Predictive ability of craniofacial anthropometry for autism spectrum disorder in a Nigerian Hausa population

Abstract

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by impairments in social interaction, communication, and repetitive behaviours. Craniofacial anthropometry has been suggested as a potential non-invasive tool for identifying phenotypic variations associated with ASD. This study evaluated the predictive ability of craniofacial anthropometric parameters for ASD among Hausa population in Kano State, Nigeria. A convenience sampling technique was used to recruit 48 participants comprising 24 ASD subjects and 24 age-matched controls aged 8 – 19 years. Craniofacial measurements were obtained using standard anthropometric techniques, and binary logistic regression analysis was employed to determine the predictive ability of the variables. The result of the study revealed that for the combined male and female participants, only the facial height/width ratio significantly predicted ASD (β = −13.93, p = 0.01), with the highest predictive strength (Cox & Snell R² = 0.189; Nagelkerke R² = 0.252) and an overall classification accuracy of 64.58%. Face height, head circumference, face width, head length, head width, head length/width ratio, and philtrum height were not significant predictors (p > 0.05). Sex-specific analysis showed that facial height/width ratio was the only significant predictor among males (β = −23.19, p = 0.01), with 69.4% classification accuracy, whereas head width was the only significant predictor among females (β = 0.25, p = 0.05), with 75.0% classification accuracy. Overall, the models demonstrated moderate predictive ability. The study concludes that facial height/width ratio and head width may serve as useful supplementary anthropometric markers in predicting ASD among Hausa adolescents in Kano State, Nigeria.