Predictors of insomnia in Peruvian college students: A regression analysis from a Bayesian approach

This study aimed to determine the prevalence, associated factors, and predictors of insomnia in Peruvian college students. A cross-sectional study was conducted with 501 students aged 18 to 27 years (56.7% women). The Athens Insomnia Scale (AIS-5) was used, and descriptive analyses, chi-square tests, and a Bayesian ordinal logistic regression model were performed. The overall prevalence of insomnia was 82.8%, with 25.95% of cases at a high level, 56.89% at a moderate level, and 17.16% at a minimal level. The variables significantly associated with the level of insomnia were age, academic year, time spent using electronic devices, perceived stress level, and time of day when studying. In the regression model, perceived stress and preference for studying at night were significant predictors. It is concluded that these two predictors increase the likelihood of higher levels of insomnia.