Cross-validation of item selection and scoring for the SF-12 Health Survey in nine countries: results from the IQOLA Project. International Quality of Life Assessment

Data from general population surveys (n = 1483 to 9151) in nine European countries (Denmark, France, Germany, Italy, the Netherlands, Norway, Spain, Sweden, and the United Kingdom) were analyzed to cross-validate the selection of questionnaire items for the SF-12 Health Survey and scoring algorithms for 12-item physical and mental component summary measures. In each country, multiple regression methods were used to select 12 SF-36 items that best reproduced the physical and mental health summary scores for the SF-36 Health Survey. Summary scores then were estimated with 12 items in three ways: using standard (U.S.-derived) SF-12 items and scoring algorithms; standard items and country-specific scoring; and country-specific sets of 12 items and scoring. Replication of the 36-item summary measures by the 12-item summary measures was then evaluated through comparison of mean scores and the strength of product-moment correlations. Product-moment correlations between SF-36 summary measures and SF-12 summary measures (standard and country-specific) were very high, ranging from 0.94-0.96 and 0.94-0.97 for the physical and mental summary measures, respectively. Mean 36-item summary measures and comparable 12-item summary measures were within 0.0 to 1.5 points (median = 0.5 points) in each country and were comparable across age groups. Because of the high degree of correspondence between summary physical and mental health measures estimated using the SF-12 and SF-36, it appears that the SF-12 will prove to be a practical alternative to the SF-36 in these countries, for purposes of large group comparisons in which the focus is on overall physical and mental health outcomes.     

A cross sectional assessment of health status instruments in patients with rheumatoid arthritis participating in a clinical trial. Minocycline in Rheumatoid Arthritis Trial Group

OBJECTIVE: To (1) validate the Short-Form Health Survey (SF-36) as a generic functional health status measure in patients with rheumatoid arthritis (RA); and (2) assess correlations between the SF-36 and other outcome measures used in the Minocycline in Rheumatoid Arthritis (MIRA) Trial. METHODS: We conducted a cross sectional analysis of the final visit outcome measures from the 48 week, multicenter, placebo controlled, double blind MIRA trial. Multitrait scaling analyses assessed convergent and discriminant validity and internal consistency reliability of the SF-36 in the study patients. Responses to comparable items on the SF-36 and modified Health Assessment Questionnaire (M-HAQ) regarding physical functioning were compared and questions from both instruments were also compared to other RA outcome measures. RESULTS: In patients with RA, the SF-36 had high internal consistency and reliability, high discriminant and high convergent validity. Moderate correlations were observed (r = -0.46 to -0.61, p < 0.01 in each case) for comparable items on the SF-36 and M-HAQ regarding dressing, walking, and bending. Joint tenderness score correlations with items on the M-HAQ and SF-36, and joint tenderness score correlations with the SF-36 scales were higher than for joint swelling scores. Physician and patient global assessments were most highly correlated (r = 0.58 and 0.66; p < 0.01, respectively) with the SF-36 bodily pain item. CONCLUSION: The SF-36 is a valid instrument for this RA population. The SF-36 correlates with the M-HAQ and the physician and patient global assessments. The usefulness of the SF-36 in measuring change in RA clinical trials requires testing in longitudinal studies.     

Translation and psychometric evaluation of a Chinese version of the SF-36 Health Survey in the United States

Outcome measures are rapidly becoming standard tools in the assessment of clinical effectiveness and in the measurement of health status in populations. In this article we document the development of a self-administered Chinese version of the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) and report the results of psychometric testing among 156 adult Chinese Americans in Boston, Massachusetts. Following the standard guidelines, a Chinese version of the SF-36 was developed through forward-backward translation techniques and committee review. We used psychometric methods to test assumptions underlying construction and scoring of scales and to evaluate the reliability and validity of the Chinese SF-36 as a measure of health status. The preliminary results indicated that missing value rates for the 36 items were consistently low. Item-discriminant validity was high (over 90% scaling successes) for six of the eight scales (Physical Functioning, Role-Physical, Bodily Pain, General Health, Role-Emotional, and Mental Health). Cronbach's alpha coefficient was above 0.70 criterion for all scales except Social Functioning. Reliability estimates also appeared to vary by sample characteristics. We discuss the implications of these findings and identify where further work will be required.     

Population reference values of the Spanish version of the Health Questionnaire SF-36

BACKGROUND: Perceived health status measurements reference values an important information source for health services research. Population-based norms have been proposed to increase their interpretability. In this paper, we have obtained the norms of the Spanish version of the SF-36 Health Survey and have compared them with US norms the questionnaire. MATERIAL AND METHODS: Data were obtained in the home interview survey on drug consumption (February 1996). This is a cross-sectional study of a multi-stage, stratified random sample of non-institutionalized individuals 15 and older residents in Spain. The final sample included 9,984 individuals, but the analysis is based on those individuals 18 or older (n = 9,151). Personal home interviews were carried out. Information included: the SF-36, legal and illegal drugs consumption, and socio-demographic data, among others. Central trend and dispersion statistics were estimated for each of the SF-36 dimension scores according to gender and age group. Cronbach's alpha coefficients were calculated to estimate the reliability of scores. RESULTS: For most SF-36 dimensions, scores were higher (better) among men and among younger age groups (p < 0.01). There was a monotonic score gradient by age which was more intense for physical function and bodily pain. All Cronbach's alpha coefficients were higher than 0.7 (ranging from 0.78 to 0.96). Spanish norms were very similar to those obtained in the US. CONCLUSIONS: Results presented should be considered the population-based norms of the Spanish version of the SF-36 Health Survey and may be useful for interpreting the questionnaire scores. These norms, which are very similar to the original US questionnaire both in absolute values and in the gender and age group distribution patterns, should be carefully used. Considerations for use discussed in the paper should be taken into account.     

Tests of scaling assumptions and construct validity of the Chinese (HK) version of the SF-36 Health Survey

Few health-related quality of life (HRQOL) survey instruments are available to the Chinese, although many have been developed for Western populations. This article describes the testing of the acceptability, conceptual equivalence, scaling assumptions and construct validity of a Chinese (HK [Hong Kong]) version fo the MOS SF-36 Health Survey. A Chinese (HK) SF-36 survey form was developed by an iterative translation process. It was administered to 236 Chinese subjects who also rated the understanding, difficulty, relevance, and acceptability of each question. The scores were tested against the original scaling assumptions. The SF-36 profile of our subjects was compared to U.S. results for conceptual equivalence. Most subjects did not have any problem in understanding and answering the SF-36. Item means were generally clustered as hypothesized. All but a few items satisfied all scaling assumptions. The shape of the eight-scale SF-36 profile was similar to that of American patients, suggesting conceptual equivalence. We conclude that the Chinese (HK) version of the SF-36 Health Survey has achieved conceptual equivalence and satisfied the psychometric scaling assumptions well enough to warrant further use and testing, using the standard scoring algorithms.     

Use of structural equation modeling to test the construct validity of the SF-36 Health Survey in ten countries: results from the IQOLA Project. International Quality of Life Assessment

A crucial prerequisite to the use of the SF-36 Health Survey in multinational studies is the reproduction of the conceptual model underlying its scoring and interpretation. Structural equation modeling (SEM) was used to test these aspects of the construct validity of the SF-36 in ten IQOLA countries: Denmark, France, Germany, Italy, the Netherlands, Norway, Spain, Sweden, the United Kingdom, and the United States. Data came from general population surveys fielded to gather normative data. Measurement and structural models developed in the United States were cross-validated in random halves of the sample in each country. SEM analyses supported the eight first-order factor model of health that underlies the scoring of SF-36 scales and two second-order factors that are the basis for summary physical and mental health measures. A single third-order factor was also observed in support of the hypothesis that all responses to the SF-36 are generated by a single, underlying construct--health. In addition, a third second-order factors, interpreted as general well-being, was shown to improve the fit of the model. This model (including eight first-order factors, three second-order factors, and one third-order factor) was cross-validated using a holdout sample within the United States and in each of the nine other countries. These results confirm the hypothesized relationships between SF-36 items and scales and justify their scoring in each country using standard algorithms. Results also suggest that SF-36 scales and summary physical and mental health measures will have similar interpretations across countries. The practical implications of a third second-order SF-36 factor (general well-being) warrant further study.     

A symptom-based measure of the severity of chronic lung disease: results from the Veterans Health Study

STUDY OBJECTIVES: We developed a symptom-based measure of severity for chronic lung disease (CLD) that can be readily administered in ambulatory care settings and be used to supplement general health-related quality of life (HRQoL) assessments and pathophysiologic indicators in research and clinical care. DESIGN: Cross-sectional data from the Veterans Health Study, an observational study of health outcomes in patients receiving Veterans Affairs (VA) ambulatory care. SETTING: Four VA outpatient clinics. STUDY SUBJECTS: Two hundred ninety-two participants with CLD were identified on the basis of patient report of having a physician's diagnosis of chronic bronchitis, emphysema, or asthma and either using inhaled medications or having a productive cough on most days for 3 months. MEASUREMENTS AND RESULTS: Participants were scheduled for an in-person interview in which they completed a CLD questionnaire and measurements of peak expiratory flow rate (PEFR). They were also mailed an HRQoL questionnaire, the Short Form Health Survey (SF-36). The CLD questionnaire included six symptom items chosen by an expert panel (two items each for dyspnea, wheezing, and productive cough). The combination of these items yielded a CLD severity index that correlated significantly with all eight scales of the SF-36 (range of r, -0.19 to -0.37; p<0.01). In contrast, PEFR had statistically significant correlations only with two SF-36 scales: physical functioning and bodily pain. CONCLUSIONS: The CLD severity index is a reliable and valid patient-administered instrument that may be used to evaluate the effects of CLD on general HRQoL and predict future health services utilization.     

Differential item functioning in the Danish translation of the SF-36

Statistical analyses of Differential Item Functioning (DIF) can be used for rigorous translation evaluations. DIF techniques test whether each item functions in the same way, irrespective of the country, language, or culture of the respondents. For a given level of health, the score on any item should be independent of nationality. This requirement can be tested through contingency-table methods, which are efficient for analyzing all types of items. We investigated DIF in the Danish translation of the SF-36 Health Survey, using two general population samples (USA, n = 1,506; Denmark, n = 3,950). DIF was identified for 12 out of 35 items. These results agreed with independent ratings of translation quality, but the statistical techniques were more sensitive. When included in scales, the items exhibiting DIF had only a little impact on conclusions about cross-national differences in health in the general population. However, if used as single items, the DIF items could seriously bias results from cross-national comparisons. Also, the DIF items might have larger impact on cross-national comparison of groups with poorer health status. We conclude that analysis of DIF is useful for evaluating questionnaire translations.     

Psychometric and clinical tests of validity of the Japanese SF-36 Health Survey

Cross-sectional data from a representative sample of the general population in Japan were analyzed to test the validity of Japanese SF-36 Health Survey scales as measures of physical and mental health. Results from psychometric and clinical tests of validity were compared. Principal components analyses were used to test for the hypothesized physical and mental dimensions of health and the pattern of scale correlations with those components. To test the clinical validity of SF-36 scale scores, self-reports of chronic medical conditions and the Zung Self-Rating Depression Scale were used to create mutually exclusive groups differing in the severity of physical and mental conditions. The pattern of correlations between the SF-36 scales and the two empirically derived components generally confirmed hypotheses for most scales. Results of psychometric and clinical tests of validity were in agreement for the Physical Functioning, Role-Physical, Vitality, Social Functioning, and Mental Health scales. Relatively less agreement between psychometric and clinical tests of validity was observed for the Bodily Pain, General Health, and Role-Emotional scales, and the physical and mental health factor content of those scales was not consistent with hypotheses. In clinical tests of validity, the General Health, Bodily Pain, and Physical Functioning scales were the most valid scales in discriminating between groups with and without a severe physical condition. Scales that correlated highest with mental health in the components analysis (Mental Health and Vitality) also were most valid in discriminating between groups with and without depression. The results of this study provide preliminary interpretation guidelines for all SF-36 scales, although caution is recommended in the interpretation of the Role-Emotional, Bodily Pain, and General Health scales pending further studies in Japan.     

Translation, validation, and norming of the Dutch language version of the SF-36 Health Survey in community and chronic disease populations

The primary objectives of this research were to translate, validate, and generate normative data on the SF-36 Health Survey for use among Dutch-speaking residents of the Netherlands. Translation of the SF-36 into Dutch followed the stepwise, iterative procedures developed by the IQOLA Project. Following extensive pilot testing, the SF-36 was administered to: (1) a random sample of adult residents of Amsterdam (n = 4172); (2) a random, nationwide sample of adults (n = 1742); (3) a sample of migraine sufferers (n = 423); and (4) a sample of cancer patients undergoing active anti-neoplastic treatment (n = 485). Data quality across the four studies was consistently high. The rates of missing data ranged from 1% to 5% at the item level, and from 1.2% to 2.6% at the scale level. Multitrait scaling analysis confirmed the hypothesized scale structure of the SF-36 and associated scale scoring in all four samples. Cronbach's alpha coefficients surpassed the 0.70 criterion for group comparisons in all but one case (the Social Functioning scale in the cancer sample), with a mean alpha coefficient across all scales and samples of 0.84. Known-group comparisons yielded consistent support for the validity of the SF-36. In the two community samples, statistically significant differences in SF-36 mean scale scores were observed as a function of age, gender, and the prevalence of chronic health conditions. In the migraine and cancer samples, mean SF-36 scale scores varied significantly as a function of various indicators of disease severity. The SF-36 profiles for the two community samples were highly similar. The cancer sample yielded the lowest SF-36 scores, with the migraine sample holding an intermediate position. On-going studies will generate data on the responsiveness of the SF-36 to within-group changes in health over time. Efforts are underway to translate and validate the questionnaire for use among ethnic minority groups in the Netherlands.     

Cloning and characterization of the CDKN2A and p19ARF genes from Monodelphis domestica

The psychometric properties of the Belgian Dutch and French translations of the SF-36 Health Survey were evaluated in a sample of 4448 Belgian patients with angina enrolled in a 6-month treatment study. Missing data were rare (<2%), and tests of both item internal consistency and item discriminant validity were satisfactory in both languages. Cronbach's alpha coefficient ranged from 0.81 to 0.91 (Dutch) and 0.82 to 0.92 (French). SF-36 scales discriminated between groups of patients differing in age and in the number of weekly angina attacks, change over 6 months in the number of weekly angina attacks and physician assessment of change in physical condition both were significantly related to changes in SF-36 scale scores. On average, scale scores for French-speaking patients were lower than for Dutch-speaking patients, most notably for Vitality and Mental Health. The average change in SF-36 scale scores over 6 months, in relation to change in clinical criteria, was similar in both language groups. The psychometric properties of the Belgian Dutch and French translations should be tested further in Belgium to determine whether the generally favorable results reported here can be replicated in other samples.     

Evaluating translations of health status questionnaires. Methods from the IQOLA project. International Quality of Life Assessment

There is growing demand for translations of health status questionnaires for use in multinational drug therapy studies and for population comparisons of health statistics. The International Quality of Life Assessment (IQOLA) Project is conducting a three-stage research program to determine the feasibility of translating the SF-36 Health Survey, widely used in English-speaking countries, into other languages. In stage 1, the conceptual equivalence and acceptability of translated questionnaires are evaluated and improved using qualitative and quantitative methods. In stage 2, assumptions underlying the construction and scoring of questionnaire scales are tested empirically. In stage 3, the equivalence of the interpretation of questionnaire scores across countries is tested using methods that closely approximate their intended use, and empirical results are compared. Data analyses from Sweden and the United Kingdom, as well as other research cited, support the feasibility of cross-cultural health measurement using the SF-36.     

Testing the validity of the Euroqol and comparing it with the SF-36 health survey questionnaire

There is an interest in the consequences of deriving a single index measure of health for validity and sensitivity. This paper presents the results of testing a recent example of a general health measure designed to derive a single index, the Euroqol (EQ), and presents a comparison with a new, influential profile measure, the Short Form 36 (SF-36) Health Survey Instrument. The EQ and an anglicised version of the SF-36 health survey, both designed for self-completion, were included in a postal survey of a random sample of 1980 patients from two practice lists in Sheffield, UK. The response rate for the EQ questionnaire was 83%, and the rate of completion over 95%. Evidence was found for the construct validity of the EQ dimension responses and the derived total EQ health score in terms of distinguishing between groups with expected health differences. Considerable agreement was found between EQ responses and the total EQ score, and the UK SF-36 profile scores. There was substantial evidence of EQ being less sensitive at the ceiling (i.e. low levels of perceived ill-health) and throughout the range of health states. A recent restructuring of the EQ, may help overcome some of the problems with the physical dimensions by reducing their skewness.     

Performance of a self-administered mailed version of the Quality of Well-Being (QWB-SA) questionnaire among older adults

OBJECTIVES: The Quality of Well-Being questionnaire is a measure of health-related quality of life (HRQoL) that has several desirable properties. Its widespread use has been hindered because it is difficult to administer. To overcome this limitation, a new self-administered form has recently been developed. This study examined the feasibility of using the Quality of Well-Being-Self-Administered (QWB-SA) questionnaire in an older population. METHODS: The Quality of Well-Being-Self-Administered questionnaire was sent to 430 community-dwelling individuals aged 65 years and older who were randomly selected from primary care physicians' offices. Response patterns, scaling distributions, and the acceptability of the survey were examined for all respondents. The results of the QWB-SA questionnaire were compared to the Sickness Impact Profile (SIP) and the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36) for those individuals who also had completed the latter two surveys approximately 10 months earlier and whose health had not changed substantially in the meantime. RESULTS: Three hundred and one older adults (70%) responded. The mean QWB-SA questionnaire score was 0.7035. The scores were not skewed, and there were no floor or ceiling effects. The mean time to complete the QWB-SA questionnaire was 14.2 minutes, which was significantly shorter than for the SIP (19.3 minutes) but significantly longer than for the SF-36 (12.5 minutes). Subjects rated their satisfaction with the QWB-SA questionnaire somewhat lower than for the SIP and similar to SF-36. Correlations between the QWB-SA questionnaire and the SIP and SF-36 were moderate and were generally stronger for measures of physical health than for other domains such as mental health. CONCLUSIONS: The self-administered QWB questionnaire was acceptable to older respondents, and it correlated with other measures of health-related quality of life. It can be considered as a candidate for some research applications among older adults.     

The Swedish SF-36 Health Survey III. Evaluation of criterion-based validity: results from normative population

Assumptions of the variation in SF-36 scale scores were tested in relation to external criteria in 8930 respondents comprising the Swedish norming population. Physical health scales were strongly associated with age, while small differences were found for the Mental Health scale across age groups. Females reported poorer health than males, particularly in ages between 30-40 and over 70. Worse health profiles were associated with social risk factors (unemployment, divorce, etc.). The disability pension rate was strongly related to reduced Physical Functioning and increased Bodily Pain. The use of medical care was reflected in general health scores (i.e., the lower the scale score, the higher the care consumption). Self-reported physical and psychological symptoms were selectively related to SF-36 scales. All SF-36 scales, except Mental Health, were more strongly related to ratings of health satisfaction than to global quality of life. Combinations of the SF-36 well-being scales explained a substantial part of the variance of these ratings. In conclusion, the performed criterion-validity tests support the cross-cultural stability of the SF-36.     

The factor structure of the SF-36 Health Survey in 10 countries: results from the IQOLA Project. International Quality of Life Assessment

Studies of the factor structure of the SF-36 Health Survey are an important step in its construct validation. Its structure is also the psychometric basis for scoring physical and mental health summary scales, which are proving useful in simplifying and interpreting statistical analyses. To test the generalizability of the SF-36 factor structure, product-moment correlations among the eight SF-36 Health Survey scales were estimated for representative samples of general populations in each of 10 countries. Matrices were independently factor analyzed using identical methods to test for hypothesized physical and mental health components, and results were compared with those published for the United States. Following simple orthogonal rotation of two principal components, they were easily interpreted as dimensions of physical and mental health in all countries. These components accounted for 76% to 85% of the reliable variance in scale scores across nine European countries, in comparison with 82% in the United States. Similar patterns of correlations between the eight scales and the components were observed across all countries and across age and gender subgroups within each country. Correlations with the physical component were highest (0.64 to 0.86) for the Physical Functioning, Role Physical, and Bodily Pain scales, whereas the Mental Health, Role Emotional, and Social Functioning scales correlated highest (0.62 to 0.91) with the mental component. Secondary correlations for both clusters of scales were much lower. Scales measuring General Health and Vitality correlated moderately with both physical and mental health components. These results support the construct validity of the SF-36 translations and the scoring of physical and mental health components in all countries studied.     

The French SF-36 Health Survey: translation, cultural adaptation and preliminary psychometric evaluation

This article reports on the main developmental stages and on the preliminary psychometric assessment of the final French version of the SF-36. A standard forward/backward translation procedure was followed. When translating survey items, the emphasis was placed on conceptual equivalence. When translating response choices, we attempted to select a set of response choices that replicate the U.S. version. The distance between the response choices was checked using visual analogue scales (N = 30). The adaptation procedure also included formal ratings of the difficulty of the translation, of the quality of the translation, and of the equivalence between the American source version and the French target version. The face validity was checked during lay panel sessions at which the translated questionnaire was administered to subjects from the general public, hospital employees, and subjects with a low level of education. Standard psychometric techniques were used to evaluate the cultural adaptation of the SF-36, using data from a general population survey. The main objective of this analysis was to determine how well the scaling assumptions (summated rating or Likert-type scaling construction) of the SF-36 were satisfied. The results support the claim that the scaling properties of the French version of the SF-36 are adequate and that health outcomes may be reliably assessed using this version of the instrument.     

The MOS SF-36 health survey questionnaire in severe chronic airflow limitation: comparison with the Nottingham Health Profile

This study documents the cross-sectional, health-related quality of life (HRQOL) measures obtained at baseline for patients with severe chronic airways limitation (CAL) being assessed for home oxygen therapy (HOT) at the Flinders Medical Centre, Adelaide, South Australia. Two generic quality of life instruments, the Nottingham Health Profile (NHP) and the Medical Outcomes Study (MOS) short form 36-item questionnaire (SF-36), were administered by interview to the same patients to permit comparisons to be made between the two instruments. SF-36 mean scores were also compared with scores obtained in separate studies of a South Australian elderly general population and of groups of Australian subjects with various medical and psychiatric conditions. NHP mean scores were compared with scores from an elderly group of Adelaide residents from a household survey. HRQOL measures were obtained for 60 patients, 32 males and 28 females. At assessment for HOT, patients with severe CAL were experiencing severe impairment in their quality of life in comparison to age-matched South Australian norms, with physical disability the major limitation. There were several significant correlations between the domains of the SF-36 and the NHP which were predominantly gender-specific. Only small decrements in mental health were found with the SF-36 questionnaire. The SF-36 and the NHP appear to provide discrepant information for severely disabled CAL patients for the subjective domains of emotional and mental health.     

Deriving a preference-based single index from the UK SF-36 Health Survey

This article presents the results of a study to derive a preference-based single index from the SF-36. The study was an attempt to reconcile a profile health status measure, the SF-36, with the "quality adjusted life years" approach. The study undertook a parsimonious restructuring of the SF-36 using explicit criteria to form the SF-6D health state classification. A sample of multidimensional health states defined by this classification were valued by a convenience sample of health professionals, managers, and patients, who responded to a set of visual analogue scale ratings and standard gamble questions, with highly complete and consistent answers. Statistical models were estimated to predict single index scores for all 9000 health states defined by the new classification. The resultant algorithms can be applied to existing SF-36 data sets and used in the assessment of the cost-effectiveness of health technologies. This preliminary work forms the basis of a larger study currently being undertaken in the UK.