Exploring patterns and insights in a comprehensive diabetes dataset

Abstract

Diabetes, a pressing global health concern, imposes significant economic and healthcare burdens on millions worldwide. Understanding the multifaceted factors contributing to diabetes is pivotal for effective prevention strategies. In this study, we leverage a comprehensive dataset (952 instances, 17 predictor variables) and employ a multifaceted statistical approach to explore the intricate interplay among stress levels, blood pressure (BP), body mass index (BMI), age, sleep quality, and physical activity in relation to diabetes, with a focus on classification and predictive implications. Our research begins by establishing fundamental relationships between discrete variables using crosstabs and chi-square tests. We uncover close associations between stress levels and BP, heightened diabetes risk with increased BMI values, and the influence of age on sleep quality. Subsequent analysis, based on descriptive data, reveals a robust correlation between physical activity and stress levels, with the paradoxical observation that excessive exercise may increase stress levels. Factor analysis further elucidates the pivotal roles of sound sleep and regular exercise in diabetes prevention, supported by asymptotic significance levels below 0.05. To culminate our study, we construct a logistic regression model with an impressive 89.3% accuracy rate for predicting diabetes risk. Notably, age, family history of diabetes, and regular medication usage emerge as the most influential factors, with regular medication demonstrating significant potential for reducing diabetes risk. Our research underscores the intricate web of factors shaping individual health and offers valuable insights for a comprehensive understanding of health and well-being in the context of diabetes prevention. Moreover, it highlights the importance of considering multiple factors in health-related research. Future research could delve into the long-term effects of interventions targeting the identified risk factors, explore the impact of socio-economic factors on diabetes risk, and investigate the potential role of emerging technologies in personalized diabetes prevention strategies.