Sex determination using foot dimensions and bimalleolar breadth: A cross-sectional study of a Nigerian population

Background Forensic identification of human remains in mass disasters often relies on body parts like the foot, which are frequently protected and preserved by footwear. Due to population-specific variations in anthropometry, locally derived standards are necessary for accurate biological profiling. This study aimed to investigate the accuracy of sex estimation using foot dimensions and bimalleolar breadth in a Nigerian university sample. Methods This cross-sectional study was conducted among 816 undergraduate students (409 males, 407 females) aged 18 - 25 years at Delta State University, Nigeria. Bilateral measurements of foot length, foot breadth, and bimalleolar breadth were obtained using a digital Vernier caliper, and the foot index was derived. Data were analysed using independent t-tests, Discriminant Function Analysis (DFA) with leave-one-out cross-validation, binary logistic regression, and Receiver Operating Characteristic (ROC) curve analysis to evaluate sexual dimorphism, predictive classification accuracy, and diagnostic validity. Results Males had significantly higher mean foot length, foot breadth, and bimalleolar breadth across both feet than females (p < 0.001), while foot index showed no significant sexual dimorphism. In univariate DFA, left-sided variables generally demonstrated higher sex classification accuracy than right-sided variables, with Left Foot Length emerging as the best single predictor (83.6%; AUC = 0.934, 95% CI: 0.915–0.953). The highest accuracy was obtained using a stepwise multivariate model combining Left Foot Length, Left Foot Breadth, and Left Bimalleolar Breadth, achieving 85.9% cross-validated accuracy. Conclusion Foot dimensions, especially multivariate models incorporating bimalleolar breadth, show moderately high accuracy for sex estimation in the studied population and offer practical adjunctive forensic tools requiring broader validation.