SF-6D population norms for the Hong Kong Chinese general population

Abstract

Purpose: To estimate population norms for the SF-6D health preference (utility) scores derived from the MOS SF-36 version 1 (SF-36v1), SF-36 version 2 (SF-36v2), and (SF-12v2) health surveys collected from a representative adult sample in Hong Kong, and to assess differences in SF-6D scores across sociodemographic subgroups. Methods: A random telephone survey of 2410 Chinese adults was conducted. All respondents completed questionnaires on sociodemographics and presence of chronic diseases (hypertension, diabetes, chronic rheumatism, chronic lung diseases, stroke, and mental illness), and the short-form 36-item health survey (SF-36) version 1, and selected items of the SF-36v2 that were different from those of SF-36v1. Responses of short-form 12-item health survey (SF-12) were extracted from responses of the SF-36 items. SF-6D health utility scores were derived from SF-36 version 1 (SF-6DSF-36v1), SF-36 version 2 (SF-6DSF-36v2), and SF-12 version 2 (SF-6DSF-12v2) using Hong Kong SF-6D value set. Results: Population norms of SF-6DSF-36v1, SF-6DSF-36v2, and SF-6DSF-12v2 for the Hong Kong Chinese were 0.7947 (± 0.0048), 0.7862 (± 0.0049), and 0.8147 (± 0.0050), respectively. Three SF-6D scores were highly correlated (0.861–0.954), and had a high degree of reliability and absolute agreement. Males had higher health utility scores (SF-6DSF-36v1: 0.0025; SF-6DSF-36v2: 0.025; SF-6DSF-12v2: 0.018) but reported less problems in all the dimensions than women. Respondents with a higher number of chronic diseases had lower SF-6D scores. Among all respondents with one or more chronic diseases, those with hypertension scored the highest whereby those with mental illness scored the least. Conclusions: The SF-6D utility scores derived from different SF-36 or SF-12 health surveys were different. The population norms based on these three health surveys enable the normative comparisons of health utility scores from specific population or patient groups, and provide estimates of age–gender adjusted health utility scores for health economic evaluations.