[Multimorbidity and use of health services by individuals with restrictions on habitual activities: the Pró-Saúde Study].

The study aimed to assess the use of health services and the association with different measures of multimorbidity. This was a cross-sectional study nested in the Pró-Saúde Study, a longitudinal study of municipal technical and administrative employees in Rio de Janeiro, Brazil. Data were analyzed from phase 2 (2001-2002), and the study population consisted of 733 individuals who reported restrictions on habitual activities due to health problems in the 15 days prior to the data collection. The search for a health service (outcome variable) was used as the proxy for use of health services. Multimorbidity was assessed by simple count and the Cumulative Illness Rating Scale, generating four exposure variables: number of self-reported diseases, multimorbidity (2 or more diseases), and total score and number of systems affected. The analyses stratified by sex used Poisson regression models with robust variance, adjusted by age and schooling. Women showed higher mean values than men for all the measures, with 51% classified as having multimorbidity. In men, multimorbidity increased by 43% (95%CI: 1.11-1.84) the probability of using health services, while there was no statistically significant association in women. For men, each additional disease increased the probability of use of a health service by 14% (95%CI: 1.05-1.24). There were evident differences in the use of health services and multimorbidity according to sex. Explaining these patterns becomes relevant for the provision of efficient, coordinated, and safe care for persons with multimorbidity.