A Combined Predictive and Causal Approach for Neighborhood-Level Diabetes Detection

Objective: Develop a neighborhood-level framework using machine learning and causal inference to identify socioeconomic and behavioral drivers of Type 2 diabetes for targeted public health interventions. Materials and Methods: Data from 1,149 Census Tracts in Toronto were integrated, linking demographic, health, and marginalization indices. Seven machine learning models classified neighborhoods with high diabetes prevalence. Feature engineering mitigated skewness and correlation, while Causal Forests estimated the Conditional Average Treatment Effect (CATE, τ) for predictors such as work stress, smoking, and mental health. Results: Predictive models achieved over 90% recall and high AUC metrics on both test and external validation datasets. Key predictors included obesity, overweight status, physical activity, and log-transformed median age. Causal analysis further indicated that elevated work stress (τ = 0.312) and daily smoking (τ = 0.155) increased diabetes risk, while stronger mental health (τ ≈ −1.1) was protective. Discussion: While genetic and clinical factors often dominate the conversation on diabetes, data is often restricted to confirmed diagnoses or not readily available for prevalence analyses. Our study shows how neighborhood contexts, including walkability, stress levels, and socioeconomic differences, help drive rising disease rates. We integrated machine learning classifiers with causal inference to examine how interventions, such as active transportation and adjusted work stress, could shift diabetes risk. Conclusion: This integrated method offers a blueprint for precision public health by clarifying how modifiable neighborhood factors affect diabetes risk. It can help tailor interventions to community needs and is applicable to other areas facing similar chronic disease challenges.