Substitution of SF-36 by SF-12 Among Hong Kong Chinese Older Adults: Secondary Analysis of Randomized Controlled Trials

Purpose This study aims to examine the appropriateness of substituting the Short-Form 36 (SF-36) by its shortened version (SF-12) in measuring health-related quality of life (HRQoL) in older Chinese population. Methods Secondary analysis of two transitional care management programs, conducted from 2009 to 2012, were analyzed (n=1188, aged 60-97). Participants were discharged patients with respiratory disease, type 2 diabetes, cardiac disease, and renal disease, and were classified according to number of chronic diseases. SF-36 was administered at baseline and 4-week follow-up. Results Both overestimations and underestimations of HRQoL by SF-12 were found. Most domain scores of SF-36 and SF-12 were highly correlated (Spearman correlation (rho) > 0.85), with the exception of General Health (rho=0.64) and Vitality subscales (rho=0.82). Multiple linear regression adjusted for demographic characteristics showed that the four out of eight domains of SF-36 and SF-12 were equivalent in measuring the difference across numbers of chronic diseases (all p < 0.05). Paired sample t tests in 989 (83.2 %) who completed the SF-36 survey 4 weeks after baseline showed that SF-12 overestimated the 4-week changes in most of the domains. Conclusion The use of the Chinese version of SF-12v2 for reporting the change over time in quality of life among medical patients after hospital discharge may need to be interpreted with caution. The SF-12 tends to underestimate the difference when compared with the SF-36.