Psychometric Properties of the Arabic Version of the Pittsburgh Sleep Quality Index in Hemodialysis Patients

Purpose The Pittsburgh Sleep Quality Index (PSQI) is widely used in screening for self-reported sleep disturbances in clinical groups or healthy populations. The literature lacks rigorous psychometric evaluation of the Arabic Pittsburgh Sleep Quality Index (A-PSQI) in hemodialysis (HD) patients. This study aims to estimate reliability, extract, and test the underlying factor structure of the A-PSQI in HD patients, and determine its suitability for this clinical group. Methods In a sample of 461 HD patients (287 males, 62.3%) recruited from all health districts in Kuwait., inter-components, between A-PSQI components, and its global score correlations were estimated. Component analysis was conducted to identify inconsistent components with the rest. Cronbach’s α and McDonald’s ω estimated reliability. Stability and convergent validity were assessed. Construct validity was explored by exploratory factor analysis (EFA). Confirmatory factor analysis (CFA) was conducted to test factor structure. Results Patients had a median (Interquartile Range) age of 54 (22) years. The A-PSQI had a mean (SD) score of 7.1 (3.65). The Cronbach α = .634 and McDonald ω = 0.62 indicated moderate reliability. EFA resulted in two factors highly correlated (0.83, PV < 0.001) with A-PSQI global score indicating two distinct constructs. CFA provided weak evidence supporting the one-factor model, but the two-factor model (explained 51.8% of total variance) without the “use of sleep medication” item, which has the smallest corrected-item total correlation, showed a good fit with the smallest AIC. Conclusion The A-PSQI seems moderately reliable to screen for sleep disorders in HD patients. Future research on scoring in the two-factors model is warranted. It can help clinicians better understand sleep disturbance and improve the instrument’s sensitivity.