OBJECTIVES: This study aimed to validate the Short Form 12-Item Survey version-2 (SF-12v2) in an older (65 years or greater) US population as well as estimate disutilities associated with relevant conditions.

METHODS: Data from the Medical Expenditure Panel Survey longitudinal panel (2014-2015) were utilized. The physical component summary (PCS) and mental component summary (MCS) scores were examined for reliability (internal consistency, test-retest), construct validity (convergent and discriminant, structural) and criterion validity (concurrent and predictive). The Short Form Six Dimension score was used to estimate disutility scores.

RESULTS: The sample consisted of 1,040 older adults. PCS and MCS demonstrated high internal consistency (Cronbach alpha: PCS:0.87, MCS:0.86) and good and moderate test-retest validity, respectively (Intraclass correlation coefficient: PCS:0.79, MCS:0.59). PCS had a moderately strong association with perceived health (r=0.58) and a fair relation with perceived mental health (r=0.35), Kessler scale (r=0.39) and Patient Health Questionnaire (PHQ-2)(r=0.33). MCS had a moderately strong association with Kessler scale (r=0.66) and PHQ-2 (r=0.59) and fair relation with perceived health (r=0.31) and perceived mental health (r=0.37). The questionnaire demonstrated sufficient convergent and discriminant ability. Confirmatory factor analysis showed adequate fit with the theoretical model and structural validity (goodness of fit=0.9588). A 1-unit increase in MCS score was associated with decreased odds of future social limitations (Odds Ratio (OR): 0.948(0.930, 0.965)) and cognitive limitations (OR: 0.920(0.903, 0.937)). Correspondingly, a 1-unit increase in PCS score was associated with decreased odds of future activity limitations (OR: 0.885(0.870, 0.900)) and functional limitations (OR: 0.877(0.863, 0.891)). Concurrent criterion validity as well as predictive criterion validity were demonstrated. Activity limitations, functional limitations, arthritis, coronary heart disease, diabetes, myocardial infarction, stroke, angina and high blood pressure were associated with a disutility of 0.17, 0.15, 0.06, 0.07, 0.07, 0.06, 0.09, 0.06, and 0.08 respectively.

CONCLUSIONS: The SF-12v2 is a valid and reliable instrument in an older US population.