Health Promotion



Hajj Health

142

Institutional Analysis

Table 20: Ministry of Health features

Identification of Need Involved in

Role

Demand - side role

Supply- side role

Outcome

The MoH is either involved in, or leads the country’s response to all aspects of the health landscape.

The MoH should be focused on policy making, monitoring, evaluation, regulation, facilitation (pembinaan), governance and managing strategic workforce planning (7 basic medical specialists) and managing hospital class A, the MoH should have the capacity to regulate and oversee health service delivery at a national level. There should be awareness that the ultimate responsibility and accountability for health matters lies with central Government through the MoH). In the context of a unified State, the relationship between MoH and local government health agencies is hierarchical. Therefore it remains the responsibility of the MoH to supervise and to facilitate (pembinaan) local government health agencies.

Relationships The MoH has a complex set of relationships, some of which are bilateral (e.g. with BPJS), some of which are hierarchical (e.g. with Type A hospitals) and some which are “diagonal” (e.g. with District Governments). The key ones are with BPJS as the payer for curative services, with Professional Associations, with the Media and with other Ministries. The complexity of this network of relationships leads to a number of both benefits and difficulties. Benefits include the wide-ranging involvement and influence in all matters to do with health and especially including the health–related activities of other Ministries. The difficulties include the variety of mechanisms available to exercise influence. In the context of a decentralised structure, the leverage available to the MoH to engineer the implementation of policy which the MoH has set depends on negotiated relationships with bilateral and “diagonal” partners. The outcome of this may be inconsistency across the country and the weakening of the policy leadership for which the MoH is responsible (and accountable).

Processes

The Ministry of Health is a large organisation which combines many roles and types of role. It is simultaneously regulatory, executive, supervisory, strategic, operational, technical and supporting. Its advocacy role is confused by its supervisory role.

Policy setting (including the contents of the basic package of health 143

Institutional Analysis

services, encouraging greater equity). Establishing and regulating standards. Technical support to Provinces, Districts and providers. Co-ordination of different aspects of the SKN. Advocacy at Government-level. Managing Type A hospitals. Direct intervention in health care development (through allocations).

Transactions

Contribution to RPJMN Exercise of oversight of decentralised institutions. Advocacy (especially around resource generation). Negotiation of service provision with providers and BPJS.

Risk

Mitigation

The role of the MoH vis-à-vis the BPJS will represent a significant conceptual and practical issue. Particularly, unanswered questions around what mechanisms will exist to enable MoH to exercise its policy and planning duties, (and to ensure implementation), what mechanisms will exist for MoH to encourage BPJS to reflect national policy and plans through its contracting and re-imbursement processes, what mechanisms exist to capture policies around equity and reaching the poor, vulnerable and indigent communities, what mechanisms will exist to ensure that the outcome of MoH (and local) licensing of facilities is reflected in BPJS contracting agreements and what mechanisms exist to align the quality assurance duties of the MoH and the quality requirements established by BPJS through its service agreements.

A Service Level Agreement with the BPJS. Reduce the scale of the MoH to enable a focus on its principal functions. Dispense with its managerial role over Type A hospitals. Exercise of its leadership function through the requirement for Provincial Governments to produce Business Plans. Use Business Plans as the instrument of supervision.

144

Institutional Analysis

The MoH, as an organisation, is heavy. The heaviness may produce a lack of agility and a slow response to prevailing challenges.

There is some evidence that Ministries of Health worldwide have been unable to cleanly distinguish the policy/strategic role from an executive role66.

Opportunity

Opportunity exists in a number of directions, including the opportunity to promote the development of primary care through the new mechanisms to finance Puskesmas, through to an opportunity to influence supply-side agents (BPJS and through the private sector).

Impact

A SLA with BPJS would have the effect of capturing and formalising a relationship between supply and demand-side factors, so the alignment/harmonisation of both supply and demand sides would be strengthened (providing that BPJS is inclined to negotiate such an agreement).

A reduction in the scale of the MoH (allied with distancing itself from the Type A hospitals) would have a number of effects, positive and negative. The negative ones include the need to create a hospital management agency and the (temporary) destabilisation of the MoH. The positive ones include a broadcast message about the policy leadership which the MoH would capture, a more agile and responsive organisation, and a reduction in the conflicts of interest inherent in being both a regulator and provider of services.

66

Ham, C, (2000), The Politics of NHS Reform 1988-97: Metaphor or Reality?, King's Fund, London

145

Institutional Analysis

1.

Description of BPJS Identification of Need ? Involved in

Role

Demand - side role

Supply- side role

Outcome

Stimulating demand through social health insurance schemes and engaging suppliers of services to

The BPJS has a number of roles which will change and develop over the coming 5 years. The BPJS has two important roles; first it is a reimbursement agency responsible for reimbursing health care providers for services rendered, and second for recruiting 100% of the population into the BPJS health protection scheme. In theory it is not involved in setting health policy, which is the responsibility of the MoH, but as a singlepayer mechanism the role of the BPJS is de facto strategic.

Relationships The BPJS has business relationships with all public sector health care providers and with many private sector providers. These relationships are governed by contracts.

It also has a relationship with the MoH in their role as a service provider (Type A hospitals) and with the MoH as the policy lead in the health sector.

Processes

The main activities of the BPJS include:

Consolidating existing Insurance Schemes into a consistent single scheme. Recruiting all of the population into the UHC scheme by 2019. Reimbursing providers for services rendered. Admitting qualified providers into the panel of approved providers (“credentialling”). Negotiating contract terms with providers, representing tariff, quality and volume. Managing reporting and service tracking.

146

Institutional Analysis

Transactions

The main transactions in which the BPJS are engaged include: Absorbing existing health financing schemes. Admitting providers through credentialling. Service delivery contracts. Reimbursement. Citizen registration. Information collection, processing, disbursement and analysis. Negotiating tariff based on INA-CBG. Intervening in reimbursements for outlier patients.

Risk

Mitigation

The casualisation of parts of the workforce.

SLA/Contract relationship with MoH.

Not recruiting all citizens.

More sophisticated contracts with providers to more adequately represent quality and access The costs of caring for presently uninsured issues. citizens may risk an overspend if those patients represent a higher need segment Positive discrimination of tariffs in favour of of society. underserved Districts; the “excess” to be deployed specifically for service development. Unclear/contestable contents of the basic package. Publish a more specific list of the contents of the basic package, specifying exactly Ensuring continuous quality improvement. inclusions and exclusions. Mechanisms to rectify inequity are not Ensure that MoH is well-represented at the clear. BPJS Board level. Tensions between BPJS and MoH.

Develop formal relationships between BPJS local offices and DHOs and the DistrictThe relationship between BPJS and subappointed Business Managers assigned to national Governments is problematical. villages. No obvious mechanism to harmonise reimbursement and the wider health economy, or to deploy BPJS engagement in the development of local strategies to deliver community-wide health gains. Perverse

incentives

inherent

in

the 147

Institutional Analysis

decentralised system.

Opportunity

Once UHC has been achieved, a mechanism exists to balance the health care economy to reduce inequity, to enable access to care for those most vulnerable and to minimise the risk of catastrophic health expenditure. Through the contract mechanism to support the development of Primary Care. Through the contract mechanism to remove the perverse incentive (upcoding, upward referrals). Through the contract mechanism to secure efficiency gains at both provider ad system levels.

Impact

Improved access. Improved care quality. Efficiency gains. Harmony between MoH, local and BPJS policies.

148

Institutional Analysis

a. Description of other stakeholders; sub-national governments. Identification of Need

Involved in

Demand - side role

Supply- side role

Outcome

Provincial, District and Village entities have a constitutionally-secured devolved range of powers and responsibilities. They all represent significant degrees of service provision and responsibility for many aspects of public health and health protection. Many provide social protection schemes (which will probably be absorbed into BPJS). Therefore, they all represent all four interests.

Role

The role of the sub-national health offices range from direct management of the Type B hospitals, the Types C and D hospitals and Puskesmas services, quality assurance, management of APBD and DAK incomes, licensing of facilities, deployment of human resources and advocacy with MoH/PHOs etc.

Relationships Provincial and District Governments enjoy organisational relationships across all elements of the health sector, and bilateral relationships with other parts of the economy (public and private). Particularly significant is their relationships with the MoHA as the main funding channel into subNational governments.

Processes

The management of hospitals and Puskesmas. Planning and management of the human assets. Advocacy “upwards”. Public consultation. Licensing.

Transactions

Budgeting. Capital allocations. (Some) manage Jamkesda.

149

Institutional Analysis

Risk

Mitigation

Anecdotally, it is reported that there is limited capacity within PHOs and DHOs to exercise an assertive and leadership role in an environment characterised by a complex mix of centrally managed social health insurance allied with the underlying and non-negotiable features of a decentralised governance and government framework.

Opportunity

Acceleration of the completion and implementation of the competence framework. Development of a national Quality Strategy. Development and improvement of the Service Standards Framework. Development of SLA/Contract relationships both hierarchically and bilaterally.

The Provincial level has the potential to develop as the key link in the chain of command; sufficiently local to ensure sensitivity to local needs and priorities, and of sufficient scale to interact with central Government meaningfully. A formal Constitutional role for Provincial Governors who are the representatives of the Centre gives them opportunity to exercise central prerogative locally. The new Provincial support agencies and District “business managers” present the opportunity to share and extend capacity and skill, and to make local services more responsive to locally-specific needs. The role of Provinces and Districts as the interface between providers and the BPJS will become critical (there are too many direct contracts for BPJS to manage directly and in detail).

Impact

Greater equity. More able local management. More autonomous provider institutions. Improved service quality.

150

Institutional Analysis

b. Description of other stakeholders; providers. Identification of Need

Demand - side role

Supply- side role

Outcome

Involved in

Public sector hospitals, public sector primary care providers and nongovernment providers operate from single-practitioner settings (usually in the non-government sector) through to the large teaching, Type A hospitals (usually public sector). Other providers include virtually vertical programmes (Family Planning), and state enterprise providers.

Role

The roles of providers include: The provision of personal services (curative, preventive and promotive). The provision of population-level services (public health, epidemiology, health education, surveillance, disaster preparedness).

Relationships Many relationships exist, including: Policy: •

MoH



Provincial Government



District Government.



Village.

Reimbursement/payment: •

MoH



MoHA



Jamkesmas/Jamkesda



BPJS



Patient (OoP)



Provincial/District Governments

Bilateral: •

Family planning

151

Institutional Analysis

Processes



Public health laboratory services



Medical Schools

The delivery of hospital and primary care services. The delivery of public health services. Management of the human assets. Quality management. Asset management.

Transactions

Referrals between tiers of health care provision. Revenue collection and disposal. Patient – level reporting. Public health – level reporting.

Risk

Mitigation

Anecdotally, it is reported that there is Provincial support agencies. limited capacity to effectively manage resources in a relatively autonomous District “business managers” setting (when finance will flow directly to The maturing of the contracting culture Puskesmas). The loss of any “top up” income from District Governments.

The competence framework

Efficiency improvements are available (particularly in secondary care) but will take skilled leadership to secure them. Systems capacity (for HR, invoicing and financing management).

Opportunity

The contracting mechanism creates the opportunity for providers to determine their own performance, based on outputs rather than inputs, reflecting local needs and capacities. If the “business managers” and Provincial Support Agencies develop a 152

Institutional Analysis

mutuality with their “client” District and Village counterparts, then this will strengthen the relationship with Provinces/Districts.

Impact

Greater equity. More able local management. More autonomous provider institutions. Improved service quality.

153

Institutional Analysis

2.

Description of other stakeholders; government agencies. Identification of Need

Involved in

Role

Demand - side role

Supply- side role

Outcome

Other government agencies include the Government itself, fellow Ministries (MPW, MoE, MoHA , MoF etc) and BAPPENAS. All have both an interest and a role in contributing to health outcomes. Equally all have an interest in the contribution which the health sector can make to their principal purposes.

The nature of the decentralised structure means that there are many outcomes which require coordination and harmony across sectors and between agencies. Relationships are both vertical and horizontal, and generally are exercised through negotiated mechanisms. For example, the role of potable water and sanitation supervised by the MPW has a dramatic effect on health status; the MoE has a role in supervising and planning higher education (and therefore the supply of qualified health workers to populate roles in the health sector), the MoF has an interest in both the budget planning/ management process and the macro-economic impact of the health sector, and MoHA is the main financing channel for expenditure at sub-national level. BAPPENAS is responsible for planning across all sectors and so has a critical function in ensuring harmony between agencies and the coordination of financing and investments. BAPPENAS also leads the preparation of the RPJMN.

Relationships Many relationships exist, including: Policy: •

Government/Cabinet determines policy, the legislative programme and the major macro-economic decisions.



MoE determines the planning and delivery of public sector Higher Education; MoH has a partnership role in influencing the labour economy for health professionals, and health care providers have a role in providing placements for the vocational aspects of the learning.



MoH provides the health-specific contribution to the RPJMN.

Reimbursement/payment: •

MoF and MoHA determines the budget and expenditure allocation for most of the decentralised functions (including health).

154

Institutional Analysis

Bilateral:

Processes



MPW undertakes major infrastructure investments, and the healthrelated elements are agreed through a negotiated process.



Similar arrangements operate with other Ministries.

RPJMN. Annual budget process. Contribution to the legislative programme. Budgeting for the revenue consequences of capital schemes.

Transactions

RPJMN planning, consultation and implementation. Legislative programme. Investment negotiation. Advocacy.

Risk

Mitigation

New cabinet.

RPJMN 2014



A new Cabinet in 2014 will present Independence of BPJS risks around the continuation of the of the reform programme (UHC etc), the Consolidation legislation 2014 planned increase in health expenditure and the reform of the MoH (assuming a new Minister). The Cabinet (and Minister) may be new to the health sector and so in conditions of some political turbulence may be less likely to continue the programme to which predecessors were committed.

decentralisation

The political implications of any inability to recruit 100% of the population to UHC by 2019. A lack of awareness interrelationships between influencing health status.

about the the factors

155

Institutional Analysis

Opportunity

RPJMN •

The complex planning and budgeting processes required to assemble the RPJMN will be very difficult to jettison and replace without destabilising the economic and service delivery systems across the government sector.

New decentralisation Law •

Encapsulates the sub-national roles, limiting the risks associated with central Government immaturity.

New MoH structure •

Impact

More agile and more flexoible structure may support a new administration without the antibodies associated with “inheritance”.

Greater equity. Improved service quality. Improved health status outcomes

156

Institutional Analysis

3.

Description of other stakeholders; the fifth estate. Identification of Need

Demand - side role

Supply- side role

Outcome

The fifth estate is that sector of the economy which has a legitimate interest in health outcomes and the health sector, but which lies outside the immediate structures of governance and delivery. It includes: Involved in

The Media Professional Associations The Legal sector

Role

The role of these organisations and interests is to: Influence policy. Represent their stakeholders. Hold the public sector to account. Reflect public or sectional interests.

Relationships Many relationships exist, including: Professional Associations will represent many individuals in positions of leadership and influence; they also have a role in determining professional standards and influencing educational practices. The media will have an interest both nationally and locally in questioning and challenging the actions and performance of the health sector, and will also have a role in public engagement particularly around public education and information-giving. In an election year, then the media will have a higher profile. The legal profession will have a role which will increase as the amount of legislation in force increases, particularly around medical litgation, access to services to which citizens are entitled and in representing interests such as the private nd pharmaceutical sectors.

Processes

Reporting

157

Institutional Analysis

Lobbying Use of the legal framework Membership of fora (committees etc)

Transactions

Publication Actions of litigation Submissions and activity in consultation processes

Risk

Mitigation Public consultation processes

That sectional interests may have a Transparency in matters of public interest disproportionate impact on wider public Involvement of the professions and the private interest sector in advisory, consultation and planning systems.

Opportunity

Use of the media as a positive instrument of health promotion

Impact

Greater public and political awareness of health issues. Reduction in examples of vexatious litigation. Greater commitment from the professional cadres. Other health outcomes including better lifestyle choices

158

Institutional Analysis

Annex 3: Contract types and how do we provide incentives to develop efficiency and quality? Common Elements of Purchaser/Provider Contract: It can be expected that the overall form of the purchaser-provider contract for Government funded services will be established by MoH and the purchasing authority (BPJS) and that this will be “not negotiable”. Equally, it will be an evolving process and there will be variations to the form and content from one year to the next, particularly in the early years. Providers should seek to be involved in the process of deciding the form of the contracts. The major elements of a contract between a purchaser of health services and the provider organisation would include, •

Details of the types or categories of services to be provided (including pharmaceuticals)



Volumes of each type or category



Prices of services including limits of overall funding (if any),



Sanctions and rewards to be invoked in the event of volume and/or quality variations and the process of managing variations



Coverage – who is entitled to what services



Principles of access to be observed (e.g. frequency of clinics, outreach services, emergency care etc)



Scope of the overall service umbrella – health priorities, preventative, curative, promotion etc and a reasonable definition of the overall service environment to be provided



Indicators of quality (covering clinical, efficiency, customer service, financial) that will be required to be measured and the standard of performance that will be required to be met.



Specification of how quality indicators will be measured, i.e., data requirements and formulas)



Reporting requirements



Obligations and rights of both parties including obligations that a provider organisation will be required to impose on the individual health service professionals that it employs or contracts.



Conditions under which a contract may be varied.



Conditions under which a contract may be terminated.

159

Institutional Analysis

Table 21: Main contract types

Block •

Simple contract



Fixed price for agreed volume



Cap and collar variances agreed

Structure

Features



Simple to manage



Low needs





Suitable for

Cost and Volume •

Relatively complex to negotiate



Simple to manage

Lacks versatility for • turbulent environments Helps cash flow in providers



Stable services



Long term services

Balloon of investment at start of contract • period, and thereafter a volume related element



information

Fee for Service/Spot

Cost per service.

item



Flexible.



Simple needs



Complex needs



Unpredictable flow.

Less predictable cash • flow.

of

information

financing

cash

Complex costing



Opportunity negotiation



May create cash flow problems for providers



Complex monitoring of providers



New services



Services being • purchased in response to newly emerging needs/demands

Established services, but for which demand is very variable.



Complex needs

information

for

re-

The usual mechanisms to secure quality are: 1. Accreditation/licensing (i.e. To institutions, expecting that they would exercise control over

the professionals in their role as employers.) 2. Professional revalidation/re-licensing (i.e. To individual professionals in their roles as

professionals.) 160

Institutional Analysis

3. Through the market - referrers/ contracting/tariffs/third party payers /payment by results.

3.1. Through business-to-business relationships, captured in some form of contract mechanism. 3.2. Through the marketplace (either managed or free) with agents (often primary care professionals) acting as agents of the state or community. 3.3. Through a free market, where patient choice rewards quality and efficiency (assuming price control). This in turn generates a number of levers/mechanisms to manage the incentive. Mechanism 3 also requires that there is: •

A form of contracting mechanism, usually based on one or more of three main forms of contract



A tool for the analysis of the provider in its market place - often the GR-MS framework applied to health care contracting to help analyse market opportunity.



Agreed contract currencies based on the delivery of:



Inputs



Outputs



Outcomes



DRGs



HRGs



Pathways

Naturally, the two main strategies are sanction for non-conformity and/or reward for conformity to quality outcomes. Most countries apply some combination of the 3 mechanisms, and will apply both sanction and reward. The mechanisms: 1. Accreditation/licensing (i.e. To institutions, expecting that they would exercise control over the professionals in their role as employers.) In many countries the funding agency (government, MoH, Social Insurer, Commercial Insurer, HMO etc) will apply some form of regulatory framework to providers. The expectation will be that the right to treat patients will be earned only through conformity to some standards of clinical quality, resource efficiency (payers will not wish to reward providers' inefficiency), and often also conformity to environmental considerations and to good employment practices. This requires some form of inspection regime, allied with sanction and reward. In such cases, what is usually regulated is inputs and outputs; only rarely is outcome a part of the regulatory framework. 161

Institutional Analysis

Typical factors judged in the process of regulation will include: •

Clinical effectiveness



Clinical safety



The outcomes of clinical audit



How risk is managed



How quality is assured



Organisational and staff development



Patient satisfaction



Resource efficiency

This is a costly and data hungry process, but there is much evidence that it produces continual improvement in health care outcomes67. 2. Professional revalidation/re-licensing (i.e. To individual professionals in their roles as professionals). In many countries, there is a requirement that clinical practitioners expose themselves to some form of relicensing/revalidation at regular intervals. The process is usually applied to establishing clinical competence, and to evidence that competence is continually developing. Many mature health care systems have criteria against which a clinician's competence to practice is judged. Usually these are applied to an individual clinician's status as a clinician as opposed to their status as an employee, although these, of course are related. Usually, the judgment requires that a clinician produces convincing evidence that they are safe in terms of patient management, and that they are making visible effort to maintain competence. The problem is finding appropriate, verifiable and meaningful evidence. Typically, the judgments will be made against areas such as: Table 22: Typical criteria for judging clinical quality

Duties of a clinician

A.

67

Dimension

1. Delivery of good

medical care

Typical evidence •

Re admission rates



Post operative infection rates



“Excess” mortality (judged against benchmarks)



Existence of an evidence base for practices

Institute of Medicine, http://www.iom.edu/) 162

Institutional Analysis

A.

Dimension

Typical evidence

2. Maintaining

good medical practice

3. Working

relationships with colleagues

4. Relations

with

patients

5. Teaching

training

6. Probity

and



Audits



etc



Evidence of formal learning (courses, scientific meetings).



Informal learning.



Informal teaching.



etc



Cancelled theatre sessions



Cancelled outpatient clinics



Inappropriate referrals



Failure to refer when clinically indicated



etc



Complaints



Compliments



Self re-referral



Cancelled theatre sessions



Cancelled outpatient clinics



etc



Student feedback



Success rate of learners in examinations



Cross-professional training



etc



Effective use of resources (time, others’ time, equipment, money, pharmaceuticals, capital assets).



Personal integrity.



etc 163

Institutional Analysis

A.

Dimension

Typical evidence

7. Health

B.

Dimension

Duties of an employee

relevant)

Days lost to ill health



Ability to deliver to job description



Rate of advance of limiting conditions



etc

Typical evidence • Membership of committees

1. Management activity

2. Research





(if •

Role as an appraiser Publication in refereed journals



Contribution to the evidence-base

3. Service



Leadership of clinical developments

development 4. Organisational effectiveness

••

Leadership developments etc Commitmentofinservice terms of hours

• Commitment in terms of behaviour etc As a general principle, departure from what is acceptable will produce remedies which vary in severity. Departure in A1, A4, A6 and A7 will need immediate remedy, possibly as far as preventing or limiting an individual's freedom to practice. A2, A3 and A5 may require developmental remedies (retraining etc). B2 may need to be remedied if it concerns the quality of research which determines clinical practice by self or others. B1, B3 and B4 may require organisational remedies which do not limit the freedom to practice medicine. The contract will conform to a variant of one of three basic types (see:"main contract types.pdf", Annex 3) The negotiation process will be at an institutional level, and will seek to produce outcomes which are equitable, stable and fair. The marketplace (3.1 & 3.2) is usually managed in terms of tariff, co-payments and access, and there will be special arrangements for acquisition to capital. Often exposure to the regulatory framework is enshrined in legislation. For 3.3., there are few mature economies which rely on this model to secure a wide range of health care services, but almost all apply this model for some aspect care; for example, the 164

Institutional Analysis

provision of aesthetic surgery, cosmetic dentistry, some aspects of optical services (particularly the purchase of fashion spectacles), some over the counter pharmaceuticals etc. Such markets are rarely efficient, mainly because the purchaser is an unsophisticated purchaser, lacking in good quality information. The usual mechanisms to secure efficiency are: Typically, currencies will be some combination of: •

Inputs (activity, beds, operations, staff hours, XRay units, Path tests etc)



Outputs (discharges, waiting times, adverse incidents, vaccination coverage etc)



Outcomes (clinical or population outcomes - prevalence/incidence, QALYs, DALYs etc., secured through identifying units of activity related to DRG/HRG/Integrated Care Pathways etc)

Some systems require conformity to clinical protocols and frameworks arising from best practice analysis at a national level (e.g. NIHCE in the UK, NIH in the U.S etc). Currencies: Whichever regulatory model is used, there remains the problem of what currency is being used to make judgments of quality and efficiency in the regulatory framework.

165

Institutional Analysis

Annex 4: Health care as a market (inc. market failure and gaming) Alternative models of health care provision In general, international judgments (e.g. WHO, World Bank etc) about health care are based on eight characteristics: •

Price consciousness: Can individuals tell whether they are getting good value for money?



Social solidarity: Does the system enable the poor to enjoy a high standard of care?



Consumer empowerment and patient satisfaction: Does the method of financing put the consumer in a weak or strong position? For instance, pre-payment may make it harder to escape bad service in the short run, whereas paying an insurer who will then pay any doctor of the customer’s choice is empowering.



Quality of care: What incentives do providers have to offer a high standard of care?



Clinical autonomy: Does the system impact on the professional duty of clinicians to act in the best interest of patients and does it encourage ‘best practice’?



Conflicts of interest with the third party payer: Any system will partly involve the pooling of risk by a third party payer, but whose interests are served by this third party? Is it a consumer-dominated organisation, or is it producer, government, or unduly shareholder dominated?



Responsiveness: Has the system proved able over time to bring the expectations of individuals into balance with the capacity of providers to treat patients?



Fiscal viability: Is the burden on public finances affordable?

The WHO has identified 11 types of system which are distinguished by applying a judgment of those characteristics: •

General taxation (e.g. the UK).



Local taxation combined with provider management by local councils (e.g. Denmark).



Social insurance paid by the employer and the employee with multiple non-competing autonomous insurers (e.g. France).



Social health insurance paid by the employer and the employee, with competing insurers (e.g. Germany).



Compulsory social insurance paid by individuals with competing insurers and a governmentapproved insurance plan (e.g. Switzerland).



Voluntary insurance, with tax subsidies, paid by employers (e.g. USA)



Voluntary insurance, paid by individuals, with tax subsidies (e.g. Australia). 166

Institutional Analysis



Voluntary insurance with either no, or very limited tax subsidy; supplementary insurance (e.g. France), substitutive cover (e.g. Germany and the Netherlands).



Catastrophe insurance plus a tax-protected savings account (e.g. Singapore).



Compulsory ‘exceptional and costly’ health insurance with German-style social insurance for acute personal medical services (e.g. the Netherlands).



Health care purchasing co-operatives (e.g. USA).

Table 23: Examples of models of health care system



Universal access to health care to all residents through Medicare. Individuals eligible for Medicare receive free ambulatory medical care and free accommodation and medical, nursing and other care as public patients in state funded hospitals. Alternatively they may choose treatment as private patients in public or private hospitals, with some assistance from Medicare.



Mainly Government financed apart from a small element of co-payment.



Universal coverage and stresses preventative care. Provision is mainly private, with 80% of Japan's hospitals and 94% of its physician run clinics being privately owned.



Government acts as regulator and insurer and determines a fee schedule in consultation with providers and consumers. All doctors receive the same salary regardless of experience. It also subsidises health care spending for the elderly, small business employees, and the self-employed.



The system covers all residents.



All essential services, including hospital and out-patient, are provided free through the public health system, with the exception of dentistry and ophthalmology.



Government is a purchaser and provider of healthcare and has responsibility for legislation and general policy matters.



General taxation with minor cost sharing



Integrated with Social Services



Developed primary and community services



Developed services for the MH and LDS

Australia

Japan

New Zealand

Sweden

68

68

There are few, if any, comparable systems to compare with Indonesia, with similar scale and complexity. The other large economies (India, China, US, Brazil, Nigeria) are largely Federal Sates and so have a completely different perspective on local service delivery.

167

Institutional Analysis

France

Germany

UK

Italy

Ex-planned economies



Public sector dominated



Secondary care driven



Interesting relationship between public and private sector



Poor Health Promotion services



Management is very HQ-centric



Capital readily available



Tricky IR



Hub and spoke distribution of acute services



Secondary care driven



High commitment to specialism (specialist hospitals are common)



Effect of integrating East Germany into a mature system



Effect of new economic challenges



Private and public sectors co-exist (but largely separately)



Established mature system



Universal access, free at the point of delivery



Primary-care driven



Decentralised service planning



Politically vulnerable



Health care cost inflation outstripping budget allocations



Similar social/political history to the UK



Well-developed primary care



Effect of unstable political environment



Effect of marked geographic differences.



High levels of public involvement in public sector



Economies starting from scratch, but with western levels of expectation.



Have learned from others.



Poor levels of investment, so often innovative actions to respond to patient

168

Institutional Analysis

needs. •

Low costs.

Conditions for a well-functioning market to prevail On the supply side, under certain conditions, providers will produce their services at the lowest possible resource costs (i.e., they will be technically efficient and cost-effectively efficient). What are these conditions? First, in the market for the service in question, there needs to be many providers, each producing a similar product. Second, consumers must be aware of the quality of the product and the prices charged by each provider. Under these conditions, providers are disciplined to produce at as low a cost as possible. If a provider wastes resources in the production process, (i.e., it is technically inefficient) or uses an input mix other than the least cost combination (i.e., it is not cost effectively efficient), to cover costs the provider must then charge a price higher than competitors who are efficient. Such a provider would be unable to succeed because consumers will purchase from lower-priced competitors. The presence of many competitors and consumer awareness of quality and prices also ensures that a provider, even if producing at least cost, cannot charge a price that exceeds the cost of production (which includes the normal returns to their entrepreneurship) in an attempt to increase profit. For a market to sustain a large number of providers, two conditions must be met. There must be free entry and exit of providers in the market. That is, there must not be barriers to entry or exit, either artificially created by government regulation or naturally occurring because of the nature of the good and its production. An example of the latter is large fixed costs in production. If there are large fixed costs, then the market might be able to sustain only one or a few providers to serve customers while operating at lowest cost capacity. Third, information on production methods and input costs (such as wages) must be common knowledge. Any provider who has a technological advantage, or can purchase inputs at lower costs than others, can produce at a lower cost and hence charge lower prices than others, expanding market share and gaining market power. Finally, to produce at efficient output levels, providers must consider all of the costs of their production methods. The societal costs of production involve the consumption of any valuable and finite resource, including not only labour services and capital, but also "free" goods such as clean air and water. If all such costs are taken into account by the provider, then the cost to providers equals the cost to society. If, on the other hand, providers do not have to pay for the use of scarce resources, a production externality is generated. In this case, the costs faced by the provider are less than the real cost to society. In the case of Indonesia, it is unlikely that these conditions will be met. Not surprisingly, these conditions for efficiency are seldom strictly met in real markets, but quite large departures from them also persist in many markets, generating concern about efficiency. Some examples of large and small deviations from these conditions are as follows. 169

Institutional Analysis



Consumers are not always aware of the prices charged by each supplier and of the quality differences among providers. The reason is that gathering information is not free - it requires that the consumer spend time and energy. Hence, consumers may not be fully informed about all prices and quality, and indeed, they sometimes make the mistake of assuming that higher price means higher quality.



There are often significant barriers to entry and exit. Physicians, for example, are regulated, which restricts entry. To practice medicine, an individual must first obtain a licence to practice from a regulatory agency, such as a College of Physicians. Hence, in these markets, the number of competitors is restricted, conferring a degree of market power on those who do gain entry.



The number of efficient providers can also be limited by the physical nature of the production process for the service, such as what occurs with large fixed costs. The fixed costs associated with building major hospitals preclude more than a small number of suppliers in most countries. Similarly, most market areas cannot sustain a large number of tertiary care hospitals given the relatively small need for highly specialised health care services and the large fixed costs involved in producing them.



The health care provider whose clinical waste is absorbed into a community consumes not only materials, labour and machinery, but also a clean environment. This imposes external costs on others in the community. If the provider is not forced to compensate for the value of the environmental impact, however, its costs are lower than the societal costs.

On the demand side, allocative efficiency requires that providers supply services which are most highly valued by consumers. A key condition for this is that all consumers are able to judge accurately the value of a good or service to themselves. Indeed, it is conventional to use the value derived by consumers of goods and services as a measure of their value to society. A related requirement for efficiency is that consumers have the same information on the quality of a good or service as do providers. This ensures that neither party to a market transaction can take advantage of the other. If one party has better information about the quality or other relevant characteristics than the other, then an informational asymmetry is said to exist. In these cases, it is possible that one party use this information to exploit the other party for financial gain. Allocative efficiency also requires that one individual's consumption of a service does not impose any costs or confer any benefits to others; the consumer is the sole recipient of these costs and benefits. If this is satisfied, then the benefits gained by the consumer are equal to the benefits gained by society. If this is not satisfied, then consumption externalities are produced. Consumption externalities will cause the quantity consumed to be different than the efficient quantity consumed. Finally, no one consumer can account for a "large" proportion of consumer demand - if there were a dominant consumer, then that consumer would hold disproportionate market power and might be able to influence prices. By 2019 BPJS will be a very large consumer. As with the production side, these conditions for efficiency are not likely to be met.

170

Institutional Analysis

Consumers (including institutional consumers such as BPJS) face significant problems with respect to the quality and the value to them of health care. Indeed, informational problems that make it difficult for consumers to judge the value of such services is one reason why suppliers of them are often regulated. Consumption externalities, both positive and negative, frequently occur. When a person obtains vaccination against a communicable disease, s/he obtains a personal benefit but it also provides a benefit to everyone else in the community by reducing their risk of contracting the disease. Large consumers, especially in local markets can exert significant market power may exert their market share power to negotiate lower prices for health care through preferred provider organisations and other arrangements. A final set of conditions is that there be a "full" set of markets. That is, there must be a market for all services a consumer desires and is both willing and able to pay for at a price sufficient to sustain production. All of these conditions, on both the production and consumption side, must be met if a market is to allocate resource efficiently. Supply and demand A market is an environment where providers and consumers exchange goods and services at a market price. These exchanges lead to particular patterns of resource use: providers choose a production method (the production technology, mix of inputs used) and how much to provide given the prices of inputs used in production and the price of their output; they are responsible, however, for assuming the cost of production. Consumers individually decide how much of each type of commodity to purchase on the basis of market prices and their incomes. The price they pay for the commodities are the revenues of providers. Through exchange of goods and services, consumers and providers seek to further their own ends. Consumers might value a good or service more than the price they have to pay, hence they are better off. Providers, on the other hand, might be able to produce a service for less than the price they receive, hence they too are better off. What factors affect the amount providers wish to produce? A critical factor is the opportunity to earn a profit, which depends on the cost of producing the good and the price of the good. As the price of good increases, the revenue from producing and selling it increases, increasing profits and providing incentives for the provider to produce more. Per-unit costs of producing the good may, however, increase as production increases (e.g., the providers might need to pay workers overtime wages), especially in the short-term. A provider will find it profitable, therefore, to increase production to the point where the cost of producing one more unit (marginal cost) equals the added (marginal) revenue obtained by selling the additional unit. Providers, therefore, tend to provide more at higher prices. Price and its relation to costs of provision are the key determinants of the supply of a service. Consumer demand for a service depends critically not only on the prevailing epidemiology in a community but also on market mechanisms driven by preference. Individuals will prefer to observe capital investment in personal care rather than in health promotion. For preferences to 171

Institutional Analysis

be translated into demands, however, consumers must have the ability to pay, hence, income and wealth are important determinants of demand. Demand and supply schedules A demand schedule depicts how prices, income and preferences affect the quantity of a service demanded by consumers. The market demand schedule summarises how the total demand of all consumers react to changes in price and other factors. The two key characteristics of the demand schedule are its slope and position. The slope of the schedule reflects how quantity demanded of a service changes with the price of the service. If the schedule is very steep, then desired consumption does not change dramatically when prices change. The sensitivity of demand to changes in price is known as the price elasticity of demand. The price elasticity of demand for a service is the percentage change in quantity demanded due to a small (e.g., 1%) change in its price. If the price elasticity of demand is less than one in absolute value, demand is said to be price-inelastic. If, on the other hand, the price elasticity of demand is greater than one in absolute value, demand is said to be price-elastic. The larger (in absolute value) is this number, the greater is the responsiveness of the quantity demanded to changes in price. Efficiency of the market The net benefit to society of a commodity produced and consumed is the total value of its consumption less the cost of the resources used in its production. Under the potential Pareto criterion, efficiency is achieved when the net benefit to society of each commodity consumed is maximised. For each good that is traded, this occurs at the quantity where the additional or marginal benefit derived from consuming an additional unit is equal to the marginal costs associated with producing an additional unit. If the technical conditions outlined above are satisfied, then amounts supplied and demanded in a well-functioning market will satisfy this condition. As long as the conditions hold, providers must produce at least cost and that the costs that providers face are equal to the full social costs of production. In a well-functioning market, the market price will adjust so that quantities supplied are just equal to quantities demanded. Because there are no externalities in such a market, supply reflects demand and demand reflects. Markets and Market Failure When one or more of the conditions discussed above is not present, a market will not result in an efficient allocation of resources. This is market failure. Market failure does not mean that a market for the good or service will fail to exist, that it cannot or will not operate (though that is one possibility); rather, it means that the production and/or allocation of the service that results from the operation of the market will not be efficient. To understand why the failure of any of the conditions to hold can lead to an inefficient allocation of resources (market failure), it is useful to refer to the relation between marginal costs, marginal benefits and prices under an efficient allocation of resources in a market. A corrective intervention must be crafted to fit the problem that is generating the market failure. In the case of market failure generated by an external production cost, the solution may be a corrective tax on the provider equal to the amount of the external cost, thereby forcing the provider to consider the full social cost when deciding how much to produce. In the case of an 172

Institutional Analysis

informational problem, the response may be for government to provide the information or for it to adopt a regulatory approach that requires a seller to reveal all relevant information to a buyer. In the case of a monopoly, it may be anti-trust regulations to prevent monopolistic practices. Regulatory Government action is itself often subject to problems. It is not always possible, for example, to determine the level of a corrective tax (subsidy) that will just equal the external cost (benefit); and sometimes a regulatory agency (BPJS) created to protect the public interest is "captured" by the industry and creates policies more to the benefit of the industry than the public. Some have labelled this a problem of "government failure". The point is to ensure that, when a market failure has been identified, the intervention proposed to "fix" it will in fact improve the situation. Health care as a commodity In most countries of the world, health care resources are not allocated wholly, or even primarily, through private markets. In nearly all countries there is significant governmental involvement in the health care sector, often to ensure widespread access to health services for all citizens (UHC). Substantial government involvement in ensuring access, regulating health professionals, and in public or private health care insurance is a common feature of health care systems around the world. Apart from equity considerations, which may be important, these institutional features (and others) commonly found in the health sector can be traced to the characteristics of health care as an economic commodity that create market failure that would result in an inefficient allocation of health care resources by free markets. Amongst the most salient elements of health care provision is uncertainty. Uncertainty is an inherent characteristic of many types of illnesses and accidents -- they (and their associated costs) are unpredictable. Health care insurance cannot eradicate (or even influence) the uncertainty of illness, but it can reduce an individual's exposure to financial risks associated with illness. But there is also uncertainty regarding the effects of treatment - its effectiveness for any single individual remains uncertain. This uncertainty is an inherent aspect of health care services. These two types of unpredictability mean that health and health care decisions must be made in a context of risk. The consumer patient or purchaser) very often has neither the information nor the knowledge to judge what clinical intervention is needed and what, if anything, is required to restore health. The health care professional has this expertise and directs the patient as to what services can be expected to restore their health. Hence, there is an imbalance in the level of information, or an informational asymmetry, between the consumer and the provider. Informational asymmetries are a significant source of market failure in the health care sector because they give the providers considerable market power. Licensing, professional regulation, and professional ethics attempt to mitigate problems that can arise from informational asymmetry. They ensure minimum quality standards and encourage providers to act as agents for their patients' interests rather than exploit the informational asymmetry for their own economic advantage. The demand for health care derives from the demand for health. The demand for health care is therefore a derived demand -- it is derived from the demand for health. In many cases the direct effect of health care on well-being is actually negative. Health care is only one of many possible ways to maintain or improve health. Health care may be a vital input once an individual is ill, but non-health care factors may be far more effective than health care in maintaining the health of a 173

Institutional Analysis

population (i.e., avoiding illness). Many non-health care interventions might improve health. The demand for health care must be placed in the context of these other health-influencing activities. Second, the demand for a health care service depends critically on the expected effectiveness of the service in producing health. Finally, health care often creates external effects beyond those which accrue to the recipient of care, which result in inefficient resource allocation in private markets. There can be external costs, such as when the consumption of antibiotics leads to the development of antibioticresistant strains of bacteria. These features of health care as a commodity -- uncertainty, asymmetry of information, derived demand, and externalities -- are not individually unique to health care. These features are generally prevalent in the health care sector, and their combination makes health care "different" from other goods. These features also frequently lead to market failure and make resource allocation in health care through private, unregulated markets inefficient, inequitable, and cost-increasing. Health care markets Informational asymmetry between patients and health care professionals creates significant supply-side market power for providers. Because a provider often determines what services are needed and then provides those services, informational asymmetry creates an interdependence between the supply and demand sides of health care markets, undermining a basic condition for efficient, well-functioning markets. In health care markets, a distinction can be made between demand, need and utilisation. Health care demand refers to the amounts of various health care services a consumer desires at given money prices, incomes, etc.; health care need is based on the amount and types of services that are effective in improving health; and health care utilisation refers to the types and amounts of different services actually consumed. Although need is a primary determinant of demand, a person may demand services that are not "needed" (an antibiotic for a viral infection), a person may fail to demand a needed service (anti-hypertensive drug in the presence of high blood pressure), and the amount of health care actually utilised may differ from that demanded and/or needed by a patient, the most important reason of which is provider influence. Because the demand for health care derives from a demand for health, a primary determinant of the demand for health care is ill-health. The demand for a health care service also depends on the total cost an individual faces for the service (including the price of the service to the consumer, travel costs, lost income while receiving the service, and non-money costs such as the time. Although insurance (public or private) often lowers the financial price a consumer pays for a service, other costs can be a significant factor in the decision to seek health care. Once an individual visits a health care provider, these demand-side factors interact with supplyside influence to determine the care actually utilised. Acting as agent for the patient, the provider assesses and recommends what care is needed to restore health. In this sense, the provider plays an active, indeed often a leading, role in treatment decisions, "inducing" patients to consume certain services. When done in the interest of patients, this inducement is beneficial. Providers, however, do not always act as perfect agents for their patients. Their recommendations are sometimes influenced by self-interest, or the interest of the organisation for which they work. Though the potential for such self-interested supplier-induced demand is 174

Institutional Analysis

widely acknowledged given the informational asymmetry between patient and provider, its policy importance remains an on-going area of controversy. Proponents of the view that inducement is quantitatively important in affecting behaviour and therefore an important policy issue argue that the market power conferred on providers by informational asymmetries must be considered when predicting the effects of many types of policy actions. For example, it could be argued that an analysis of the effects on utilisation of introducing co-payments for services must take into account possible supply-side responses to the policy. That is, the initial effect of the price increase to individuals is just as would be predicted by a normal demand analysis -- the quantity of physician services demanded by individuals’ falls. But if the drop in the quantity of services demanded causes provider incomes to fall, they may respond by exerting their market power, derived from informational asymmetry between providers and patients, by increasing the demand for their services (e.g., shift the demand curve out). The phenomenon of supplier-induced demand has a direct bearing on the validity of quasimarket mechanisms in challenging the proposal for a national network of publicly funded primary care centres to meet health care needs in rural areas. The predicted effect of a policy of increasing the supply of providers will be incorrect, however, if providers can induce demand for their services, allowing providers to maintain a practice even after the market has become "saturated" with providers, a phenomenon that generally has been observed in the health care systems of many countries. Informational asymmetries also may invalidate the assumption of "consumer knows best" which underlies evaluative policy assessment in health care. In the presence of informational asymmetries, the health care utilisation observed reflects both patient and provider preferences, rather than just patient preferences. It may also reflect considerable consumer ignorance. Conclusion: An approach to appraising the health systems is via a more qualitative assessment of the three core dimensions of universal coverage (WHO, 2008), namely its depth (what proportion of the population is covered by some form of insurance or pre-payment?), its breadth (the range of services or interventions that are available to members of the insured pool), and also its height (the proportion of total costs covered by pre-payment). A summary for 9 countries is given.Error! Reference source not found. shows that efficient health care systems, measured in terms of life expectancy related to per capita total health expenditure do not depend on a market mechanism, but have significant, skilled government intervention to manage and regulate the markets operating. The institutions involved in the specification and provision of health care services in Indonesia are ill-equipped to mitigate the causes and effects of market failure (information asymmetry, ignorance, service are differentiated, resource immobility, provider market power, inadequate provision (services would or could not be provided in sufficient quantity by the market in sparsely populated areas), the existence of external costs and benefits and the presence of inequalities). The main arguments for and against market mechanisms in health care include: For •

Encourages rational use of services 175

Institutional Analysis



Helps raise revenue



Improves quality of services

Against •

Low-income groups more price sensitive than higher income groups



Quality improvements may not be realised and poor may be deterred from using services



Often customers ≠ users

The implication of all of this is that the quasi-market mechanism to be introduced in Indonesia is unlikely to deliver equity and efficiency across the country without significant government intervention to regulate the impact of market failure. All institutions on the demand side (MoH, BPHS, PHOs and DHOs) will need to develop systems, processes and competences to recognise and influence the impact of market weaknesses inherent in the health care context. Figure 15: Correlation between health expenditures and outcomes

69

69

Chisholm et al, op.cit

176

Table 24: Dimensions of universal coverage in 'high-performing' countries

70

70

Chisholm, D et al (2010), Improving health system efficiency as a means of moving towards universal coverage, World Health Report (2010) Background Paper, No 28

Institutional Analysis

Annex 5: Determining the strength of individual Puskesmas Table 25: Illustration of how Puskesmas may be judged

Area of interest

Aim

Examples

Safety

Patient safety is Protect patients and staff through systems that enhanced by the use of health care processes, a) identify and learn from all patient safety incidents and other reportable incidents, and make improvements in practice based on local and national experience and information derived from working practices and the analysis of incidents; and systemic activities that prevent or reduce the b) ensure that patient safety notices, alerts and other communications concerning patient safety risk of harm to patients. which require action are acted upon within required time-scales. c) all risks associated with the acquisition and use of medical devices are minimised; d) ensure that all reusable medical devices are properly decontaminated prior to use and that the risks associated with decontamination facilities and processes are well managed; e) ensure that medicines are handled safely and securely

Clinical and Cost Patients achieve health Effectiveness care benefits that meet their individual needs through health care decisions and services based on what assessed research evidence has shown provides effective clinical outcomes Governance

Ensure that a) Care is delivered by healthcare professionals who make clinical decisions based on evidencebased practice. b) clinical care and treatment are carried out under supervision and leadership; c) clinicians continuously update skills and techniques relevant to their clinical work; and d) clinicians participate in regular clinical audit and reviews of clinical services.

Managerial and clinical a) apply the principles of sound clinical and corporate governance; leadership and 178

Institutional Analysis

Area of interest

Aim

Examples

accountability, as well as the organisation’s culture, systems and working practices ensure that probity, quality assurance, quality improvement and patient safety are central components of all the activities of the health care organisation.

b) actively support all employees to promote openness, honesty, probity, accountability, and the economic, efficient and effective use of resources; c) undertake systematic risk assessment and risk management; d) ensure financial management achieves economy, effectiveness, efficiency, probity and accountability in the use of resources; e) challenge discrimination, promote equality and respect human rights; and f)

meet the existing performance requirements set out in the SPM

g) have a systematic and planned approach to the management of records to ensure that, from the moment of creation until ultimate disposal, that it serves its purpose and is disposed of appropriately when no longer required. h) undertake all appropriate employment checks and ensure that all employed or contracted professionally qualified staff are registered with the appropriate bodies; and i)

User and Health care is provided partnership focus in partnership with patients, their carers and relatives, respecting their diverse needs, preferences and choices, and in partnership with other organisations (especially

require that all employed professionals abide by relevant published codes of professional practice.

systems exist to ensure that a) staff treat patients, their relatives and carers with dignity and respect; b) appropriate consent is obtained when required for all contacts with patients and for the use of any patient confidential information. c) staff treat patient information confidentially, except where authorised by legislation to the contrary. d) Patients have access to suitable and accessible information about, and clear access to, 179

Institutional Analysis

Area of interest

Aim

Examples

social care procedures to register formal complaints and feedback on the quality of services and; organisations) whose services impact on e) are not discriminated against when complaints are made. patient well-being. f) information is made available to patients and the public on their services, provide patients with suitable and accessible information on the care and treatment they receive and, where appropriate, inform patients on what to expect during treatment, care and after-care. g) Patients and service users, particularly those with long-term conditions, are helped to contribute to planning of their care and are provided with opportunities and resources to develop competence in self-care. h) Community empowerment is evidenced by support for UKBM. Accessible and Patients receive services a) The views of patients, their carers and others are sought and taken into account in designing, Responsive Care as promptly as possible, planning, delivering and improving health care services. have choice in access to services and treatments, b) Health care organisations enable all members of the population to access services equally and offer choice in access to services and treatment equitably. and do not experience unnecessary delay at c) Health care organisations ensure that patients with emergency health needs are able to access any stage of service care promptly and within nationally agreed timescales, and all patients are able to access delivery or of the care services within national expectations on access to services. pathway. d) Health care organisations plan and deliver health care which reflects the views and health needs of the population served and which is based on evidence or best practice; e) maximises patient choice; f)

uses agreed guidance, guidelines or protocols for admission, referral and discharge that accord with the latest national expectations on access to services. 180

Institutional Analysis

Area of interest

Aim

Examples

Care Environment Care is provided in and Amenities environments that promote patient and staff well-being and respect for patients’ needs and preferences in that they are designed for the effective and safe delivery of treatment, care or a specific function, provide as much privacy as possible, are well maintained and are cleaned to optimise health outcomes for patients.

services are provided in environments which

Public Health

promote, protect and demonstrably improve the health of the community served, and narrow health inequalities by

Programmes and services are designed and delivered in collaboration with all relevant organisations and communities to promote, protect and improve the health of the population served and reduce health inequalities between

promote effective care and optimise health outcomes by being a) a safe and secure environment which protects patients, staff, visitors and their property, and the physical assets of the organisation; b) supportive of patient privacy and confidentiality. c) well designed and well maintained with cleanliness levels in clinical and nonclinical areas that meet the national specifications.

a) co-operating with each other and with local authorities and other organisations; b) ensuring that national policy informs local policies and practices c) making an appropriate and effective contribution to local partnership arrangements d) have systematic and managed disease prevention and health promotion programmes which meet the requirements of national policy

181

Institutional Analysis

Area of interest

Aim

Examples

different population e) identify and act upon significant public health problems and health inequality issues groups and areas. f) implement effective public health programmes to improve health and reduce health inequalities, conforming to best practice g) protect communities from identified current and new hazards to health h) take into account current and emerging policies and knowledge on public health issues in the development of their public health programmes, health promotion and prevention services

182

Annex 6: Conceptual model used to describe institutional relationships Figure 16: Analytical model

Table 26: Definitions of the model components

External: Societal

Technological

Expectations of the community

Professional standards

Risk management

Developing technologies

Economic

Political

Fiscal space

Policy and legislation

Budgets and cost control

The economy

Internal: Formal organisational arrangements

Tasks and transactions

financial systems

the jobs to be done

management

professional standards

geographical distribution of services

the distribution of work

management systems

expenditure management

corporate

income management

effectiveness/governance

business planning

performance indicators

direct service delivery

Institutional Analysis

etc.

etc.

People, capacity and skills

Informal culture

skill mix

quality of information given and shared

rôles

territorial issues

quality of disclosure and exchange

risk tolerance/aversion

staffing levels

etc.

etc

184

Annex 7: Summary of actions These actions are listed in the order in which they appear in the report, and refer to the relevant sections in which they appear. There are some instances of repetition because they appear in several places in the analysis. Table 27: Summary of actions

Action

This will require substantial capacity development at all levels of the SKN to ensure that the practice of business planning and contracting as a means of doing business and managing complex relationships is embedded in systems, capability and culture. The boundary and role anomalies in the distribution of hospitals need to be resolved by re-assigning the stewardship of hospitals to their respective tiers of governance. WHO recommended Indonesia to continue establishing networks in collaboration with other countries. (Communicable Disease Control). Additionally, the country should have a 5-year plan and budget completed with an annual work plan. (Communicable Disease Control). It is a priority to devise mechanisms to enable MoH and MoE to influence the providers of training in both public and private sector settings equally to ensure that that graduates are produced in the right numbers, with the highest degree of competence and deployed in the most needy parts of the country. It would be beneficial to agree a definition of what constitutes “management” expenditure so that it can be measured and managed in the future. The MoH should surrender its role as direct provider of personal services to an alternative agency. It would be a step forward to appoint a Chief Medical Officer, a Chief Nursing Officer, a Chief Pharmaceutical Officer and a Chief Scientific Officer whose roles would be around the development of their professions rather than as Ministry functionaries A desirable model would be that the various lines of accountability (horizontal, vertical and bi-lateral) are represented by service level agreements, the contents reflecting the roles of the institutions involved. What is required is a “light touch” mechanism for MoH to be able to exercise its legitimate roles through the end-point deliverers in a decentralised

Institutional Analysis

Action framework, based on agreement rather than diktat. Improve targeting for poor and near-poor, as more than half of Jamkesmas beneficiaries were not from these groups. Improve socialisation (social marketing) to increase public awareness. Ensure supply-side availability and readiness. The experience of Jamkesmas highlights the significant disconnect between entitlements on paper versus what the system will deliver. Ensure sustainability through improving efficiency and effectiveness of implementation. Jamkesmas was entirely financed through central government taxes, and premiums were not based on sound actuarial calculations. Supply-side constraints and supply-side subsidies gave the false impression that financing of Jamkesmas was sufficient, when the reality was that the programme did not reimburse the full cost of services and relied heavily on supply-side subsidies. Make provider payment mechanisms more results-focused. Under Jamkesmas, payments to providers were basically fee-for-service (including diagnosis-related groups for hospital-based care). At present, there are no additional incentives to improve quality and provider performance. Providers are not given incentives to achieve targets. The BPJS regime does not resolve this issue. Establish a robust and reliable information system to support monitoring and evaluation, and continuously update the calculation of the programme’s costs. Learn lessons from selected provinces and districts that have attained virtual universal health coverage, such as Bali, Aceh, and Jakarta. The Government can learn from these regions’ experiences and estimate costs from existing samples. A concern is the quality of coding at the clinical level. The DRG-based classification operates a very high level of aggregation - for secondary care, the locally-derived DRG deploys 800 categories of discrimination (compared with 12,000+ in ICD 11) creating opportunity for contestability and frequent opportunities for the “ad hoc” management of outliers. It is reported that the coding is inconsistent across the country, and that the training of coding clerks is often poorly conducted. This will need to be rectified. There is an urgent need to unify and align health information systems across the sector. The contracting mechanism needs to be developed and adjusted to recognise issues of gaming, quality assurance and policy alignment and to 186

Institutional Analysis

Action mitigate their effects. In terms of health care practice, there is evidence of inefficient operating practices, including, for example, a very high dependency on in-patient care, a very low level of use of 24 hour – surgery, high drug costs and a reimbursement regime which is likely to aggravate this. There are three mechanisms to help manage this; first a contracting framework which encourages and rewards efficient clinical practices, and second a set of credible and agreed clinical guidelines for the management of the major causes of morbidity and finally unifying the various drug lists into a common approved formulary would help to manage drug costs. In terms of operating efficiencies, assuming a more rigorous contracting process is introduced, then Provincial Health Offices should be required to produce and submit business plans every 3 years, leading to annual operational plans. In planning terms, these would become the currency of supervision, in the same way that contract performance is the currency of management. This will require support in terms of developing capacity/skills in business planning (Figure 10). The “ad hoc” movement of money (both revenue support and capital) is conducted on a “grace and favour” basis, which has the effect of distorting priorities, encourages fungibility at District level and consumes energy and effort. The annual budget setting process is a negotiation process at echelon 2 level. The linkage between budgets and policies, priorities and plans is at best opaque, and at worst, weak. The issues could be regularised by formalising the process in a consistent and transparent way, by making allocations to Provinces/Districts (which are irregular or “ad hoc”), and internally within the MoH subject to a robust business case making process. The revitalisation of Puskesmas as part of the Health Systems Strengthening based on Primary Health Care Reform is a necessity focusing on institutional, management and provision of health resources. Whilst it is difficult to determine management overhead precisely, the established figure of 26% representing “administration” is, by any measure, very high. Even if much of the 26% is inappropriately defined as “administration”, then, a management overhead in health care systems, internationally, of 10% would be considered high, and is more typically 7/8%.Efforts should be made to reduce this administrative burden; the “natural wastage” process will take a very long time to have a substantial impact on management overhead. It is suggested that providing incentives for MSS/SPM implementation for local governments would improve implementation. Therefore the central government should provide monetary incentives that induce competition. Moreover, it should publish MSS/SPM achievement data in order to increase voter’s knowledge about local governments’ performance. 187

Institutional Analysis

Action The finalisation and implementation of this framework for new appointees should be accelerated, and the same framework applied to existing postholders to guide personal development needs. The management and leadership role of Provincial Government could be developed. It has the potential to be the pivotal instrument of downwards supervision and leadership and upwards reporting. The preparation of a competence framework, jointly specified by MoHA (management and governance) and MoH (technical) is underway. It is not clear whether this framework will be applied only to new recruits, or whether it will also be applied to existing post-holders so that personal development programmes can be agreed and implemented. The latter option is to be preferred. Formal performance partnerships rather than the exercise of authority or patronage are a mechanism to reconcile the tension inherent between the principles of decentralisation and need to deliver services consistent with national priorities and policies Across all areas of the landscape is the pressing issue of creating an effective performance management process If proposals to require employers in the formal sector to contribute to health insurance are introduced, then this may have the effect of encouraging employers to “casualise” their workforces, possibly by using zero-hours contracts. The practical effect of this would be that it would add to the size of the informal economy, would mean that the burden of health care costs would regressively fall more heavily on those in greatest need, and would increase the risk of exposure to catastrophic health expenditure. Government has recognised this risk and so has reserved the right to alter the balance of personal/corporate/government contributions to the risk pool. Government will need to be vigilant to ensure that this balance is managed Since the Law will be followed by the Government Regulation as the replacement of PP 38/2007, the roles of each tier in health promotion should be clarified to avoid confusion and overlap. The clarification remains a “work in progress” at MoH. There is a need to clearly distinguish, between tiers of government, respective roles in health promotion and to establish SPM standards which encourage appropriate levels of expenditure. The MoH should design a comprehensive policy and strategy for health promotion, based on their responsibility for the provision of public merit services There needs to be a social marketing mechanism to both recruit citizens into HCIS and inform those eligible of their eligibility status and expected benefits. 188

Institutional Analysis

Action A significant anomaly arises from the conditions in which a public sector political/policy entity (MoH, Province or District) is also a provider of services under contract to BPJS. At MoH level this should be resolved by removing the management role of the MoH over Class A hospitals and delegating the management role to an executive agency. The practical proposition is to create an internal firewall within Provincial and District Health Offices which will explicitly distinguish the supervisory, legislative and regulatory functions from the management functions (Class B, C and D hospitals and Puskesmas). In terms of direct “health giving” activity, the MoH will need to develop its capacity and its energy in public health functions such as: 5. Health surveillance 6. Health status surveillance 7. Water, sanitation and hygiene 8. Epidemic surveillance 9. Health promotion/behaviour change communication (road safety, healthy life style promotion etc) 10. The provision of public “merit” goods In terms of indirect health system oversight and development, the MoH will need to focus on; 11. Regulatory systems. 12. Governance standards. 13. Planning and policy making. 14. Harmonising medical and technical education production and curriculum development with health system needs. 15. Nationwide licensing regulation (esp. for health professionals).

189

Institutional Analysis

Action 16. Advocacy (with BPJS, MoF, MoHA, Parliament, MPW, Provincial HOs, District HOs, BAPPENAS etc). •

HMIS

In terms of system leadership, the MoH should assume responsibility for the development of; •

Clinical protocols/care pathways.



Health Technology Assessment.



Emergency/ disaster/epidemic management.



The development of clinical standards.



The drug regulatory framework (pharmaco-vigilance, DRA, product licensing, QC, GMP, GDP etc).



(Possibly) management of the central purchasing of clinical materials and commodities).



Supervision and co-ordination of donor relationships (e.g. GAVI, GFATM initiatives).

Finally, BPJS needs to be accountable and responsive to the community it serves and which finances it (a known problem of single payers in other countries), either through political leadership and engagement, or through an operating relationship with central and local governments The organisational relationship between the BPJS and MoH could be captured in a number of ways; at a strategic and governance level through ensuring that the MoH is well represented at Board – level within the BPJS, that the MoH is scrutineer for BPJS business plans, and that the operational relationship is captured in a Service Level Agreement (Figure 10) The relationship should also include a degree of influence by MoH about which providers (public and private) could/should be admitted to an approved list of providers based on some judgement of quality (accreditation, for example). If the licensing role of local health offices were standardised and strengthened and supported by a robust quality assurance framework, then the licensing (and possibly accrediting role) of Provinces could be deployed by BPJS in the appointment of providers in the UHC scheme 190

Institutional Analysis

Action If health experience and health care are considered to be significant elements of human development, then what needs to be recognised is the necessity to consider the wider network of inter-connected factors which deliver health benefit or health hazard. Some Governments (admittedly none known to be of the scale and complexity of Indonesia) have engineered planning mechanisms which are designed to reflect and capitalise on the interconnectivity. The Governments of Kazakhstan and of Wales, for example, have designed cross-sector planning mechanisms which require Ministries to identify ways in which their activities can contribute to the achievement of the objectives of other Ministries, and to commit to jointly planning and investing in such activities. This could be a model adopted by BAPPENAS to promote “joined-up” national development. As the market develops, then it will be necessary to enable and encourage non-government provision by neutralising competitive disadvantage experienced through differential criteria for admission to providers’ lists and through minimising the effect of the economic burden of capital financing. There are three ways to achieve this. One is to give tax benefits to private sector to compensate for the cost of capital acquisition. The other way is to apply a capital charging overhead on the public sector, reflecting the public dividend capital invested in public assets. The third is a differential tariff to reflect the lower costs experienced by the public sector. Private sector representatives should be included in the negotiations. A risk assessment needs to be undertaken, and contingency plans developed to confront the predictable conditions in which absorptive capacity (fiscal space, skills, information systems, inability to recruit Jamkesda etc) jeopoardises the implementation of UHC. This requires that the Minister of Home Affairs decree stipulates that in the pursuit of operational efficiency and effectiveness that the head of the health centre can be held by a functional health worker, rather than uniquely by a structural, bureaucratic occupant. Health activity required of Puskesmas should be grouped into mandatory actions such as health promotion, maternal and child health, communicable disease and NCD prevention, nutrition, environmental health, emergency and essential health services, and discretionary/desirable health actions accounting for locally- specific health problems in the area (and accounting for the availability of resources). Hence, the existing regulatory frameworks71 need to be applied more rigorously to assure the shifting of services away from curative care, but Puskesmas need to be strengthened to encourage preventive and promotive activities, i.e. towards encouraging behavioural change through all the

71

Ministerial Decree 118/2004, and the strategic plan 1994 are designed to strengthen and support promotive and preventive care.

191

Institutional Analysis

Action Community Based Health Actions such as Posyandu, Desa Siaga, Pos Obat Desa, Pos Malaria, etc. The contract with BPJS offers the mechanism to implement this objective It should be a local government responsibility to calculate the quantity, and skill mix health of the workforce required to equip Puskesmas with the human assets required to deliver its duties. The status of the head of Puskesmas should be recalibrated not as a structural position but as a functional position. BPJS needs to engineer a contract portfolio which is both manageable and which enables the proper supervision of the performance of many thousands of providers through a blend of contract types (fee-for-service, block and cost & volume) reflecting local specificities, and to have relationships with local institutions (e.g. DHO’s) to whom contract conformity monitoring and reporting can be delegated. Contracts should include a series of quality standards associated with access, clinical quality, efficiency, reporting requirements and coding accuracy. It may be that more sophisticated targets for BPJS are required, which combine, consolidation, absorption of existing schemes and the recruitment of citizens operating in the informal sector. The least unacceptable solution would be to consolidate schemes upwards, but for Government to invest a proportion of the planned growth in health care expenditure to subsidise the risk pool until such time as they are naturally equalised. The solution, avoiding large scale, short term capital investment from the public purse, would be to encourage the independent sector to contract with BPJS. The obvious solution has been proposed earlier (see 0); that is, that the MoH exercises its supervisory and guidance roles by focusing on desired outcomes – conformity to standards etc – through being the audience for business plans created at a local level, which then needs to be negotiated and agreed. The outcome of the agreement is then captured in a Service Level Agreement which identifies costs, quality, access and reporting requirements. Community involvement in the preparation of the plan would be a requirement. The business plan would represent Provincial concerns, and would be informed by District Business Plans. This solution protects the interests of the principles of decentralisation because it represents an agreement made between the Province and the 192

Institutional Analysis

Action Centre; not accession to a greater authority. It involves the community because it is transparent. It would be an evidence-informed plan, and the MoH has the means to exercise its responsibilities through the SLA. Referral systems should be strengthened, and driven by a protocol/pathway determined by the MoH exercising it’s leadership and guidance role. Standard Operating Procedures/Protocols for required primary care activities should be developed. Health financing needs to be more predictable, and needs to be defined as the responsibility and authority of both government and society. Although health is the 'right' of communities, the community should be encouraged to participate in the financing of health (both personal and public health). Whilst co-payment is an established principle for personal services, it is less so for public merit services. A mechanism to secure shared responsibility for the health status of communities will be to ensure that the planning and budgeting is governed by the technocratic, district government and community through the community empowerment mechanisms and using SMD and MMD as the instruments of decision making. The community should be encouraged to participate in the financing of health (both personal and public health). There is a need to reform the regulation to provide more flexibility to manage the budget to meet Puskesmas needs. This means substantial skills and capacity development at District-level to support the management and governance of primary care, both directly through the business managers, and through District-wide activities (health promotion, community engagement etc). The relationship between Districts and Provinces is not dissimilar to the relationship between the relationship between Provinces and Districts, and the solution is similar; that the relationship between Districts and Provinces is captured and formalised through a Business Plan – SLA arrangement. This has the added advantage of creating opportunity for the Community Empowerment mechanisms to operate at a service planning and delivery level. Local government should include achievement of the SPM in the five year development plan and implemented under the annual plan and budget. The MoH should monitor and supervise the achievement of the SPM. A series of suggestions about engaging with the non-Government sector. “On the other side of the equation, the non-government sector must be prepared to expose itself to conformity to quality standards (including clinical quality), to sharing activity and reach data and to sharing clinical information at an aggregate level if they wish to benefit from “cash cow” publically 193

Institutional Analysis

Action subsidised income streams. In terms of private sector educators (see 6.6), a similar proposition exists – that in return for exposure to supervision and regulation of quality conformity, and the regulation of supply, the private sector will get access to public clinical environments for clinical placements.” Align reporting/information flows with accountability and financing flows. All institutions on the demand side (MoH, BPHS, PHOs and DHOs) will need to develop systems, processes and competences to recognise and influence the impact of market weaknesses inherent in the health care context. Alter the BPJS share of allocation to include semi-variable cost elements in the tariff values as well as the planned variable costs. Engineer contract or service-based agreements between local Governments and Provinces, and between Provinces and the MoH. The agreements would capture the intentions set out in the business plans, would require that the agreements are two-way (funds one way; performance the other), and that they would capture agreements around quality, access, equity and efficiency (Figure 10). Extend the concept of matched funding to create incentives for Districts to conform to central priorities, and to provide an instrument for MoH to encourage greater equity. Require Provincial Health Agencies and Local Government Health Agencies to produce locally-specific costed business plans which reflect both RPJMN priorities and local specificities, so that funding can be aligned with health gain objectives, and in which allocations, releases and budgets are transparent (as would connections/disconnections between them). Re-visit the structure and roles of units within the MoH, Provinces and Districts to ensure that no conflicts of interest exist (e.g. between regulatory functions and management functions). An alternative strategy would be to consider creating a regulatory authority which would operate vertically and horizontally throughout the system. This would require that the authority has the instruments and leverage to actively manage the outcome of its regulatory role, and may require legislative intervention to create the entity empowered to intervene. Whilst in theory a regulatory body exists (Badan Pengawas RS) and should be operated at provincial level, it is not compulsory. At national level is only a coordinating mechanism. However BPRS has little power of enforcement and operates more as an arbitration instrument. The District and Provincial Health Offices have the power of hospital licensing and recommendation, but there are structural role conflicts and anomalies present in these structures and so enforcement is very weak.

194

Suggest Documents