Transforming generic HRQOL instruments to a summary utility index is useful for deriving quality adjusted life years (QALY) in any cost/QALY analysis. Nevertheless, relatively little research has been done to map or revalue generic HRQOL measures to utility scores. OBJECTIVE: The purpose of this study was to investigate the role of the SF-12 in predicting utility scores derived from Health Utility Index (HUI) and Visual Analogue Scale (VAS). METHOD: Data were obtained from 6 ,000 randomly selected managed care patients age 25 to 95 years in the US. The SF-12, HUI-III and VAS were used in the survey to assess health status. The SF-12 items were used to predict HUI and VAS scores using least square regression, with disease and sociodemographic covariates. A second model entered each SF-12 item as categorized responses to avoid any arbitrary assumption about the differences in quality of life between different item response alternatives in the SF-12. Model heteroscedasticity was corrected by White variance-covariance matrix. A 'Bounded Influence Estimation' procedure was also used to capture the effect of any extreme utility score. RESULTS: The SF-12 items and socio- demographic covariates accounted for 47% to 52% of the variations in the HUI-III and VAS. Age and items of the SF-12 were significantly (p<0.05) associated with variations in utility scores. Model specification satisfied the RESET test and the linear model showed a favorable result compared to the log-log or semi-log model. CONCLUSIONS: This research provides support that an algorithm can be derived from the SF-12 to predict HUI and VAS, and thus utility scores. This is also consistent with the previous findings on predictive algorithms using the SF-36 and HUI to derive utility scores.