Psychometric properties of the SF-12 (Hebrew version) in a primary care population in Israel

OBJECTIVE. To evaluate the psychometric properties of the 12-item Medical Outcomes Study (MOS) Short-Form Health Status Survey (SF-12) (Hebrew version) in a large primary care patient sample in Israel. SUBJECTS. The sample comprised 3631 adült primary care patients who were screened for depression in a longitudinal design. MEASURES. SF-12 yielding two scores: the Physical Component Summary (PCS-12) and Mental Component Summary (MCS-12). For comparison, the Center for Epidemiological Studies - Depression scale (CES-D), two subscales from the 90-item Hopkins Symptom Scale (SCL-90), Quality of Life Depression Scale (QLDS), World Health Organization Quality of Life Measure - Bref (WHOQoL-Bref), and interviewer-administered Composite International Diagnostic Interview (CIDI) were also administered. RESULTS. Score distribution was satisfactory, the amount of missing data was minimal and item-to-item correlations were satisfactory. Floor and ceiling effects were minimal in items with more than three response options. A confirmatory factor analysis supported the two-dimensional model of health. Test-retest reliability was good for both summary scales in a nondepressed population and for PCS-12 in the depressed population, but only moderate for MCS-12 in the depressed population. In regard to convergent validity, MCS-12 correlated negatively and significantly with the CES-D, QLDS, and two SCL-90 anxiety subscales, and positively and significantly with four WHOQoL-Bref domains. PCS-12 correlated positively and significantly with the WHOQoL-Bref physical domain. Discriminative validity was established in that both PCS-12 and MCS-12 showed meaningful effect sizes between groups with various degrees of physical and mental health problems. Sensitivity to change was established in that both PCS-12 and MCS-12 change scores were significantly different in subjects who had been depressed but were no longer so, as compared with those who were still depressed at follow-up. CONCLUSIONS. The SF-12 (Hebrew version) is a reliable and valid measure, particularly in a nondepressed population.