The study aimed to verify the efficacy of expressive therapies (ET) in sleep disorders in hyperfrequent elderly (HE) of Primary Health Care (PHC). Quasi-experimental study, with 69 elderly people assisted at PHC in a metropolitan region of the Midwest, divided into two groups: intervention (elderly with high attendance) and control (low-frequency elderly). Sleep questionnaires were used: Pittsburgh Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), Insomnia Severity Index (ISI) and STOP-bang (SB), in addition to the investigation of sociodemographic variables, anthropometric and health service-related. For data analysis, chi-square tests, multivariate analysis of variance and Wilks' Lambda test were used, considering p ≤ 0.05. The intervention group (IG) showed a decrease in PSQI scores (p = 0.003), ESE (p = 0.006), IGI (p <0.001), SB (p = 0.002) with significant differences between groups. ET were effective in the female and male IG, attenuating sleep disorders. Thus, by reducing sleep disorders in the group of HE people with the use of non-pharmacological intervention in PHC, it is possible to improve sleep quality and, consequently, reduce the use of health services, reducing financial costs for the health system public.