Health Registries and Quality Geir Joner

Institute of Health Management and Health Economics

Health Registries - 1 • Regulated by law – Cancer Registry of Norway

• Regulated by “forskrift” – The Medical Birth Registry of Norway

• Etablished through individual regulations – A licence has been granted by application to the The Data Inspectorate of Norway (“konsesjon”) – A well defined but limited purpose, limited in time and volume – Typical for medical quality registries (about 50)

Health Registries - 2 • • • • •

Disease registry Epidemiological registry Research registry Quality registry Mixed….

Health Registries - 3 • Aims – Epidemiological: • Insidence/prevalence of disease(s) • Time trends, monitoring • Burden of disease • Typical: Cancer Registry – Research • Risk factors /causes of disease • Complications • Typical: Medical Birth Registry

Health Registries - 4 • Aims - continued – Economic/financial: • Norwegian Patient Registry • Financial basis for hospitals: 40-60% based on production of DRG points – Quality of care • Monitoring quality – Follow-up/prospective/cross-sectional – Direct monitoring of actual treatment or care

Quality Registries - 1 • Definition: – registry with aim to etablish direct monitoring of medical treatment (process or outcome) – most quality registries limited to one disease (eg.diabetes or AMI) or one medical procedure (treatment of hip fractures)

• National or regional – 2005: 50 registries • 36 national (nationwide) • 14 regional • 6-8 are funded by Regional Health Authorities from 2004

Quality Registries - 2 • What is typical? – Started by ”pioneers” – Doctors working within a network or within a professional organizaton – No long-term funding

Benchmarking •

•

General definition: Benchmarking (also "best practice benchmarking" or "process benchmarking") is a process used in management and particularly strategic management, in which organizations evaluate various aspects of their processes in relation to best practice, usually within their own sector. This then allows organizations to develop plans on how to adopt such best practice, usually with the aim of increasing some aspect of performance. Benchmarking may be a one-off event, but is often treated as a continuous process in which organizations continually seek to challenge their practices. Competitive benchmarking is used in competitor analysis. When researching your direct competitors you also research the best company in the industry.

• The Norwegian Arthroplasty Register – The Norwegian Cruciate Ligament Register – The Norwegian Hip Fracture Register

• http://www.haukeland.no/nrl/

Mortality within 30 days after hospitalization as quality indicator in Norwegian hospitals

Norwegian Knowledge Centre for the Health Services Institute of Health Management and Health Economics

20

Hip fracture

15

10

Percent dead within 30 days

Mortality 1997 til 2001 AMI, Stroke and Hip fracture index 25

Myocardial infarction Stroke

5

NOV 01 SEP 01 JUL 01 MAY 01 MAR 01 JAN 01 NOV 00 SEP 00 JUL 00 MAY 00 MAR 00 JAN 00 NOV 99 SEP 99 JUL 99 MAY 99 MAR 99 JAN 99 NOV 98 SEP 98 JUL 98 MAY 98 MAR 98 JAN 98 NOV 97 SEP 97 JUL 97 MAY 97 MAR 97 JAN 97

Month & year of admission

30D – Acute Myocardial Infarction 40

Shrinkage, risk adjusted Risk adjusted Unadjusted case fatality rates (1997-2001) Indifference limits

30

Density

Alert limits

20

10

0

0.00

0.05

0.10

0.15

0.20

0.25

Probability of death by hospital - AMI

0.30

0.35

30D – Stroke 12 Shrinkage, risk adjusted Risk adjusted Unadjusted case fatality rates (1997-2001) Indifference limits

10

Density

8

Alert limits

6

4

2

0

0.00

0.05

0.10

0.15

0.20

0.25

Probability of death by hospital - stroke

0.30

0.35

30D – Hip fracture 25

Shrinkage, risk adjusted

20

Risk adjusted Unadjusted case fatality rates (1997-2001) Indifference limits

Density

Alert limits 15

10

5

0

0.00

0.05

0.10

Probability of death by hospital - hip fracture

0.15

Mortality 30D after hospitalization

•

AMI 15,6% - 19,0% ; median 16,6% – One hospital have significant higher 30D than mean – 11 hospitals: internal evaluation

•

Stroke 11,5% - 23,6% ; median 15,9% – 6 hospitals have significant higher 30D than mean – 8 hospitals: internal evaluation

•

Hip fracture 5,4%-13,3%; median 8,2% – 4 hospitals have significant higher 30D than mean – 12 hospitals: internal evaluation

Prel. results • •

Small differences for AMI Huge differences for stroke and hip fracture – 246% for hip fracture – 205% for stroke

• •

Reliable results? No final conclusion

Results should be interpreted with caution • Old data • Poor data quality (diagnoses and procedures) – – – – –

Lack of valid diagnosis (PAS problems) Wrong diagnosis Priority of diagnoses not correct To few diagnoses (leads to underreporting of co-morbidity) Lack of procedures (surgery)

Further plan – 30D • New dataset (2000-2005) • Evaluation of diagnosis quality – Read 2000 patient records in 15 hospitals – Collect data about symptoms, clinical findings, lab etc – Score leading to some classification • Eg. AMI: 1) AMI 2) probably AMI 3) no AMI

• Adjust for age, gender, comorbidity etc • Is death 30D suitable as quality indicator

We need data sets with unique patient identification • Norwegian Knowledge Centre for the Health Services has large national data sets which may be employed in quality assessment • Additional information by linkage to other registries (Statistics Norway) – Death or alive – Causes of death – Socioeconomic status, ethnicity etc.

Norsk Pasientregister with identifiable patients (NPR) from mid 2007 • NPR give us new possibilities– patients may be followed through several episodes of hospitalization and from one hospital to another • Course of the disease: complications and death compared between hospitals and • Linkage to other registries • Valuable data when employed by Quality registries

• Local initiatives – http://www.helse-vest.no/sw6034.asp

Proportion of patients with yearly eye check according to international guidelines – by hospital Andel som har gjennomgått øyekontroll

Andel av pas i avd (%)

100 90 80 70 60 50

2004 2005

40 30 20

1

10 0 A

B

C

D

E

F

Sykehus

DEMO

G

H

I

J

Mean HbA1c (”blood glucose”) in children by hospital and year Gj.snitt HbA1 pr sykehus 9,2 9 HbA1c (%)

8,8 8,6 2004

8,4

2005

8,2 1

8 7,8 7,6 A

B

C

D

E

F

Sykehus

DEMO

G

H

I

J