A-170 Prevalence and risk factors in obesity phenotypes in a rural-urban population in a developing country

Abstract

Obesity is a chronic disease strongly associated with cardiovascular disease and type 2 diabetes; its prevalence has increased in recent years due to changes in diet and lifestyle. Currently, there are different obesity phenotypes related to the amount of body fat and the presence or absence of metabolic abnormalities. These are classified as: metabolically healthy normal weight (MHNW), metabolically unhealthy normal weight (MUNW), metabolically healthy obesity (MHO), and metabolically unhealthy obesity (MUO). On the other hand, the PERU MIGRANT study is a cohort of 989 subjects over 30 years of age living in rural or urban areas. The aim of study is to determine the prevalence and risk factors of the obesity phenotypic profile in a rural-urban population using data from the PERU MIGRANT study. Secondary data analysis was performed using a cross-sectional, retrospective design. Subjects with complete sociodemographic and metabolic data were included, while those with diabetes and/or hypertension were excluded. The sample size consisted of 820 subjects, and the outcome variable was obesity phenotype, defined according to the Adult Treatment Panel III (ATP III) criteria and body mass index. Descriptive and inferential analyses were performed using the Chi-square, Kruskal-Wallis, and Dunn's post-hoc tests. Additionally, a robust multinomial regression with standard errors, adjusted for sex and urbanization, was conducted to obtain the relative risk ratio (RRR) with its respective 95% confidence intervals (95% CI). The analysis was performed using Stata version 16. A total of 820 subjects were included, of whom 53.05% (n = 435) were female, 78.66% (n = 645) resided in an urban area, and 45.58% (n = 371) had a high level of physical activity. Regarding obesity phenotypes, 15.73% (n = 129) presented a MUO phenotype, and 40.85% (n = 335) were MHO. In the bivariate analysis, obesity phenotypes were associated with sex (p<0.0001); women with the MUO phenotype represented 22.9% (n = 100), while males represented 7.53% (n = 29). Likewise, an association was found with the level of urbanization (p<0.001). In rural areas, the MHNW phenotype predominated at 78.29% (n = 137), while in urban areas, the MHO phenotype was more prevalent at 47.44% (n = 306) and 19.22%, respectively. Physical activity (p <0.001) and levels of high-sensitivity C-reactive protein (p <0.0001) were also associated with the phenotype. In the multinomial model, urbanization was found to be significantly associated with a higher risk of MHO phenotype (RRR = 7.25; 95% CI 4.68–11.23; p < 0.001) and MUO phenotype (RRR = 18.41; 95% CI 7.17–47.24; p < 0.001) compared to MHNW, and male sex was associated with a lower risk of MUO (RRR = 0.22; 95% CI 0.13–0.36; p < 0.001). The predominant phenotype was MHO, and living in an urban area increases the risk of developing both the MHO and MUO phenotypes. On the other hand, physical activity didn´t have a significant influence when adjusted to the multinomial model.