Abstract
The coexistence of two or more morbidities, including at least one mental morbidity, is defined as mental-physical multimorbidity (MP-MM). It is linked to significant poor outcomes, such as a high burden of healthcare utilisation, particularly in the elderly. To evaluate the complex connections between the 16 physical and mental morbidities among Brazilian older people from the Brazilian Longitudinal Study of Ageing, we performed a network analysis (NA), a sophisticated multivariate statistical technique to estimate all relationships between morbidities represented by an undirected grafus. The objective was to estimate patterns in a complex set of multiple aleatory variables and display them in a network map within nodes and edges representing the variables and the interrelationships among them. In this study, we applied the NA to model interrelationships among chronic physical morbidities and depression. We utilised data from 6.104 participants of the second wave (2019-2020) of the Brazilian Longitudinal Study of Ageing (ELSI-Brazil). The data were adjusted according to the Ising model with the IsingFit function by R Software. Centrality and stability measures were assessed by the bootstrap method through the bootnet library. In this network, depression, low back pain, and hypertension were the morbidities that had the most effects on the network's overall structure, according to an examination of the centrality metrics of the nodes (strength, proximity, and betweenness). Depression was the morbidity with the higher betweenness. The model's interpretation indicates that depression is the illness that has the highest influence on the model and would likely be the most beneficial area for intervention.