Abstract
Based on a national representative sample of the population aged 50 years or older, this study aimed to estimate the prevalence of frailty among men and women, identify associated sociodemographic and health factors, and estimate the population attributable fraction. Data from the second wave (2019-2021) of the Brazilian Longitudinal Study of Aging (ELSI-Brazil) were used. Frailty was classified based on the number of positive items among unintentional weight loss, exhaustion, low level of physical activity, slow gait, and weakness. The main analyses were based on multinomial logistic regression stratified by sex. The prevalence of frailty was lower in men (8.6%; 95%CI: 6.9; 10.7) than in women (11.9%; 95%CI: 9.6; 14.8), with the most frequent item being the low level of physical activity in both. Age and schooling level were the sociodemographic factors associated with pre-frailty and fragility among men and women. The population attributable fraction was different for frailty between genders. In men, the highest population attributable fraction was due to not having a partner (23.5%; 95%CI: 7.7; 39.2) and low schooling level (18.2%; 95%CI: 6.6; 29,7). In women, higher population attributable fraction values were due to memory deficit (17.1%; 95%CI: 7.6; 26.6), vision deficit (13.4%; 95%CI: 5.1; 21.7), and diabetes mellitus (11.4%; 95%CI: 4.6; 18,1). Similar population attributable fraction levels were observed for heart disease (8.9%; 95%CI: 3.8; 14.1 in women and 8.8%; 95%CI: 2.0; 15.6 in men). Strategies aimed at physical activity have the potential to prevent frailty in both men and women, and the prevention of chronic conditions is more important in women.