Psychometric Evaluation of the Pittsburgh Sleep Quality Index in Korean Breast Cancer Survivors: A Confirmatory...

Background/Objectives: Poor sleep quality is a prevalent and burdensome concern among breast cancer survivors. However, its assessment relies heavily on the Pittsburgh Sleep Quality Index (PSQI), whose latent structure has shown inconsistent support across populations. This study aimed to examine the underlying factor structure and reliability of the PSQI among Korean breast cancer survivors using confirmatory factor analysis. Methods: A cross-sectional survey was conducted with 386 non-metastatic breast cancer survivors recruited from a university cancer center in South Korea. Ten competing one-, two-, and three-factor models were identified in previous studies and tested using confirmatory factor analysis with maximum likelihood estimation. Model fit was assessed with χ²/df, Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR), and model parsimony was compared using Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). Results: The mean global PSQI score was 7.46 (SD = 3.95), and 72.8% of participants were classified as poor sleepers. Among the tested model, a three-factor solution provided the best fit (χ²/df = 0.795, CFI ≈ 1.000, TLI ≈ 1.000, RMSEA ≈ 0.000, SRMR = 0.017) and achieved the lowest AIC and BIC values. This finding indicates the most favorable balance between fit and parsimony. This three-factor model delineates three distinct but related domains: perceived sleep quality, sleep efficiency, and daily disturbances. The global PSQI demonstrates acceptable reliability. Conclusions: These findings support the three-factor structure of the PSQI as the most valid representation of sleep quality among Korean breast cancer survivors. These results underscore the importance of population-specific validation of sleep measures and confirm the clinical utility of this measure as a multidimensional tool for assessing sleep in survivorship care.