Cochrane Health-related Quality of Life Methods Group (HRQoL MG): Toward Registration

t ea n lin om ! o le .c ilab h-inst a v r a arc ette i-rese l s New .map QoL www NEWS N U M B E R I LETTER 26 JANUARY-JUNE 2001 NTRODUCING Coc...
Author: Aubrey Bryan
3 downloads 2 Views 689KB Size
t ea n lin om ! o le .c ilab h-inst a v r a arc ette i-rese l s New .map QoL www

NEWS N U M B E R

I

LETTER

26

JANUARY-JUNE

2001

NTRODUCING

Cochrane Health-related Quality of Life Methods Group (HRQoL MG): Toward Registration Véronique Staniek, PhD, Regulatory Department, MAPI Research Institute, Lyon, France HRQoL MG Members

Convenors: Bernard Jambon, Dick Joyce, Donald Patrick Co-ordinator: Véronique Staniek Members: Ruthy Acosta, UK (Eyes and Vision Group) Xavier Badia, Spain Dorcas Beaton, Canada Adrian Bottomley, Belgium Olivier Chassany, France Desmond Curran, Ireland (Statistical MG) Helen Dayus, UK Paul Dieppe, UK Pennifer Erickson, USA Peter Fayers, UK Steven Fullerton, USA Morton Glanz, (Statistical MG) Sally Green, Australia (Musculoskeletal Group) Gordon Guyatt, Canada (Cochrane Screening and Diagnosis Tests Methods Group) Asha Hareendran, UK Sonja Hunt, USA Amid Ismail, USA (Oral Health Group) Veronica Kennedy (Ear, Nose & Throat Disorders Group) Madeleine King, Australia Leslie Lenert, USA Elaine McColl, UK Patrick Marquis, France Keith A. Meadows, UK Richmal Oates-Whitehead, New Zealand (New Zealand Cochrane Center) Grace Paterson, Canada Donald L. Patrick, USA Danny Ruta, UK Jeff Sloan, USA Marianne Sullivan, Sweden Sergei Varshavsky (Russian Cochrane Center) Mike Wheeler, UK

What is the Cochrane Collaboration? The Cochrane Collaboration, launched in 1993, is an international organisation that aims at helping people to take well-informed decisions about health care (‘evidencebased medicine’) by preparing, maintaining and promoting the accessibility of systematic reviews (or meta-analysis) of the effects of healthcare interventions. About fifty Collaborative Review Groups (CRGs) prepare and maintain Cochrane Reviews. People working in Centres, Methods Group, Fields and the Consumer Network support the CRGs. Methods Groups are established to provide policy advice, to develop methodology and provide training and support to the Collaboration. For more information, please visit the website of the Collaboration: www.cochrane.co.uk

I N Introducing 1-2

Cochrane Health-related Quality of Life Methods Group (HRQoL MG): Toward Registration by Véronique Staniek

Report 2

9

Assessing Health-related Quality of Life – What can the Cognitive Sciences Contributes? An Advanced Seminar by KA Meadows Education in Health-related Quality of Life by Marissa Vicari

24-25 A Decisive Step Towards the

Recognition of Patient Reported Outcomes in Clinical Trials by Catherine Acquadro

Discussion 3-4

Towards a More Restricted Use of the Term ‘Quality of Life’ by Erik Nord et al.

T H I S

A Cochrane HRQoL Methods Group: Why? For most CRGs, HRQoL is a stated outcome of interest in systematic reviews. But difficulties, such as lack of HRQoL data, pooling of data, validity of the scales and interpretation, are encountered by CRGs when including HRQoL data in reviews.

The objective of the Cochrane HRQoL MG is to advise Cochrane reviewers about when and how to incorporate HRQoL data into systematic reviews. The HRQoL MG will refine methods for searching HRQoL studies in the literature, will develop methods for systematically reviewing HRQoL studies and refine methods for meta-analysis of HRQoL data.

(continued on p 2)

I S S U E

Work in Progress 5

A Study to Develop and Evaluate a Patient-Centred, Stroke-Specific Quality of Life Measure by Deborah Buck et al.

6-7

Assessing Population Utilities for New Health Care Interventions by Amid.I Ismail et al.

15-16 The Three Time Dimensions Synoptic

Scale (3TSS) for Depressive Population by Nicole de Leval

Results 13-14 State Trait Anxiety and Fear of Labour

in First and Subsequent Pregnancy by Eleonora Bielawska et al. 17-18 General Population Norms for Romania

8

Assessing Quality of Life in Primary Biliary Cirrhosis by Anne Rannard et al.

Instruments 10-11 Health-related Quality of Life and

Obesity by Marianne Sullivan et al. 12-13 A Comparison between the Hebrew

Versions of the Quality of Life Enjoyment and Satisfaction Questionnaire and the Lancashire Quality of Life Profile in Severely Mentally Ill Patients by M. Ritsner et al.

using the Short Form 36 Health Survey (SF-36) by Valentina Mihaila et al. 20-21 Quality of Life and Lower Urinary Tract

Dysfunction by Dudley Robinson et al. 22-23 Differences in Health-related Quality of

Life among Stroke Subtypes by Maree L. Hackett et al.

Publications 4-6-8-18-19 Website 19 Meetings 26-27-28 Key Word Index 28

QOL

2

QoL Newsletter, 2001; 26

I

NTRODUCING

R

EPORT UK

Cochrane Health-related Quality of Life Methods Group (HRQoL MG): Toward Registration

Assessing Health-related Quality of Life What can the Cognitive Sciences Contributes?

(continued from p 1)

KA Meadows, Head, Health and Survey Research Unit, Applied Statistics Center, University of Hull, England

From creation to registration The creation of the HRQoL MG was initiated during an exploratory meeting held in Lyon on July 6th, 1998 attended by members of the French Cochrane Centre and researchers interested in HRQoL issues. In order to contribute to the work of the CRGs in the most relevant way, the HRQoL MG has conducted a survey in order to estimate the CRGs’ potential needs regarding HRQoL data. In addition, an analysis of the systematic reviews produced by the CRGs has been conducted to examine how HRQoL outcomes are currently included in the reviews. These results helped us to prepare a research agenda according to the difficulties and the needs of the CRGs, and working groups have been established to work on specific topics: Terminology, Instrument validity and Statistics issues. About thirty members (see the list on page 1) have now joined the HRQoL MG and will work on a voluntary basis to contribute to the Cochrane Collaboration. The group will first work with these members, before additional people could join the methods group, if needed. The final registration is expected for the first quarter of 2001. MAPI Research Institute, a non-profit organisation whose objectives are to facilitate and advance the science and practice of HRQoL evaluation and to promote and support collaborative research initiatives, supports this initiative. Additional sponsors aware of the importance of the evidence-based medicine and interested in supporting the HRQoL Methods Group (travel expenses, meetings of members) will be welcome. • For more information, please contact Véronique Staniek, PhD MAPI Research Institute, 27, rue de la Villette, Lyon 69003, France. Tel: +33 (0) 4 72 13 66 67 Fax: + 33 (0) 4 72 13 66 82 E-mail: [email protected]

An Advanced Seminar

uch of the work on the design and development of health-related quality of life (HRQoL) measures over the past decade or so has focused on establishing their statistical and psychometric validity and reliability. In contrast however, variation and error in the response to questionnaire items arising from ambiguous wording and the demands placed on the respondents’ cognitive processes to provide valid answers is rarely or explicitly addressed. While interest on the cognitive aspects of respondent behavior and the design of questionnaires (generally known as Cognitive Aspects of Survey Methodology CASM) is a relatively new area of multidisciplinary research, its importance to survey research in general is rapidly being acknowledged. Nevertheless, its application to HRQoL research remains limited. So as to facilitate the strengthening of links between CASM and HRQoL research and assessment, twenty-two invited participants from the USA, Scandinavia, Holland, the UK and the Republic of Ireland with experience in HRQoL research and or CASM attended in early December a two-day advanced seminar titled ‘Assessing Health-related Quality of Life - What can the Cognitive Sciences Contribute?’ The seminar, which was convened and hosted by the Health and Survey Research Unit, Applied Statistics Centre, University of Hull in collaboration with Dr Ivan Barofsky, comprised six plenary session papers, a panel discussion on the feasibility of applying CASM to HRQoL research and three workshops in which participants were asked to identify a number of potential areas of research. The six plenary session papers were: • Cognitive psychology and self-report: methods and materials, Jared Jobe • Pretesting survey instruments: an overview of cognitive methods and Techniques, Debbie Collins • Implications of response shift for cognitive research, Carolyn Schwartz • Diversity and HRQoL assessment: cognitive equivalence and summary measures, Ivan Barofsky

M

• Cognitive models and modern psychometric model incompatible frameworks? Jakob Bjorner • A theory based instrument for evaluation of individual QoL: The SEIQoL, Hannah McGee A number of priority research areas were identified during the meeting with an overall agreement that CASM could and should be applied both to the understanding of the cognitive process and testing of existing and new measures in the assessment of HRQoL. The meeting highlighted the need for more information about some of the basic issues of cognitive research as well as a common nomenclature across the different disciplines. All participants agreed to maintain contact and to investigate the feasibility of carrying out coordinated research in a number of areas. Arising from the seminar, activities during 2001 include the publication of the seminar proceedings and monograph. Development of an infrastructure to facilitate communication between interested researchers and aid collaborative research will also be a priority over the next 12 months. It is also intended that the HSRU website (www.hull.ac.uk/hsru/) will act initially as a means of disseminating information on relevant areas of interest and activities resulting from the seminar. A second advanced seminar is planned for 2002. It is hoped that all these activities will provide a basis for developing a more formal forum for interested researchers to exchange ideas on methodological issues and collaboration with respect to CASM related QoL research. • Acknowledgements We would like to acknowledge Johnson & Johnson, NovoNordisk A/S and the University of Hull, for their financial support. For more information please contact, Dr Keith Meadows, Head, Health and Survey Research Unit, Applied Statistics Centre, University of Hull, Cottingham Road, Hull. HU7 6RX England. Tel: +44 1482 305125 E-mail: [email protected].

QoL Newsletter, 2001; 26

D

NEWS

QOL

LETTER

3

ISCUSSION

Towards a More Restricted Use of the Term ‘Quality of Life’ Erik Nord1, Trude Arnesen1, Paul Menzel2, Jose-Luis Pinto3 National Institute of Public Health, Oslo - 2Pacific Lutheran University, Tacoma, Washington - 3Pompeu-Fabra University, Barcelona.

1

Imagine that you had a chronic illness or disability and somebody asked you to fill in a questionnaire and subsequently claimed that they had ‘measured’ the quality of the life you were living with this condition. How would you feel about that? We are not sure that you would have liked it. We believe you could feel that the measurers were beyond their capacity. You might also feel that they were trespassing, as you feel when someone reads your personal mail or enters your house without being invited. In our experience in clinical and health services research, it is difficult for an investigator to give a talk about the effects of illness and disabilities on quality of life without a number of people in the audience feeling uncomfortable or getting upset. Objections have been voiced in the literature both by patients’ organizations and researchers themselves. Dahl 1 notes that reports about patients’ ‘quality of life’ leave the patients with feelings of being devalued. Gill and Feinstein2 point out that "...unlike beauty, which rests in the eye of the beholder, quality of life is inherently an attribute of the patient (or "beholdee")". Sonja Hunt, creator of the Nottingham Health Profile, expressed many of these concerns in a talk at the ISOQOL meeting in Barcelona in November 1999. The widespread use of the term ‘quality of life’ in clinical research and health economics is understandable in the light of history. Until the last couple of decades, assessment of medical outcomes was strongly focused on gains in survival time, i.e. quantity of life. However, any kind of outcome of productive activity has two main dimensions: quantity and quality. An increasing number of people were concerned that medical outcome assessment was neglecting the quality dimension in two important ways.

First, health improvements that lead to better rather than longer lives for chronically ill or disabled people are a major outcome of health care, and there was no satisfactory account of this type of outcome. Second, advances in medical technology increasingly made it possible to prolong the lives of people with fatal illness, but well-being in those gained years also needed to be taken into account. So interest in the ‘quality side’, became in, and what was then more natural than to speak of ‘quality of life measurements’? But the use of the term has gone much too far. To most people, ‘quality of life’ refers to how good, desirable and enjoyable life as a whole is felt by the person in question2. Only a small portion of all items in clinical instruments that are said to measure quality of life refer to this subjective, overall feeling. Very many refer to objective symptoms and functional status (capabilities). Those that do refer to subjective feelings often refer to short term mood, like mood on the day of measurement or in the week or two preceding it. Our short term mood does not necessarily reflect our overall view of the desirability of the life we are living, viewed as a long sequence of years with up and downs, strengths and weaknesses, hopes and challenges, frustrations and rewards. Researchers claiming to ‘measure quality of life’ may of course defend the terminology on the grounds that it is all a matter of definition. But when researchers present their results to the general public as if the public should accept those findings, their definitions have to be in keeping with common sense. Otherwise there is bound to be misunderstanding and unnecessary tensions. We believe that data about the objective conditions under which people live, be they environmental or related to their own bodies, are important for policy making. But they should not be referred to as quality of life data. People can live in primitive houses, wear worn out clothes, not afford to go to the movies, have a limp, miss an arm, see poorly

and have angina. That does not mean that they necessarily lack quality of life. Given the complexity and diversity of human nature, we believe that the term ‘quality of life data’ should be used carefully and only in connection with data about people’s subjective feelings about life. Furthermore, there is a need to distinguish between the terms ‘measures’ and ‘indicators’. We suggest that in most people’s ears the former is a more ambitious term than the latter. The difference corresponds to the difference between ‘hard’ and ‘circumstantial’ evidence in judicial language. Responses to questionnaires like the Sickness Impact Profile, the Nottingham Health Profile, SF-36, the Quality of Well-Being Scale, the EuroQol, the Health Utilities Index, the 15-D etc. may yield helpful indications of the impact of different illnesses and treatments on people’s well-being. But to the external observer they can only be indications. They are not, and can never be, measures of the nature and strength of the actual, complex overall feeling of well-being and satisfaction with life inside the patients concerned. That inside, overall feeling can only be communicated in non-standardised language chosen by the patients themselves, and perhaps even then not fully. Some would argue that techniques like the standard gamble or the time tradeoff do measure quality of life in the large sense, inasmuch as subjects are asked how much they are willing to sacrifice of life itself in order to be relieved of illness. When considering that willingness to sacrifice, subjects must of course take into account the overall quality of their life. However, responses to standard gamble and time trade-off questions are of a hypothetical nature and may therefore very well be biased. Willingness to sacrifice lifetime may also be severely restricted by feelings about life as in some sense ‘holy’ or untouchable – i.e. feelings that transcend the concept of quality of life3. (continued on p 4)

QOL

4

QoL Newsletter, 2001; 26

D

ISCUSSION

P U B L I C AT I O N

Towards a More Restricted Use of the Term ‘Quality of Life’ (continued from p 3)

Neither standard gamble nor time tradeoff responses should therefore be understood as ‘measures’ of quality of life. To accommodate some of the critique above, the term ‘health related quality of life’ is often used as distinct from ‘overall quality of life’. But the expression is awkward. One can speak of specific functional limitations – be they mental, physical or social – as being caused by illness, and thus as being health related. One can also speak of losses of quality of life as being health related. In fact, this is precisely what techniques like the standard gamble and the time trade-off purport to indicate. But one cannot turn the issue around and separate from the holistic sensation of quality of life one part which can meaningfully be referred to as ‘health related’. ‘Quality of life’ is a handy term, the misuse of which is not easy to avoid even if one accepts the above criticisms in principle. But perhaps it is time to make a stronger effort. Questionnaires like the SF-36 and the EuroQol, and techniques like the standard gamble and the time trade-off, might in future be referred to as ‘indicators of health status and well-being’ rather than ‘measures of (health related) quality of life’. We believe this change of terminology would be in keeping with common sense perceptions of the complex and inherently subjective character of the phenomenon ‘quality of life’. • Erik Nord, National Institute of Public Health, Geitmyrsveien 75, 0462 Oslo, Norway. Tel: +47 220 42 200 – Fax: +47 220 42 595 – E-mail: [email protected]

SF-8 Health Survey Manual his 220-page manual documents the new SF-8 Health Survey, a generic multipurpose short-form (SF) survey of health status published by QualityMetric, Inc. (www.qmetric.com). It is a major advance in the application of SF technology for purposes of achieving both brevity and comprehensiveness. It has only eight questionnaire items. By relying on a single item to measure each of its eight domains, the SF-8 represents the ultimate phase in the evolution of SF assessments - single-item scales. It is also the first SF survey to be constructed on the basis of empirical studies linking each item to a comprehensive "pool" of widely used questionnaire items proven to measure the same health concept. Because that pool included the SF-36 Health Survey, each SF-8 single-item scale and its summary measures can be scored on the same metric as the SF-36 scales and summary measures. Thus, the SF-8 provides the long awaited "missing link" - a practical tool for directly linking the norms from large population surveys with the results from more focused outcomes studies and monitoring efforts in everyday clinical practice. SF-8 has been translated for use in more than 30 countries and languages. Finally, SF-8 is playing a major role in the evolution of computerized dynamic health assessments that match questions to respondents' health levels. SF-8 items are the first questions asked by QualityMetric's Dynamic Health Assessment (DynHA) "engine", which uses computerized testing logic to achieve brief and accurate assessments of generic health status. Additionally, this manual also completely documents the SF-6.

T

To order the SF-8 manual, please visit www.qmetric.com/marketplace/index.cgi. The manual price is $89.95. Postal mail: QualityMetric Inc. 640 George Washington Highway, Lincoln, RI 02865 USA. Tel: +1 (401) 334-8800

P U B L I C AT I O N

STUDIES ON HEALTH-RELATED QUALITY OF LIFE IN PATIENTS WITH CHRONIC OBSTRUCTIVE PULMONARY DISEASE

Carl-Peter Engström

Studies on Health-Related Quality of Life in Patients with Chronic Obstructive Pulmonary Disease Carl-Peter Engström

This book describes Health-Related Quality of Life (HRQoL) in Chronic Obstructive Pulmonary Disease (COPD), to develop a Swedish version of the St George’s Respiratory Questionnaire (SGRQ), to study factors related to HRQoL in COPD and to study long-term effects of rehabilitation on exercise tolerance and HRQoL. Göteborg 2000

Five major conclusions were drawn:

1. Dahl G. QALY. A system that ultimately assigns different values to life. Journal of the Norwegian Association of the Disabled 1992;1:26 (text in Norwegian).

1. All aspects of HRQoL are affected by consistent determinants of the full COPD when pulmonary function is range of HRQoL measures; reduced by 50% or more; 4. Impacts of COPD can be detected 2. Physical functioning is most severely also in Swedish patients by the affected; validated version of the SGRQ;

2. Gill TM, Feinstein AR. A critical appraisal of the quality of Quality-of-Life Measurements. Journal of the American Medical Association 1994;272:619626.

3. Dyspnoea-related restriction, exercise 5. A rehabilitation programme for COPD tolerance (walking distance) and level patients has positive long-term effects of depressed mood (HAD) are on exercise tolerance.

3. Fowler FJ, Cleary PD, Massagli MP, Weissman J, Epstein A et al. The role of reluctance to give up life in the measurement of the value of health states. Medical Decision Making 1995;15: 195-200.

Carl-Peter Engström, Department of Pulmonary Medicine and Health Care Research Unit, Department of Internal Medicine, Sahlgrenska University Hospital, SE-413 45 Göteborg, Sweden.

NEWS

QoL Newsletter, 2001; 26

W

QOL

LETTER

5

ORK IN PROGRESS UK

A Study to Develop and Evaluate a Patient-Centred, Stroke-Specific Quality of Life Measure Deborah Buck, BA (Hons)1, Ann Jacoby, PhD1, Gary Ford, FRCP2 1

Department of Primary Care, University of Liverpool, Liverpool, UK Stroke Research Group, Institute for the Health of the Elderly, Newcastle upon Tyne, Newcastle upon Tyne, UK

2

here is much need for strokespecific, patient-based outcome measures so that the impact of neuroprotective drugs and other therapeutic interventions on quality of life (QoL) can be fully assessed. This requires an acceptable, relevant, valid and reliable stroke-specific measure. A recent review of currently available stroke-specific measures indicated that they lacked comprehensiveness, with little evidence of validity or reliability1. We have recently developed a strokespecific QoL measure, containing 56 items, which is being evaluated for its psychometric properties. In developing the measure we adopted a patient-centred approach. In-depth qualitative interviews were conducted with 28 stroke patients in order to ascertain the impact of stroke on QoL. The key themes that emerged were: mobility, hand function, pain, self-care, activities of daily living, sleep, fatigue, energy, incontinence, vision, hearing, cognitive problems, communication, weepiness, interpersonal relationships, sex life, religious/personal beliefs, occupation, driving, social life, and positive and negative feelings. These themes were subsequently used to inform the development of the interviewer-administered QoL instrument. Initially, a pool of 140 items was generated. These were pre-tested, with a further 30 stroke patients, using cognitive interviewing techniques in order to assess acceptability, understandability, relevance and comprehensiveness. Respondents were asked to report if any items were difficult to understand or confusing, whether they thought any could be worded better, and whether any were unacceptable. The cognitive interviewing phase proved invaluable, revealing a number of problems with wording of items and response categories available. This led to significant revision of the instrument: items that were originally in the form of

T

statements were converted into questions; some positively worded items were modified to be negative in direction; the wording of some items was amended to reduce ambiguity; and response options were revised and reduced to 4 for each item. In order to maximise the acceptability and feasibility of the measure for practical use, an item reduction phase was undertaken. The purpose was to reduce the number of items as far as possible, and to minimise the burden to patients, without compromising the measure’s comprehensiveness, validity or reliability. The revised version was therefore administered to a further 100 patients in order to identify poorly performing items (a non-response of >5%, or failure to discriminate between respondents). Factor analysis was performed with remaining items in order to identify likely domains. Where appropriate, domains were then further reduced (i.e. with the proviso that there should be no less than 5 items in the domain, and that the internal consistency (Cronbach’s alpha) of the domain should be at least .85).

Additional cognitive interviews were undertaken to pre-test the shortened version. The final measure, containing 56 items, comprises 11 domains, each with a high level of internal consistency (Cronbach’s α range .72 to .90) (see Table below). The measure was shown to be relevant and acceptable to stroke patients. It is currently being evaluated for test-retest reliability and validity with a further 100 stroke patients. We envisage that this stroke-specific QoL measure will be available for use in a wide range of health services research settings including clinical trials of neuroprotective and thrombolytic drugs, and rehabilitation and other therapeutic studies. • For more information, please contact Ms D. Buck, Research Associate, Department of Primary Care, University of Liverpool, Whelan Building, Brownlow Hill, Liverpool L69 3GB. Tel: + 44 (0) 151 794 5607 Fax: + 44 (0) 151 794 5604 - E-mail: [email protected] 1. Buck D, Jacoby A, Massey A and Ford G. Evaluation of measures used to assess quality of life after stroke. Stroke 2000; 31: 2004-2010.

Internal consistency of domains in the stroke-specific QoL measure Domain

Alpha co-efficient

No. of items

Feelings

.90

6

ADL and self-care

.90

8

Cognitive

.89

5

Mobility

.84

9

Emotional

.76

4

Sleep

.79

6

Interpersonal relationships

.84

6

Communication

.79

4

Pain/sensation

.79

3

Vision

.76

2

Fatigue

.72

3

QOL

6

QoL Newsletter, 2001; 26

W

ORK IN PROGRESS USA

Assessing Population Utilities For New Health Care Interventions Amid I. Ismail, DrPH in Oral Epidemiology, Woosung Sohn, DrPH, Oral Epidemiology School of Dentistry, University of Michigan, Ann Arbor, USA ranslation of advances in health sciences and technology requires an understanding of the societal values, educational systems, and professional environments in which the new information and new technologies will be used. In dentistry, previous experiences with adoption of new technologies and management methods, as well as new models of dental practice and education, have shown that the dental community and public at large may have unique and different perspectives that should be considered by the research community. New diagnostic technologies may challenge our current paradigms in research, health care, and education. Early detection of dental, oral, craniofacial and systemic diseases and conditions may force us to consider a number of issues related to ethics,

T

effectiveness, and preferences of patients to outcomes. Adoption of new technologies will be influenced by the public’s perception of the risks and quality of life associated with the outcomes of each new medical technology. It would also be influenced by their willingness to accept uncertainty; a factor associated with all health interventions. The revolutionary growth in communication in the later part of the 20th century presents only the beginning of a long road towards the universal access to information. The major ‘commodity’ that we deal with today and in the future in business, education, and industry will be ‘information’. As public access to health information increases, the demand for new innovations and critiquing of professional decisions will increase. Shared decision making with

P U B L I C AT I O N

Cardiovascular Disease and Hypertension Population-Based Studies on Self-rated Health and Health-related Quality of Life in Sweden Carola Bardage, MSc, PhD

he aim with this thesis was to study cardiovascular disease and hypertension, use of drugs and health from an epidemiological perspective. Various methods – Self-Rated Health (SRH), Health Related Quality of Life (HRQoL) – the 36-item short form questionnaire (SF-36) – and health utility measurements – the Rating Scale (RS) and the Time-Trade Off (TTO) methods – were employed. Data from the Swedish Adoption/Twin Study of Aging (SATSA) in 1984, 1987, 1990 and 1993 as well as a general population survey conducted in Uppsala County in 1995 were used. This thesis revealed that persons with cardiovascular disease and/or with hypertension experience poorer health than others in the population. The poor health may be caused both by the disease and/or the drug treatment. The results in this thesis also suggested that special attention and care should be directed to persons with cardiovascular disease and/or hypertension reporting ill health. This especially is important given that low HRQoL can be a risk factor for subsequent cardiovascular events or complications which in turn might result in higher mortality rate. Carola Bardage, Department of Pharmacy, Pharmaceutical Services Research, Faculty of Pharmacy, Uppsala University, Box 586, SE-751 23 Uppsala, Sweden. Tel: +46 18 471 4949 – Fax: +46 18 471 4154 E-mail: [email protected] Cardiovascular Disease and Hypertension

Population-Based Studies on Self-Rated Health and Health-Related Quality of Life in Sweden BY

CAROLA BARDAGE

UPPSALA UNIVERSITY 2000

T

patients and society will be the standard of care in the 21st century. In this new environment, success in translating new innovations into practice will require development of methods for measuring preferences of individuals and society for the outcomes that may result from the application of new technologies. This field of research has not yet been well developed in the health care sciences1. Preference for a health outcome incorporates several dimensions such as probability of risks and benefits, quality of life, time, cost, and the ease of delivery of an intervention. ‘Preference’ in the quality of life literature is used as an umbrella term referring to methods that measure values and utilities2. Values are preferences measured under certainty (100% sure that the desired outcome will be achieved). Utilities, on the other hand, are preferences measured under conditions of uncertainty where the probability of success is incorporated in the preference measurement scenarios or questions. The standard gamble theory is the method most recommended to estimate utilities1. This NIH-funded research project at the University of Michigan plans to test methods and measure population preferences of current and new treatment options of ‘toothache’. These same methods, after validation, can assess the preferences of populations to new technologies and whether the public would prefer the outcomes that are generated by using new advances in health sciences. Information on preferences for outcomes of new health technologies or knowledge would assist researchers in planning for clinical trials to test the success of new technologies using real-life data on what the public accepts as a measure of success. The Michigan study is testing the feasibility, reliability and construct validity of methods of four methods of measuring preferences of treatment outcomes: Visual Rating Scale VRS 1, Healthy Year Equivalent HYE 3, Quality Adjusted Life Years QALY 4 and Willingness to Pay WTP5. (continued on p 7)

NEWS

QoL Newsletter, 2001; 26

W

QOL

LETTER

7

ORK IN PROGRESS Assessing Population Utilities For New Health Care Interventions (continued from p 6)

The methods are being used to measure the preferences for the following treatment options of a tooth that ‘hurts’ (these options are stated using nontechnical terms): 1. Filling the tooth that hurts, which is considered the standard treatment in this study. The alternative treatments are: 2. Immediate extraction of the tooth, leaving the respondent with no pain from this tooth for the rest of his or her life. 3. Root canal therapy that involves removing the nerve of the tooth and filling the cavity, leaving the respondent with a fully functional tooth with a silver-coloured filling and without pain or discomfort for the rest of his or her life. 4. A new technology, dentin regeneration, which involves removal of the damaged part of the nerve and placement of a material inside the tooth that helps the nerve and some of the tooth to grow again. This leaves the respondent with a fully functional tooth with a silver filling and without pain or discomfort for the rest of his or her life. Over a period of one year, we have developed and tested scenarios, audiovisual presentations, and an interviewing program (MS Access®) to describe each treatment scenarios using text, sound, and photographs. After presenting the treatment scenarios, the interviewers provide each respondent with a set of outcome cards depicting each treatment described for a tooth that hurts. The respondent is instructed to order the card starting with the outcome that he/she prefers the most and followed by his or her second, third, fourth and least preferred outcome state. Two of the cards describe the socalled ‘anchor’ outcome state: 1) the best possible outcome (anchor 1) described as full oral health for the rest of life with no pain and 2) the worst possible outcome (anchor 2) described as severe and continuous pain resulting in extraction of the tooth on the same day it was filled (filling and extraction option). After the determining the visual (‘gut’) rating of each outcome state, a computer program administers a series

of options following the standard gamble method to determine utility score (the probability of occurrence of the worst outcome of the standard treatment [filling] which is considered equivalent to the probability of success of each alternative treatment). The utility scores for each treatment will be used to estimate the QALY. For the HYE, the computer program proceeds with estimating the life years equivalent to each utility score using the standard gamble method. Finally, the respondents are asked a series of questions to determine their willingness to pay (WTP) for the new health technology (dentin regeneration). The methods were tested in a series of pilot projects. Once perfected, a final pilot study was carried out to estimate the reliability of the interview scripts and the program. Two trained and calibrated interviewers interviewed 40 paid adult volunteers. Twenty subjects were interviewed a second time a week after. The Kappa coefficients for ranking of the anchors relative to the treatment options ranged between 0.57 and 1.0. There was no significant difference between the testretest HYEs of all treatments (Pearson correlations ranged between 0.52 and 0.84). The correlation coefficients of the primary and repeated WTP scores ranged between 0.96 and 0.98. Using the VRS, only 22 adults (55%) correctly ranked the anchors relative to the dental treatments. However, when the SG method was used the respondents were correct in ranking the anchors from 70 to 100%. The appropriate ranking of the anchors relative to the traditional treatment was 95% for both anchors, and for the new technology, ‘anchor 1’ was appropriately ranked 100% and ‘anchor 2’ 80%; and relative to the least favorable treatment (immediate tooth extraction), anchor 1 and anchor 2 were appropriately ranked 98% of the time. This pilot study found that the SG method yields higher percentage of appropriate ranking of dental outcomes. We are currently in the midst of conducting the study with a randomly selected sample of adults (18-69 years) with at least one tooth. Our sampled individuals reside in the Detroit Metropolitan area of Southeast

Michigan, USA. This area was selected because we plan to test hypotheses on the preferences of respondents with different educational, dental insurance, and income status. To reach over 600 adults and interview them at their homes, we have decided to use a two-stage process for sampling. In the first stage, trained telephone screeners using a list-assisted random digit dialing method contract residents in the area, ask for willingness to participate, and administer a screening questionnaire. An interviewer visits those who agree to participate. The interviewers using a notebook with a separate LCD monitor that is placed in front of the respondents to review the treatment protocols, assess the utilities, and administer a questionnaire. We have so far interviewed over 300 adults located in low, middle and highincome areas. Our interviewers have reported that the respondents were able to comprehend the content and respond to the standard gamble and other measures. The whole interview and assessment of utilities last around 45 minutes. • For further information, please contact, Dr Amid I. Ismail, School of Dentistry, University of Michigan, Room D2361, 1011 N. University, Ann Arbor, MI 48109-1078, USA. Tel: +1 (734) 647 9190 Fax: +1 (734) 936 1597 E-mail: [email protected] 1. Bennett KJ, Torrance GW. Measuring health state preferences and utilities: rating scale, time tradeoff, and standard gamble technique. In Spilker B, editor. Quality of life and Pharmacoeconomics in Clinical Trials (2nd Ed.). Philadelphia: LippincottRaven, 1996;253-265. 2. Brazier J, Deverill M, Green C, Harpter R, Booth A. A review of the use of health status measures in economic evaluation. NHS Health Technology Assessment 1999;3 (9):1-166. 3. Mehrez A. Gafni A. The healthy years equivalents: how to measure them using the standard gamble approach. Med Dec Making 1991;11:140-146 4. Weinstein MC, Stason WB. Foundations of cost effectiveness analysis for health and medical practices. New England J Med 1977;296:716721. 5. Donaldson C, Shackley P, Abdalla M. Using willingness to pay to value close substitutes: carrier screening for cystic fibrosis revisited. Health Econ 1997;6:145-149.

QOL

8

QoL Newsletter, 2001; 26

W

ORK IN PROGRESS UK

Assessing Quality of Life in Primary Biliary Cirrhosis Anne Rannard, BA (Hons), Deborah Buck, BA (Hons), Ann Jacoby, PhD Department of Primary Care, University of Liverpool, UK

Primary Biliary Cirrhosis (PBC) is a relatively rare condition of the liver in which there is destruction of the interlobular bile ducts through what is thought to be autoimmune activity1. The only cure at the present time is liver transplantation, although there is some evidence to show that disease progression can be slowed in certain patients by the use of ursodeoxycholic acid2,3. The disease is unrelated to alcohol consumption, and only becomes a true cirrhosis in its late stages4. PBC affects mainly females, and its greatest impact is evident in young females, in whom the disease is particularly aggressive. PBC is characterised by the presence of Antimitochondrial Antibody (AMA) which is often found on routine or coincidental blood tests5. Common symptoms of the disease are profound and persistent fatigue, pruritis, abdominal pain and dry eyes and mouth (Sjogrens’ syndrome). The effects of PBC on the individual can range from minor discomfort to almost total incapacity, relative to disease progression. The impact of PBC on the individual and on family life can therefore be far-reaching, affecting daily functioning, child care, relationships and occupations. However, little or no attention has been given to the devastating effects of the condition on quality of life, although clinical studies have focused on biological markers of the disease since the 1950’s6. There is scant reference to healthrelated quality of life (HRQoL) in the PBC literature and there are no PBC-specific QoL measures in existence to date. The aim of the present study is to develop and produce a multidimensional, psychometrically robust, HRQoL measure for PBC that can be used in both clinical and research settings. Purposive sampling techniques identified 40 patients from varying age groups who agreed to participate in depth interviews. Following analysis, the item generation and item reduction stage of

the process will be completed. Psychometric evaluation of the measure will be carried out on a different group of PBC patients, prior to a postal survey of the final measure and further subsequent psychometric analysis. It is envisaged the measure will be used for PBC patients in clinical and research settings, and will inform clinical trials of future drug therapy in this area. • Contact details: Anne Rannard BA (Hons), Research Associate, Department of Primary Care, Whelan Building,University of Liverpool, Liverpool, L69 3GB Tel: +44 (0)151 794 5611 - Fax: +44 (0)151 794 5604 – E-mail: [email protected]

1.

James OFW, Bhopal R, Howel D, et al. Primary Biliary Cirrhosis Once Rare, Now Common in the United Kingdom. Hepatology 1999;30:390-394.

2.

Portmann BC. Recurrence of primary biliary cirrhosis after transplantation. Eur J Gastroenterol Hepatol 1999;11(6):633-637

3.

Bateson M, Gedling P. Ursodeoxycholic acid therapy for primary biliary cirrhosis. A 10-year British single-centre population-based audit of efficacy and survival. Postgrad Med J 1998;74:482-485.

4.

Jones DEJ, Bassendine MF. Primary biliary cirrhosis. Journal of Internal Medicine 1997.

5.

Sherlock S, Scheuer PJ. The presentation and diagnosis of 100 patients with primary biliary cirrhosis. New England Journal of Medicine 1973;289:674-678.

6.

Cauch-Dudek K, Abbey S, Stewart DE, Heathcote EJ. Fatigue and Quality of Life in primary biliary cirrhosis. AASLD abstracts Hepatology October 1995; p108a

P U B L I C AT I O N

Quality of Life in Epilepsy Beyond seizure counts in assessment and treatment Gus A. Baker and Ann Jacoby, University of Liverpool, UK

hough clinical aspects of epilepsy such as seizure control are crucially important to its management, increasing attention is being given to wider quality of life issues. Epilepsy continues to be an often misunderstood and stigmatizing condition; for the vast majority of people whose seizures can be well controlled, the social and psychological repercussions are often of greater significance than the seizures themselves.

T

The increasing emphasis on the importance of non-clinical outcomes in the assessment of new treatments and management strategies for chronic conditions such as epilepsy has stimulated interest in methodological issues in assessing quality of life. This book reviews the recent literature on the impact of epilepsy on everyday experience and the methodological issues involved in assessing that impact. It also considers the perspectives of a range of health professionals involved in caring for people with epilepsy and how, through appropriate management, the impact on their lives can be minimized. Quality of Life in Epilepsy - Beyond seizure counts in assessment and treatment – 2000; ISBN: 90-5823-121-6 – US$58 / £38 / e64 To order, please contact Harwood Academic Publishers at [email protected]. For additional information please visit Harwood Academic Publishers’ website at http://www.gbhap.com

NEWS

QoL Newsletter, 2001; 26

R

QOL

LETTER

9

EPORT

Education in Health-related Quality of Life A Report on MAPI Research Institute Educational Programmes 2000 Marissa Vicari, MA, Project Manager, MAPI Research Institute, Lyon, France

Bringing people together for HRQoL The year 2000 marked six years of MAPI Research Institute Education Programmes in Health-related Quality of Life (HRQoL). From Florence, Italy, to Chantilly-Paris, France, to New York City, USA, over 80 participants this year received information on HRQoL, asked questions and engaged in discussion with experts in the field. In the last year of the century, MAPI Research Institute courses brought together practicing Physicians and Nurses, Research Scientists, Health Economists, Marketing Managers, Medical Affairs Managers, university Professors, Public Health Officers, Health Outcomes Researchers and others from countries as varied as Australia, Austria, Belgium, Brazil, Budapest, Canada, China, the Czech Republic, Egypt, France, Germany, Iran, Italy, Japan, Mexico, the Netherlands, Spain, Switzerland, Sweden, the UK, and the USA. Health-related Quality of Life in Respiratory Disease - Florence, Italy MAPI Research Institute and AstraZeneca continued into a second year of collaborating on educational programmes with the Health-related Quality of Life in Respiratory Disease session. Highlights of the course included Elizabeth Juniper’s MCSP, MSc, (QoL Technologies Ltd.) presentation on assessment of HRQoL in asthma, focusing on HRQoL questionnaires in asthma and the advantages and disadvantages of alternative versions for adults and in paediatrics, and her talk on interpretation of HRQoL data. This year we were also fortunate to have Professor Paul Jones (St. George’s Hospital Medical School) who spoke on assessment of HRQoL in Chronic Obstructive Pulmonary Disease (COPD), HRQoL questionnaires in COPD and HRQoL in clinical practice.

Linguistic validation of HRQoL questionnaires was addressed by Isabelle Mear, MA (MAPI Research Institute), and integration of HRQoL assessment into clinical trials by Elisabeth Ståhl, BSc (AstraZeneca, Lund). Over 60 participants convened at The Grand Hotel, a Florentine architectural monument that provided a uniquely beautiful setting for the session. MAPI Research Institute would like to give special thanks to AstraZeneca for making this event possible. Measuring Health-related Quality of Life in Clinical Trials - European and USA sessions In the fall, workshops on Measuring Health-related Quality of Life in Clinical Trials were held in Europe and the United States. These provided more one-on-one contact with the speakers through smaller group settings and practical group exercises, as well as a dinner out with the speakers. Participants were also given the opportunity to send in case scenarios, questions and points of interest in advance, for discussion during the course. The workshops began with a day on HRQoL assessment strategy development, psychometric validation of HRQoL instruments, linguistic validation of HRQoL questionnaires, interpretation issues and presenting HRQoL results. Day two followed with implementation of HRQoL into clinical trials, scoring and item reduction, analysis of HRQoL data, and modern psychometric methods. A panel discussion at the end provided additional time for questions and problem solving. Speakers included Linda Abetz, MA, Benoit Arnould, MA, and Isabelle Girod, MA (MAPI Values) as well as Isabelle Mear. Dorothy Keininger, RPh, MS (MAPI Research Institute) presented a summary of existing guidelines and HRQoL initiatives.

Each workshop hosted a special continent-specific segment on submitting HRQoL results, with Submitting to European Health Authorities presented at the European session by Olivier Chassany, MD (Hôpital Lariboisière, Paris, France), and submitting to the Food and Drug Administration discussed in New York with Bruce Crawford, MS (MAPI Values). It was a great pleasure for all to have John Ware, PhD of QualityMetric, Inc, developer of the SF-36, speak on interpretation issues and modern psychometric issues at the New York workshop, which took place in the distinguished Waldorf Astoria. For updated information regarding training sessions in the year 2001 please see the MAPI Research Institute website (www.mapi-research-inst.com) or contact Marissa Vicari, MA, Project Manager, MAPI Research Institute, 27, rue de la Villette, 69003 Lyon, France. Tel: +33 (0) 472 13 66 67 - Fax: +33 (0) 472 13 66 82 - E-mail: [email protected]

Call for Sponsors MAPI Research Institute would like to continue to provide quality educational programmes with leaders in the field of HRQoL and to make this resource available, and affordable, to a wider audience including clinicians, university faculty, students and people from developing countries. As a non-profit institution, the only way for us to continue to provide this increasingly demanded resource is with the support of sponsors. If you are interested in collaborating with MAPI Research Institute by being a training session sponsor please contact Marissa Vicari at the above address.

QOL

10

QoL Newsletter, 2001; 26

I

NSTRUMENTS

Health-related Quality Jan Karlsson, RA, Anna Rydén, RA, Charles Taft, PhD, Marianne Sullivan, PhD. Institute of Internal Medicine, Sahlgrenska University Hospital, Göteborg University

I. Presenting the SOS-QL Survey© A s obesity is considered a chronic and incurable disease, treatment outcome can only be assessed in terms of changes in the degree of overweight and its consequences, not in terms of cure-rate. The primary goals of treatment are to control concomitant diseases, symptoms and complaints, and minimize psychosocial adverse effects by reducing weight. Under these circumstances, the obvious measure of treatment efficacy is the effect it has on the patients' everyday life and well-being, i.e., healthrelated quality of life (HRQL). What do we know? An indication of the current state of HRQL in obesity research was obtained by examining the published literature1. In a nutshell, the main purposes and methods of the papers substantiate: 1) the newness of the field; 2) the scarcity of controlled studies; 3) the variety of selected instruments; and 4) the rapid growth in the number of epidemiological and clinical studies using HRQL methods. It is also notable that most clinical studies have been conducted to evaluate the effects of weight-reduction surgery, while only two or three have been carried out to assess quality of life change during non-surgical weight loss treatment. To date, only a few attempts have been made to develop and validate HRQL methods in obese populations. Further careful evaluations of instrument properties are needed in longitudinal field studies, where the contribution of specific questionnaires vis-à-vis generic ones can be clarified. This process will take several years to complete. A number of recent publications have measured health status in the obese using the generic SF-36 Health Survey. Due to its well-documented high psychometric standards and multinational applicability, the SF-36 will undoubtedly be increasingly used, with or without other condition-specific measures. Main features of the SOS (Swedish Obese Subjects) study. The SOS study is an ongoing nationwide, multi-center project which comprises a registry study and an intervention trial2. Since its start in October 1987 about 7000 severely obese persons have been accepted in the registry study. The intervention trial is a controlled clinical study designed to test if the negative effects of severe obesity on mortality, morbidity and quality of life are reduced during long-term weight reduction3. The outcomes of surgical vs. conventional weight reduction treatment will include 2000 surgical cases and their matched controls followed for 20 years. The SOS Quality of Life Survey© is a battery of study-specific and generic questionnaires designed to assess quality of life in the SOS study (Figure 1). Well-established generic HRQL measures, assumed to cover a broad range of health impacts from obesity, were supplemented by conditionspecific instruments, all suitable for large-scale mailout-mailback data collection. How HRQL improvements are evaluated and related to weight loss Is poor HRQL reversible after substantial weight loss, i.e., to levels of a group of healthy subjects? Are improvements maintained over time? In most instances the answer to both questions seems to be yes and definitely so regarding psychosocial functioning and mental well-being3,4. Whether impacts on physical functioning are permanently reversed needs to be evaluated further, particularly concerning how weight loss affects concomitant conditions. The SOS study will shed more light on this issue.

Brief Description of the SOS-QL Survey Self-assessment of eating behaviour The Three-Factor Eating Questionnaire, standard version, includes statements comprising three factors: restrained eating, disinhibition, and hunger. A short-form version has been developed within the SOS study5. Obesity-related psychosocial problems A study-specific module was created to assess the psychosocial burdens of obesity. It covers the perceived impact of obesity on selected activities known central to obese persons4. Physical & role functioning Four categories from the Sickness Impact Profile, SIP, i.e., ambulation, home management, work, and recreation/ pastimes, were selected to cover limitations in daily life activities. Respondents endorse those statements that describe a limitation related to their health. Psychosocial functioning Social interaction, the main psychosocial category from the SIP, was chosen to assess health-related dysfunction in social life; quality and quantity of social contacts within the family, among friends, and in the community. General health perception Overall health was measured by the current health scale selected from the General Health Rating Index. Mood disorders/distress The Hospital Anxiety and Depression scale was used to describe levels of psychological distress and to detect possible or probable mood disorder. Questions are summed to anxiety and depression scores with cutoffs for clinical cases. Mental well-being Mental well-being was measured by the short version of the Mood Adjective Check List. Responses are summed into pleasantness, activation and calmness dimension scores and an overall index. The Rosenberg SelfEsteem scale was added to include the psychological self-image. Overall Quality of Life Overall Quality of Life was assessed using a standard global question. Figure 1. Conceptual and measurement model of the SOS-QL The Swedish Obese Subjects Quality of Life Survey® (From Sullivan et al, International Textbook of Obesity. Ed. P. Björntorp. New York: Wiley, 2001; pp 485-510) Concepts: condition-specific and generic Condition-specific Complaints/ consequences

Generic Functional health: Physical/ mobility oriented consequences

Instruments: Obesity-related and generic Three-Factor Eating Questionnaire • Restraint eating • Disinhibition • Hunger

Obesity-related Problem scale • Obesity-related psychosocial problems

Sickness Impact Profile • Ambulation • Home management • Work • Recreation and pastimes

Social/ emotional/ cognitive consequences

Sickness Impact Profile • Social interaction

General health perceptions

General Health Rating Index • Current health

Mental health: Distress/ well-being

Hospital Anxiety and Depression scale • Depression • Anxiety

Summary of issues

Mood Adjective Check List • Overall mood score • Pleasantness • Activation • Calmness

• Key to success: need of both condition-specific and generic measures, long-term follow-up, large samples, matched controls. • Poor quality of life generally reversible if weight loss is substantial.

Rosenberg Self-Esteem scale • Self-esteem

• Obesity-specific measure most responsive to weight reduction. Overall Quality of Life

Global rating

NEWS

QoL Newsletter, 2001; 26

LETTER

of Life and Obesity The HRQL Group and Health Care Research Unit, Göteborg, Sweden. E-mail: [email protected]

II. Presenting the OBES-QoL© The OBES-QoL (obesity quality of life) test battery is the product of our experience in obesity research through nearly two decades, first and foremost the SOS study. The battery covers well-defined condition-specific and generic aspects of HRQL in accordance with basic scientific criteria and treatment efficacy. All obesity-specific modules included have high psychometric standards. It has advantages for clinical trials since longitudinal effect sizes are related to the magnitude of weight loss. Experience from the SOS study suggests three time points as necessary and sufficient for quality of life evaluation in obesity intervention studies: baseline, at 12 and at 24 months1,3. Tighter measurements have not proven cost-effective. Selective time points reduce patient and administrative burden and optimize data quality. Brief Description of the OBES-QoL Generic instruments - SF-36 Health Survey, v 2.0. The SF-36 is a widespread, generic short-form instrument, which comprises eight core domains of healthrelated quality of life. The new and improved version 2.0 of the SF-36 primarily increases the precision and reliability of the two role functioning scales (Role-physical and Role-emotional6) and will likely improve the ability of the instrument to detect treatment effects in clinical trials. The Swedish SF-36 v 2.0 is validated and population norms are available (Taft et al, unpublished data). - Hospital Anxiety and Depression Scale (HAD). The HAD instrument extends and refines the measurement of mental health problems in obese subjects1,2. HAD is a short and effective measure of anxiety and depression symptoms. The instrument was designed to detect mood disorders in somatically ill patients and two scale score cutoffs are available for identification of possible and probable clinical cases of anxiety or depression. An advantage of HAD over other instruments assessing psychiatric morbidity is that it does not include somatic items, e.g., questions about appetite loss and weight change. Although such questions may be accurate indicators of depression in normal weight individuals, they are likely to confound the detection of depression in obese populations. Swedish population norms are available, as well as reference values for obese subjects. - Rosenberg Self-Esteem Scale (RSE). The RSE scale is a generic measure of global self-esteem. It measures favorable or unfavorable Figure 2. Conceptual and measurement model of the OBES-QoL The Obesity Quality of Life Survey. ©HRQL Group 2000 Concepts: condition-specific and generic

Instruments: Obesity-related and generic

Condition-specific Complaints/ consequences

Functional limitations

Obesity Eating Problems scale • Restrained eating • Uncontrolled eating • Emotional eating

Obesity Functional Health Scale • Mobility • Ambulation • Sleep and rest • Home management • Work • Recreation • Social interaction • Sex life

Generic Physical/ mobility oriented consequences

SF-36 • Physical functioning • Role functioning-physical • Bodily pain

General health perceptions

• General health • Vitality

Social/ emotional consequences

• Social functioning • Role functioning-emotional

Mental health Distress/ well-being

• Mental health Hospital Anxiety and Depression scale • Depression • Anxiety Rosenberg Self-Esteem scale • Self-esteem

Overall Quality of Life

Global ratings

Obesity Psychosocial Problems scale • Obesity-related psychosocial problems

Obesity Mental Distress scale • Intrusion • Helplessness

attitudes toward oneself. The instrument has been widely used in many different contexts and populations. SOS data as well as reference values for obese subjects are available. Obesity-specific modules - Obesity Eating Problems Scale (OE). Extreme hunger feelings and problems related to eating control are frequently reported by obese subjects, especially during the course of treatment. Eating behaviour is thus a significant component of quality of life assessment in obesity. OE is a short-form instrument, developed within the SOS study to improve the measurement of obesity-related eating problems5. The module comprises three subscales: restrained eating, uncontrolled eating and emotional eating. - Obesity-related Problem Scale (see also figure 1). Obese people encounter discrimination and prejudice in their social life and commonly report severe limitations in psychosocial functioning. Hence, psychosocial functioning is one of the most important dimensions in the assessment of quality of life in obese subjects. OP is a short and effective obesity-specific module, which was specially constructed to measure the impact of obesity on psychosocial functioning2. The module comprises questions regarding regular activities in everyday life, such as going to restaurants or community activities, trying on and buying clothes, bathing in public places. The scale has proven psychometrically valid4 and responsive to weight reduction in both surgically and conventionally treated patients3,4. Results from the SOS intervention study demonstrate a distinct doseresponse relation between weight loss and changes in OP scale scores, i.e., the more favorable weight reduction the larger the reductions of obesity-related psychosocial problems1,3. The OP scale is particularly suited for evaluating quality of life effects of weight reduction in clinical trials. - Obesity Mental Distress Scale (OD). This new module, dealing specifically with obesity-related mental distress, was developed within the SOS to supplement the generic measures of mental health7. The OD module comprises two psychometrically stable factors labeled intrusion and helplessness. Intrusion is a measure of the negative impact obesity has on all aspects of life. Helplessness results in mental fatigue and lack of fighting spirit. These factors partly explain, and may predict, differences in general mental health. - Obesity Functional Health Scale (OF). This module was developed to assess functional limitations related to obesity. The SIP was used in the development work with this module. SIP is a widespread, generic long-form measure of health-related limitations in everyday life and items frequently endorsed by obese subjects were extracted to form a provisional obesity-specific functional limitations module. Some items were reformulated to better fit the specific conditions faced by obese persons and the new response format. This core set of items was further supplemented with additional items of clinical relevance to form the OF scale. - Additional obesity-specific modules are currently under development. 1. Sullivan M, Karlsson J, Sjöström L et al. Why Quality-of-Life measures should be used in the treatment of patients with obesity. In: Björntorp P. (ed.). International textbook of obesity. New York: Wiley, 2001; pp 485-510. 2. Sullivan M, Karlsson J, Sjöström L et al. Swedish Obese Subjects (SOS) - an intervention study of obesity. Baseline evaluation of health and psychosocial functioning in the first 1743 subjects examined. International Journal of Obesity 1993;17:503-512. 3. Karlsson J, Sjöström L, Sullivan M. Swedish Obese Subjects (SOS) - an intervention study of obesity. Two-year follow-up of health-related quality of life (HRQL) and eating behavior after gastric surgery for severe obesity. International Journal of Obesity 1998;21:113-126. 4. Karlsson J, Sjöström L, Sullivan M. Construct validity and responsiveness of the obesity-related problem scale (OP). Quality of Life Research 1999;8:603 (abstr.). 5. Karlsson J, Persson L-O, Sjöström L et al. Psychometric properties and factor structure of the Three-Factor Eating Questionnaire (TFEQ) in obese men and women. Results from the Swedish Obese Subjects (SOS) study. International Journal of Obesity 2000;24:000-000. 6. Jenkinson C, Stewart-Brown S, Petersen S et al. Assessment of the SF-36 version 2 in the United Kingdom Journal of Epidemiology and Community Health 1999;53:46-50. 7. Rydén A, Karlsson J, Persson L-O et al. Obesity-related coping and distress and relations to treatment preference. British Journal of Clinical Psychology, in press.

QOL

11

QOL

12

QoL Newsletter, 2001; 26

I

NSTRUMENTS Israel

A Comparison between the Hebrew Versions of the Quality of Life Enjoyment and Satisfaction Questionnaire and the Lancashire Quality of Life Profile in Severely Mentally Ill Patients M. Ritsner1,2, I. Modai1,2, R. Kurs1, H. Kostizky1, A. Ponizovsky1 Sha’ar Menashe Mental Health Center, Hadera, Israel - 2Psychiatry Department, Bruce Rappaport Faculty of Medicine, Technion, Haifa, Israel

1

Purpose Among numerous quality of life (QoL) studies of patients with severe mental disorders, we found only two that performed direct comparisons of different QoL measures within the same patient populations. Lehman and associates 1 compared convergent validity of the Quality of Life Interview (QoLI) and the Quality of Life Scale 2 among patients with severe mental illness. Psychometric properties of the Lancashire Quality of Life Profile (LQOLP) 3 , and the Life Experiences Checklist (LEC) were compared between psychiatric patients and healthy subjects. Similar studies and additional statistical testing with other widely used QoL measures are necessary to facilitate the choice of appropriate research instruments. In the present study, we compared the composition and psychometric properties of both the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q)4 and the LQOLP in psychiatric inpatients.

Method The sample consisted of subjects fulfilling ICD-10 criteria for schizophrenia (81.1%, N=56), schizoaffective disorder (10.1%, N=7) or mood disorders (8.8%, N=6). Patients with associated diagnoses of organic brain disease, serious somatic pathology, drug and alcohol abuse or dependence, and those with low comprehension skills were not enrolled. All participants signed written informed consent. The patient sample was 79.7% male, with mean age of 38 years (SD=9.1). Forty-four patients (63.8%) were never married, 14 (20.3%) were married, 11 (15.9%) were divorced/separated. Twenty patients were re-tested after a 2-week interval. The Q-LES-Q is a self-report measure of subjective quality of life (life satisfaction and enjoyment) in psychiatric patients. It consists of 93 items that are grouped into ten summary scales. Responses are scored on a 1- to 5-point scale, with higher scores indicating better QoL.

In the present study, we did not examine the 10 items referred to as School/Courses scale because they were irrelevant for psychiatric inpatients. The LQOLP is a structured interview for measuring quality of life of people with severe mental illness in community settings. The LQOLP is based on Lehman’s QoLI5 and combines objective and subjective measures in several life domains. In this study, we only used those LQOLP items, which measured the subjective aspects of the life domains (27 of 105 items).

Results The Q-LES-Q and LQOLP differ in format. Out of the 15 domains covered by both instruments, there were only 5 overlapping domains, plus the Perceived Quality of Life Index. These concordant domains included Health, Work, Leisure time activities, Social relationships, and Life satisfaction (General well being). In the LQOLP, the five instrument-specific domains were Religion, Finances, Living situation, Family relations and Legal and Safety. Instrument specific domains in the Q-LES-Q are Subjective feelings, Household duties, General activities, and Satisfaction with medication. The number of items per instrument domain is considerably greater in the Q-LES-Q than in the LQOLP, with the only exception in the Life satisfaction/General well-being domain, which consists of only item in both measures. Internal consistency measured by Cronbach’s α for all Q-LES-Q domains was consistently higher than for the LQOLP dimensions (r=0.87 to 0.94 versus r=0.32 to 0.90, respectively). Similarly, Cronbach’s α coefficients for the concordant domains of both measures were all higher for the Q-LESQ than for the LQOLP. Despite the discrepancies at the domain levels, the Perceived QoL Index of both instruments demonstrated high internal consistency (r=0.93 and 0.90, respectively). To establish test-retest reliability, 20 randomly selected inpatients completed both questionnaires with a 2-week

interval. The interclass coefficients of the LQOLP dimensions were generally in a high range (r=0.63 to 0.82), except for Work, Social relations, and Legal and Safety domains, where test-retest reliability was found to be markedly lower. Similar figures of test-retest reliability were observed for the Q-LES-Q domains (r=0.63 to 0.84), except for a lower reliability of Satisfaction with medication domain (r=0.47). For the common domains of both instruments, the Q-LES-Q demonstrated higher internal consistency than the LQOLP. To test convergent validity, Pearson correlation coefficients were calculated for the common domains. Both measures were highly correlated in the Health (r=0.66) and Life satisfaction/General well-being (r=0.72) domains. At the same time, we found a low correlation between three common domains (Work, Leisure/participate and Social relationships) in both instruments (r=0.10, r=0.28 and 0.35, respectively).

Conclusion Our purpose was to clarify relative merits and disadvantages of each of these scales to facilitate the difficult task of choosing an appropriate instrument from among the many existing QoL scales. Because of substantial differences in underlying concepts, condensed versions of the instruments were compared in this study. Comparing the instruments, we found six common dimensions covered by questionnaires, five LQOLP domains and four Q-LES-Q domains that were instrument-specific. The internal consistency of both Hebrew versions of the instruments in this study closely corresponds to those reported for the original LQOLP, and for the original Q-LES-Q. Our direct comparison of the two instruments within the same psychiatric inpatient populations, shows that the Q-LES-Q provides better internal consistency than the LQOLP. The relatively greater number of items per domain in the Q-LES-Q compared with the LQOLP could be a possible explanation for this finding. (continued on p 13)

NEWS

QoL Newsletter, 2001; 26

I

QOL

LETTER

13

NSTRUMENTS A Comparison between the Hebrew Versions of the Quality of Life Enjoyment and Satisfaction Questionnaire and the Lancashire Quality of Life Profile in Severely Mentally Ill Patients (continued from p 12)

The Q-LES-Q exhibited high 2-week testretest reliability for the six concordant dimensions and moderate to high reliability for the instrument-specific domains. Compared with the Q-LES-Q, the LQOLP- test-retest reliability coefficients were generally lower for the concordant dimensions, but they were high for 3 of the LQOLP-specific dimensions. Contrary to our expectations, generally low (if any) correlations were found between the concordant domains of the two instruments. Only Life satisfaction/General well-being and Health domains highly correlated between both scales. This finding is consistent with data on the individual measures suggesting higher consistency for the summary dimensions of perceived QoL than for specific life domains. Discrepancies in the more specific dimensions could result from the differences in definition of the dimensions and varying numbers of items across dimensions. For example, the low correlation between Social

R

relationships domains of both instruments could be explained by the fact that the LQOLP includes an additional domain of Family relations that may be confused with the definition of social relationships in that measurement. The lack of correlation between the specific domains and the higher correlation for general life satisfaction supports the idea that various instruments may present consistent evaluations of general QoL, but each is unique in its specific dimensions. To conclude, this study shows that our Hebrew versions of both instruments have reliability comparable with the reliability of the original LQOLP and QLES-Q. At the domain level, we found relatively low internal consistency for some of the domains from the LQOLP (not Q-LES-Q), due to the small number of items per domain or their heterogeneity. In general, the comparable dimensions of the Q-LES-Q provide better internal consistency and test-retest reliability than those of the LQOLP.

However, concurrent use or a synthesis of the two questionnaires may provide a more comprehensive and accurate evaluation of subjective quality of life. • For further information, please contact M. Ritsner, MD, PhD Director, Open Ward & Research Unit, Sha’ar Menashe Mental Health Center, Israel. Mobile Post Hefer 38814 - Tel: +972 4 627 8750 -Fax: +972 4 627 8045 E-mail: [email protected] 1. Lehman AF, Postrado LT, Rachuba LT. Convergent validation of quality of life assessments for persons with severe mental illnesses. Quality of Life Research 1993; 2:327-333. 2. Heinrichs DW, Hanlon TE, Carpenter WT Jr. The Quality of Life Scale: an instrument for rating the schizophrenic deficit syndrome. Schizophrenia Bulletin 1984; 10:388-398. 3. Oliver J, Huxley P, Bridges K, et al. Quality of Life and Mental Health Service. London & New York: Routledge, 1996 4. Endicott J, Nee J, Harrison W, et al. Quality of Life Enjoyment and Satisfaction Questionnaire: a new measure. Psychopharmacology Bulletin 1993;29:321-326. 5. Lehman AF. The well-being of chronic mental patients: Assessing their quality of life. Archives of General Psychiatry 1983;40:369-373.

E S U LT S Poland

State Trait Anxiety and Fear of Labour in First and Subsequent Pregnancy Eleonora Bielawska-Batorowicz, PhD, Marzena Machala, MSc Institute of Psychology, University of Lodz, Lodz, Poland arly studies indicate that fear of labour is the lowest during the second trimester of pregnancy1. There are some controversies whether previous experience of labour results in a decrease or an increase of fear of labour2. If the latter were verified, then a special preparation for women giving birth to their second child would be necessary to prevent the development of excessive fear. It is quite likely that intense fear of labour might be connected with some personality traits, i.e. trait anxiety. Therefore a study was conducted in order to explore the intensity of fear of labour and its changes over the course of the first and subsequent pregnancy, as well as its relations to state trait anxiety.

E

The study was aimed to answer to the following questions: 1. Does the intensity of fear of labour change during gestation in the same way in primiparous and in multiparous women? 2. Is the pattern of changes in the content of fear of labour the same in first and in subsequent pregnancy? 3. Is the fear of labour related to the state trait anxiety, and if so, how is it related?

Methods The sample consisted of two groups of women: 1) one group of women expecting their first baby with no

previous labour experiences (N=18, mean age 24,07); 2) one group of women who already gave birth to one child (17 subjects, mean age 29,06). Women undergoing complicated pregnancy or who underwent forceps or caesarean section delivery were excluded. Both groups did not differ significantly in their level of education, marital and material status.

Measures 1. Questionnaire of the Fear of Labour ´ (QFL) by L. Putynski It consists of 18 items examining the experiences related to labour. Each item (continued on p 14)

QOL

14

QoL Newsletter, 2001; 26

R

E S U LT S State Trait Anxiety and Fear of Labour in First and Subsequent Pregnancy (continued from p 13)

can be answered on 0 – 4 scale, where ‘0’ stands for ‘not at all’ and ‘4’ stands for ‘very much so’. The questionnaire distinguishes 6 aspects of fear of labour: course of labour, health and life of the baby, maternal role, postpartum period, health and life of the mother, loss of control during labour. 2. State Trait Anxiety Inventory (STAI) – Polish version by C. D. Spielberger, J. ´ Strelau, M. Tysarczyk, K. Wrzesniewski Scores from two sub-scales – X-1 state anxiety and X-2 trait anxiety – were used.

group there are no changes over time in fear over the course of labour and over the postnatal period. For multiparas there is no change in fear over maternal health and life. The most dramatic differences between primiparas and multiparas are recorded in fear over health/life of the baby (p

Suggest Documents