Construct Validity of the Stroke-Specific Quality of Life Questionnaire in Ischemic Stroke Patients

Abstract

Hsueh I-P, Jeng J-S, Lee Y, Sheu C-F, Hsieh C-L. Construct validity of the Stroke-Specific Quality of Life questionnaire in ischemic stroke patients. Objectives To investigate the construct validity of the commonly used 8- and 12-domain versions of the Stroke-Specific Quality of Life (SS-QOL) with a first-order factor model. The better-fitted version was further evaluated by a second-order factor structure model in order to determine whether a summary score is justified. Design Cohort study. Setting Outpatient stroke clinic. Participants Patients (N=388) with ischemic stroke. Intervention Not applicable. Main Outcome Measures The SS-QOL. We first conducted confirmatory factor analysis (CFA) to evaluate the construct validity of the first-order 8- or 12-domain versions of the SS-QOL. The better-fitted version was then validated by investigating the second-order health-related quality of life (HRQOL) factor. Results The 12-domain version, but not the 8-domain version, had sufficient goodness of fit (χ 2=2041.7, df=1061, χ 2/ df=1.9, comparative fit index [CFI]=0.98, Tucker-Lewis index [TLI]=0.98, and root mean square error of approximation=0.05). All items of the 12-domain version showed acceptable factor loadings (>0.40) and were retained. Furthermore, the second-order CFA fit indices of the 12 domains were acceptable (χ 2=2630.3, df=1115, χ 2/ df=2.4, CFI=0.97, TLI=0.97, root mean square error of approximation=0.06), indicating that a summary score was justified for representing the overall status of HRQOL. Conclusions Our results show that the construct validity of the 12-domain SS-QOL is well supported for measuring HRQOL in ischemic stroke patients. Thus, we recommend the 12-domain version of the SS-QOL for use in capturing the multiple impacts of stroke as well as overall HRQOL status on the basis of patients' perspectives.