Predictive power of a body shape index (ABSI) for diabetes mellitus and arterial hypertension in Peru: demographic and health survey analysis - 2020

• Autores: Torres Malca, Jenny Raquel, Ojeda Heredia, Andony, Zuzunaga-Montoya, Fiorella Elvira, Vera-Ponce, Victor Juan, Cruz-Ausejo, Liliana, De la Cruz-Vargas, Jhony A.
• Formato: artículo
• Fecha de Publicación: 2022
• Institución: Universidad Tecnológica del Perú
• Repositorio: UTP-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.utp.edu.pe:20.500.12867/6359
• Enlace del recurso: https://hdl.handle.net/20.500.12867/6359; https://doi.org/10.6000/1929-6029.2022.11.14
• Nivel de acceso: acceso abierto
• Materia: Diabetes mellitus; Hypertension; Anthropometry; https://purl.org/pe-repo/ocde/ford#3.00.00

Introduction: Given the relationship between obesity and type 2 diabetes mellitus (T2DM) and hypertension, an indicator of body fat, A Body Shape Index (ABSI), has been considered to have apparent predictive power for these diseases.Objective: To determine the predictive power of the ABSI for DMT2 and hypertension in Peru through the analysis of the Demographic and Health Survey-2020 (ENDES-by its acronym in Spanish-2020).Methods: Cross-sectional analytical study of the ENDES-2020. The variables evaluated were ABSI, body mass index, high abdominal waist, waist-to-height ratio, body roundness index (BRI) and conicity index (COI). Areas under the curves (AUC) together with their 95% confidence interval (95%CI) were used to present each index.Results: A total of 19 984 subjects were studied. Regarding hypertension, the highest AUC was presented by the COI: AUC=0.707 (95%CI 0.694-0.719). While the ABSI obtained the penultimate place: AUC=0.702 (95% CI 0.689-0.715). In case of DM2, the highest ABC was presented by BRI: AUC=0.716 (95%CI 0.689-0.743); while ABSI obtained the second place: AUC=0.687 (95%CI 0.658-0.717). Conclusions: The results demonstrate that ABSI is not a good predictor for hypertension and DMT2 in the Peruvian population. If these findings are confirmed by other studies, its use would not be recommended for these diseases, and other anthropometric indicators that could perform better should be further explored.