Fatores associados à mini avaliação nutricional e sua relação com desfechos de saúde em idosos após quatro anos de acompanhamento: estudo longitudinal de base populacional

Abstract

This study investigated, after four years of follow-up, the factors associated with the Mini Nutritional Assessment (MNA) and its relationship with health outcomes in community-dwelling older adults. This is a longitudinal, population-based study, based on a preliminary (2018-2019) and current (August 2019 to July 2023) study, with a sample of 584 older adults attending the Basic Health Unit in the municipality of Santana do Deserto-MG. Validated questionnaires and instruments covering sociodemographic aspects, previous medical history, cognitive assessments, physical conditions, and behavioral and modifiable factors were used. Data were analyzed using R software version 4.4.3, and descriptive statistics were performed using relative and absolute frequencies (categorical variables) and mean with standard deviation (continuous variables). Comparison between baseline and four-year follow-up of categorical (McNemar test) and continuous (paired t-tests) variables was performed to verify longitudinal differences. To identify baseline predictors of MNA scores after four years of follow-up, multivariate linear regression models were performed, including baseline MNA as a lagged variable. Unadjusted linear (continuous variables) and logistic (dichotomous variables) regression models were performed to assess the role of baseline MNA scores as predictors of several four-year outcomes (mental health, social support, quality of life (QOL), hospitalization, comorbidities, and functioning). The goodness-of-fit of the linear models was assessed using R2 , and a p < 0.05 was considered significant. After four years of follow-up, the mean age of participants was 73.1 7.04 years, with a prevalence of females (53.4%, n = 312); married (50.7%, n = 296), non-white ethnicity (53.8%, n = 314), and only 8.7% (n = 51) of the elderly were at risk of malnutrition. The comparison between baseline and after four years of followup revealed a significant worsening in cognition, QOL (social, environmental, and physical), functionality, and social support, in addition to improvements in mental health, loneliness, psychological QOL, and nutrition scores. At four years of follow-up, a higher baseline MNA score was associated with higher MNA scores, social support, functioning, and all QOL domains, while being associated with lower levels of loneliness and mental health (depression, anxiety, and stress). Baseline MNA was only able to predict MNA itself and environmental QOL (p = 0.001). Some predictors showed a trend but did not reach significance, such as depression (p = 0.082), anxiety (p = 0.062), and social QOL (p = 0.107). Thus, it can be concluded that the MAN proved to be comprehensive in relation to health outcomes in elderly people living in the community, as it allowed inferences about the various factors observed in the aging process, which are expressed in physiological, psychological and social changes, affecting the autonomy and QoL of this population.