Adaptation of a 3-factor model for the Pittsburgh Sleep Quality Index in Portuguese older adults

The present study examined the factor structure of the Pittsburgh Sleep Quality Index (PSQI) in a sample of older Portuguese adults using a cross-validation approach. Design is a cross-sectional. A convenience sample of 204 community-dwelling older adults (M=70.05, SD=7.15) were included. The global sleep quality (GSQ) score ranged from 0 to 18 with a mean of 5.98 (SD±3.45). The distribution showed that gender and perception of oneself as healthy influences GSQ in this sample. Cronbach's α was 0.69, but increased to 0.70 if the "use of sleep medication" component was deleted. Exploratory factor analysis (EFA) demonstrated two factor model is better than one factor, and a model fit with good indices (chi-square=8.649, df=8, p=0.373). Confirmatory factor analysis (CFA) was performed on the single factor, two factor, and three factor models, with and without the "use of sleep medications" component. The best model was the 3-factor model without the "use of sleep medications" component (chi-square=1.214, df=6, GFI=0.997, AGFI=0.918, CFI=0.986, RMSEA=0.046). The adaptation of the model is similar to the original model, with the only change being the exclusion of the "use of medications to sleep" component. We suggest using that component as a complementary qualitative assessment of health.