Psychometric evaluation of the 12-item short-form health survey (SF-12) in osteoarthritis and rheumatoid arthritis clinical trials

Abstract

Background: The psychometric properties of the 12-Item Short-Form Health Survey (SF-12), a subset of the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36), have been tested in the general population and certain disease states.

Objective: The purpose of this study was to evaluate the psychometric properties of the SF-12 as a generic measure of health-related quality of life (HRQoL) in osteoarthritis (OA) and rheumatoid arthritis (RA) patient populations in clinical trials.

Methods: Data were aggregated from 5 clinical trials evaluating the efficacy of non-steroidal anti-inflammatory drugs in OA (n = 651) and RA (n = 693) patients. Patient assessments in these trials were made using the SF-36 and commonly used clinical measures of OA and RA at baseline and after up to 6 weeks of treatment. For the items of the SF-36 contained in the SF-12, the item missing rate, computability of scores, floor and ceiling effects, factor structure, and item-component correlations were evaluated. Clinical variables and correlations of physical component summary (PCS-12) and mental component summary (MCS-12) scores of the SF-12 with the corresponding SF-36 component summary scores (PCS-36 and MCS-36) were also examined. Analyses were conducted separately for OA and RA patients.

Results: A low individual SF-12 item missing rate (0.29% to 2.30%) and a high percentage score computability (90.9%-94.3%) were observed at baseline. No floor or ceiling effects at baseline were observed. The scree plot confirmed the 2-factor structure of the SF-12 items. Items belonging to the physical component correlated more strongly with the PCS-12 than with the MCS-12; similarly, items belonging to the mental component correlated more strongly with the MCS-12 than with the PCS-12. The correlations between the PCS-12 and PCS-36 and between the MCS-12 and MCS-36 ranged from 0.92 to 0.96 (P < 0.001) at baseline and at week 2, 4, or 6. Significant correlations ranging from -0.09 to -0.58 (P < 0.05) were observed between the SF-12 scores and clinical variables.

Conclusion: The SF-12 appears to be a psychometrically sound tool for the assessment of HRQoL in OA and RA patients.