Predictive Modelling of Health Status Awareness Among Academic Staff of Tertiary Institutions in Kogi State, Nigeria Using Random Forest

Abstract

This study investigates the level of health status awareness among academic staff in selected tertiary institutions in Kogi State, Nigeria. The research employed both descriptive statistical methods and machine learning techniques specifically the Random Forest classification algorithm to identify key factors influencing health awareness. A structured questionnaire was administered to 316 academic staff, capturing variables such as age, gender, marital status, academic rank, frequency of health checks, access to health facilities, and presence of chronic conditions. Descriptive analysis revealed that a significant proportion of staff engage in regular health checks and physical activity, suggesting moderate to high health awareness levels. The Random Forest model demonstrated strong predictive performance, with an accuracy of 87.2%, precision of 84.5%, recall of 89.3%, and an F1 score of 86.8%. Feature importance analysis showed that frequency of health checks, age group, and academic rank were the most influential predictors of health awareness. The findings underscore the role of individual health behaviours and demographic characteristics in shaping awareness. The study recommends institutional health campaigns targeted at junior academic staff and enhanced access to health facilities. It concludes that machine learning models offer a reliable approach for profiling health awareness and guiding evidence-based interventions in tertiary institutions.