Abstract Introduction An increasing number of studies have explored the link between sleep disturbances and erectile dysfunction (ED). This work reports findings from the 2015 follow-up of the Epidemiologic Sleep Study (EPISONO) conducted in São Paulo, Brazil, comparing results from its 3rd and 4th editions (2007 and 2018, respectively). We examined the incidence and prevalence of ED and their association with potential determinants, including serum testosterone and obstructive sleep apnea (OSA). Methods Participants completed full-night polysomnography and standardized health questionnaires. Serum testosterone was also measured. The cohort included men aged 20-80 years. Longitudinal analyses calculated the incidence of ED and generalized estimating equation models were performed. Data from EPISONO 2018 were analyzed cross-sectionally through binomial logistic regression and mediation–moderation modeling. Results During follow-up, the incidence of ED was 10.55%, predominantly among men older than 50 years. In the longitudinal model, advancing age (odds ratio OR=1.09), depressive symptoms (OR=1.05), and reduced total testosterone (OR=2.69) significantly predicted ED, whereas higher psychological well-being (World Health Organization Quality of Life; OR=0.87) was protective. In the 2018 dataset, the overall prevalence of ED reached 20.06%, again higher among participants over 50. The likelihood of ED increased with age (OR=1.07), while better psychological quality of life reduced this odd (OR=0.65). Mediation analysis showed a significant indirect effect of the apnea-hypopnea index (AHI) on the relationship between age and total testosterone. When ED was set as the outcome, age retained a direct effect, but AHI did not. Conclusion These findings underscore the relevance of aging, depressive symptoms, testosterone levels, and psychological quality of life in the development of ED. Although it was associated with OSA, this link did not appear independent of age. From a longitudinal perspective, modifiable factors, such as mood, sleep quality, and hormonal status, represent key targets for preventive interventions to reduce ED risk over time. Support (if any) Associação Fundo de Incentivo à Pesquisa (AFIP), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; grants 2020/13467-8 and 2021/05920-7).
Andersen et al. (Fri,) studied this question.