Association between female genital mutilation and sexually transmitted infection in sub-Saharan Africa: Evidence from interpretable weighted machine learning.

OBJECTIVE: Sexually transmitted infections (STIs) impose a heavy public health burden on low- and middle-income countries. We aimed to examine the association between female genital mutilation (FGM) and STI and develop an interpretable machine learning framework for STI identification among women in Sub-Saharan Africa. METHODS: We analyzed Demographic and Health Survey (DHS) data from 57 548 women aged 15-49 years across 10 Sub-Saharan African countries. Eighteen prespecified sociodemographic, behavioral, health-related, symptom, and cultural features were included. Data from nine countries were used for model development and internal validation, while Sierra Leone was reserved for independent external validation. Weighted logistic regression, elastic net, random forest, and XGBoost incorporated individual-level DHS sampling weights, with hyperparameters tuned using fivefold cross-validation. SHapley Additive exPlanations (SHAP) analysis was used for model interpretation, and survey-weighted multivariable logistic regression quantified adjusted associations. RESULTS: A total of 4144 women had STIs. The selected weighted random forest model showed good discrimination in both the internal validation set (area under the receiver operating characteristic curve [AUROC] = 0.863) and the independent external validation set (AUROC = 0.878). SHAP analysis identified history of abnormal genital discharge, history of genital sores, and FGM as the leading contributors to STI identification. Survey-weighted multivariable logistic regression showed that women with FGM had 82.7% higher odds of STI than those without FGM (adjusted odds ratio 1.827, 95% confidence interval 1.630, 2.049). CONCLUSION: Female genital mutilation and genital symptom histories were associated with STI. Combining sociodemographic, behavioral, and symptom information might improve STI identification and highlight the sexual and reproductive health relevance of FGM in high-burden settings.