Modelling Diabetes Risk: Social and Behavioural Factors in BRFSS Data

This project analyses data from the 2024 Behavioral Risk Factor Surveillance System (BRFSS), a U.S. health survey of over 450,000 adults, to examine how demographic, socioeconomic, and behavioural factors relate to diabetes risk.

The goal is to identify which factors are most strongly linked to a diabetes diagnosis, and how those links change after adjusting for the others.

The project has three parts:


Overview

The key findings at a glance, with charts showing how diabetes risk varies with age, body weight, and income.

Open Overview »


Exploratory Data Analysis

Who is in the sample, how each variable is distributed, and how each one relates to diabetes before any adjustment.

Open EDA »


Logistic Regression

The full model and its results: adjusted odds ratios, model performance, diagnostics, and assumption checks.

Open Regression »