Evaluation of the Short-Form Health Survey (SF-36) Using the Rasch Model

Abstract

Introduction. Health-related quality of life (HRQOL) is an outcome variable of growing importance in chronic disease research. Many intervention-type studies seek to show improvements in HRQOL based on treatment effects. As interest grows in using HRQOL as an outcome measure, the need to investigate the measurement properties of HRQOL assessments increases in importance. Objective. The purpose of this study was to evaluate the SF-36 for proper measurement functioning using the Rasch model. Methods. A total of 634 participants completed the SF-36 HRQOL assessment. The Rasch partial credit model was used to analyze the two dominant HRQOL domains (physical and mental) of the assessment Results. Majority of the total criteria used for optimal category functioning were met for the physical health domain and all of the total criteria were met for the mental health domain. Both convergent and construct validity evidence provided substantial confirmation for the use of the Rasch physical and mental health person scores as measures of HRQOL. Conclusion. Results of this study showed that the SF-36 met stringent modern measurement criteria using the Rasch model.