The Challenge of Cross-Cultural Quality of Life Assessment

The Challenge of Cross-Cultural Quality of Life Assessment Monika Bullinger, Silke Schmidt Institute of Medical Psychology University of Hamburg

Background Interest in Health-Related Quality of Life as v descriptor of functioning and well-being of populations with and without medical conditions (epidemiological perspective) v outcome criterion for interventions (clinical perspective) v an aid for decision making in the health care field (political perspective)

Meanings of the term international v political: nation v geographical: country v anthropological: v sociological: culture society v psychological: identity language

Demands Measures of Quality of Life should be sensitive to v language and dialect v customs, beliefs and traditions v education and socioeconomic status Nat. Cancer Institute 1992

Essential questions v Is Quality of Life a relevant concept in a given nation/ culture? v Do nations/cultural groups share an identical set of concepts about Quality of Life? v Can Quality of Life concepts be assessed with instruments? v Is Quality of Life measurable across nations/cultures with the same instrument? v Can Quality of Life data be compared across nations/ cultures? v Do cross-cultural Quality of Life results provide a sound basis for decision making?

Problems v Ethnocentrism? v Normativity v Bias of concept? in assessment? v Ethical consequences?

Statement "Although some researchers may desire a scale or similar instruments for global assessments of cultures, permitting comparison of the "nature" of one culture with another (. . . ), no such scale exists. In fact, given the multiplicity of variables or domains comprising a culture, that goal is unrealistic, both theoretically and methodologically. " T. M. Johnson in Spilker 1996, p. 511

Criteria v functional equivalence adequacy of translation v scale equivalence comparability of response scales v operational equivalence standardization of psychometric testing v metric equivalence order of scale values along criterion Hui & Triandis 1985

Approaches to cross-cultural instrument development v sequential approach (transfering an existing questionnaire to another culture, e. g. SF-36 Health Survey) v parallel approach (assembling an instrument based on existing scales from different cultures, e. g. EORTC QLQC 30) v simultaneous approach (cooperative cross-cultural development of a questionnaire, e. g. WHO-QOL)

Steps in instrument development v Item development (focus groups; expert pannel; cognitive debriefing). . . v Translation (foreward, backward, piloting) v Psychometric testing (reliability, validity, responsiveness) v Norming (representative population sample, weighing)

Focus Groups v Can help to identify relevant concepts v Involve v can of potential respondents and/or experts be active in dimension/facet/item generation

Cognitive Debriefing v Is used to examine the concepts/dimensions/items from the respondent perspective v Can be performed individually or in the group v Should use standard format of presentation, discussion and documentation

Question writing Questions should: v be based as far as possible on evaluation of questions already in use in the countries or on the suggestions of experts and lay-people participating in the focus groups; v give rise to answers that are enlightening about the concepts to be measured; v reflect the meaning conveyed in the definition of the indicator and its domains/facets; v cover, in combination with other questions for a given indicator, the key aspects of that as described in the definition

Questions should. . . v use simple language, avoiding ambiguity in terms of either wording or phraseology; v be shorter rather than longer; v avoid double negatives; v be amenable to a rating scale; v ask about a single issue/facet; v be applicable to individuals with a range of health status; v be phrased as questions and not statements; v reflect the typology of questions adopted for the project

The Translation Process Difficulty 2 Raters Quality foreward 2 Raters Quality backward 2 Raters Original Target Original 2 Translaters Comparison

Testing Response Scales v Can be used to assess the conceptual equivalence of response scales v Helps to examine the interval properties of response scales within and across cultures v can be performed using visual-analogue (e. g. “none-all of the time”) scales as anchors with responses to be set between these (e. g. “sometimes”) v Ensures cross-cultural comparability of instruments

Cross- cultural psychometric testing v Data sets: l national, combined national, global v Methods: classical psychometrics (e. g. Cronbachs Alpha), l structural equation modelling (e. g. EQS), l modern approaches (e. g. Item Response Theory) l v Desired Product: l a cross-culturally usable and interpretable measure

Internationally active groups Quality of Life working groups v The Nottingham Health Profile (NHP) Group v The European Organization for Research and Treatment of Cancer (EORTC) Group v The International Quality of Life Assessment Project Group (IQOLA: SF-36) v The Sickness Impact Profile (SIP) Group v The World Health Organization Quality of Life (WHOQOL) Group v The European Quality of Life Project (EUROQOL) Group v The Functional Assessment of Cancer Treatment (FACT) Group

IQOLA Project Phases v translation v scale --> survey form construction --> scoring algorithms v validation and norming --> interpretation v publication of results --> user friendly guidelines

IQOLA Project v Prototype project for sequential approach (SF-36 Health Survey) v Was the first project to develop standards for cross-cultural QOL research v Translated and tested the SF-36 Health Survey in 15 (Phase 1) countries v Representative (norm) data available in most western European countries

Results from the IQOLA project v The psychometric properties of the questionnaire acceptable in each culture. v In norming studies, the scale scores of SF-36 scores do differ. v There is considerable overlap between western countries in the dimensional structure of the SF-36.

WHO QOL Project v Prototype project for simultaneous approach (WHOQOL) v Generated groups) v Reduced v Tested the items within each country (focus items empirically in a field study performance of WHO QOL-100 (-BREF) in several studies.

Results from the WHOQOL project v The items constructed by different cultures are comparable. v National items do not contribute significantly to the instrument's quality. v Structural equation modelling does not show substantial difference in the relationships of dimensions across cultures.

Cross-cultural Quality of Life Assessment v Conceptual level: l Representation and operationalisation of the concept v Methodological level: l Type and appropriateness of assessment v Application level l Practical considerations and feasibility in cross-cultural settings, v Policy level l Ethical considerations and interpretability of crosscultural data sets

Cross-cultural Quality of Life Assessment (Examples) v. Adults: IQOLA-Project (SF-36) WHO Qo. L-Project (WHOQOL) EUROHIS WHOQOL OLD v. Children: DISABKIDS-Project Haemo. Qol-Project ESCH-Qo. L-Project KIDSCREEN- Project

WHOQOL-OLD- Cross-cultural structural testing (pilot) Country Code nvalid CFI Scotland 1 116 . 967 England 2 145 . 970 Germany 3 354 . 982 Spain 4 271 . 979 Denmark 5 384 . 975 Czech R 7 325 ? Hungary 8 333 . 981 Canada 10 178 . 967 U. S. A. 12 295 . 967 Israel 13 250 . 979 Sweden 16 431 . 976 Uruquay 20 248 . 977 Lithuania 23 342 . 977

EUROHIS-QOL 8 item index for different countries

· Differential item functioning analyses of the 8 items of the EUROHIS -QOL 8 item index Country Gender Age group - Condition R 2 -diff RUMM 1 How would you rate your quality of life . 005 . 000 . 003 . 121 . 002 . 592 . 001 . 078 2 How satisfie d are you with your health . 012 . 000 . 001 . 186 . 017 . 000 . 020 . 034 3 Do you have enough energy for everyday life . 008 . 076 . 003 . 954 . 004 . 065 . 003 . 288 4 How satisfied are you with your ability to perform you r daily activities . 016 . 000 . 002 . 272 . 009 . 853 . 003 . 234 5 How satisfied are you with yourself . 027 . 000 . 001 . 647 . 006 . 039 . 003 . 596 6 How satisfied are you with your personal relationships . 010 . 072 . 003 . 335 . 003 . 048 . 0 06 . 082 7 Have you enough money to meet your needs . 027 . 000 . 001 . 013 . 015 . 006 . 102 8 How satisfied are you with the conditions of your living place RUMM . 014 . 000 . 004 . 919 . 027 . 977 . 010 . 078

Summary v International efforts to assess Quality of Life crossculturally exist. v Many instruments need to be reviewed for their crosscultural performance. v First results suggest a cross-cultural applicability of instruments. Considerations in Qol research in developing countries v value to all collaborating parties v compatibility with ressources/energies v compliance with ethical and moral standards of collaborators/subjects

Ethics in cross-cultural Quality of Life research v transparency of underlying concepts v modesty in using specific measurement approaches v correctness data in applying instruments and analyzing v responsibility publication for the results also after their

Conclusions v Quality of life seems to be a universal human concept as concerns its relevant dimensions. v Different individual behaviors, societal conditions and cultural regulations may apply, but these concern the means rather than the results of pursuing well-being. v Although cultures do differ in the basic conditions provided to strive at a favorable Quality of Life, the person's subjective perception is not a linear reflection of these conditions.

Future Provided that the Quality of Life of citizens is a mayor concern in the given country, Quality of Life data v may give information about the respective status of populations, v may thus suggest plans to improve the Quality of Life status of populations by specific interventions, v can be used to measure the effects of such interventions v and can contribute to minimizing the gap between the "developed" and the "developing" world.