Intersectional disparities in clinical obesity in Peru: A MAIHDA analysis of national survey data

Objectives: To examine the distribution of clinical obesity across intersectional social strata in Peruvian adults and evaluate heterogeneity in prevalence across combinations of social positions using a Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) approach. Materials and methods: Cross-sectional study using pooled data from the Peruvian Demographic and Health Survey (ENDES) 2021–2024. Clinical obesity was defined following the Lancet Diabetes & Endocrinology Commission criteria. Seventy-two intersectional strata were constructed by crossing sex, age group, household wealth, area of residence, and self-perceived ethnicity. Two-level logistic regression models and stratum-specific absolute risk due to interaction (ARDI) were estimated. Results: A total of 112,319 adults ≥20 years were analyzed. The estimated prevalence ranged from approximately 1% to 49% across strata, was higher among older adults, women, wealthier households, urban residents, and non-native individuals. Most between-stratum differences were captured by the additive main effects of the strata-defining variables, consistent with a modest role for intersectional interactions. Nonetheless, synergistic effects were concentrated among older women, with non-native poor urban women aged ≥60 years exhibiting the greatest excess burden beyond additive expectations (ARDI=+14.87 percentage points). Conclusions: Clinical obesity in Peru is unequally distributed across intersectional social strata. While additive effects predominate, intersectional analysis identifies priority subgroups whose burden exceeds additive predictions, underscoring the value of MAIHDA for informing targeted public health strategies.