DISCRIMINATIVE POWER OF THE HEALTH-STATUS QUESTIONNAIRE-12 IN RELATION TO AGE, SEX, AND LONGSTANDING ILLNESS - FINDINGS FROM A SURVEY OF HOUSEHOLDS IN GREAT-BRITAIN

Study objective-To assess the ability of the health status questionnai re 12 (HSQ-12) to discriminate between older and younger age groups, i ts appropriateness for use with an older population in terms of the sp read of responses across categories, floor or ceiling effects, and its ability to discriminate between those with and without a reported lon gstanding illness and type (sensitivity and specificity). Design and s etting-The vehicle for the study was the Office for National Statistic s (ONS) omnibus survey in Great Britain. The sampling frame was the Br itish postcode address file of ''small users'', stratified by region, and socioeconomic factors. This file includes all private household ad dresses. The postal sectors were selected with probability proportiona l to size. Within each sector 30 addresses were selected randomly. The number of selected addresses was 3000. Participants-Altogether 1912 a dults aged 16 and over were interviewed in person in their own homes, giving a response rate of 72%. Measures-The HSQ-12, and the ONS genera l household survey questions on longstanding illness; the ONS omnibus standard sociodemographic items. Main results-There were exceptionally high rates of item response in all age groups. The score differences by construct (eg, age group, sex, longstanding illness) were in the ex pected directions with statistically significant age gradients. Age wa s associated with most of the HSQ-12 domains, although this associatio n had interactions with longstanding illness or sex. The differences i n HSQ-12. scores with reported longstanding illness and type of longst anding illness made theoretical sense, which supports the discriminati ve power of the scale. The frequency distributions for HSQ-12 items in relation to age and sex, and by reporting of longstanding illness are also presented here in order to demonstrate ceiling effects. Most res pondents in all age groups achieved high (good) scores on the ''social functioning'' subscale. The HSQ-12 had good results for specificity w hen tested against reporting of a longstanding illness, although this was at the expense of sensitivity. Conclusions-The results support the use of the HSQ-12 with older populations, particularly for those with chronic illnesses, although it will reveal relatively few problems am ong younger populations. The results presented here indicate that it w ill require supplementation with more sensitive disease and/or domain specific scales in the areas of interest or intervention, but it provi des an acceptable, brief, core measure of health related quality of li fe. This paper presents the first British normative data using the HSQ -12.