# Gender-Based Comparative Study of Type 2 Diabetes Risk Factors in Kolkata, India: A Machine Learning Approach

**Authors:** Rahul Jain, Anoushka Saha, Gourav Daga, Durba Bhattacharya, Madhura Das Gupta, Sourav Chowdhury, Suparna Roychowdhury  
**arXiv:** [2311.10731](https://arxiv.org/abs/2311.10731)  
**Date:** 2023-10-15  
**Summary page:** https://www.summarizepaper.com/en/paper/2311.10731v1/

## Key points

- Type 2 diabetes mellitus is a significant global health concern
- Understanding risk factors is crucial for prevention and management
- Study conducted in Kolkata, West Bengal, India
- Aimed to investigate gender-specific differences in risk factors
- Sample consisted of 428 patients from Belle Vue Clinic
- Machine learning models used to predict risk of diabetes
- Age-related increase in risk for both males and females
- Differences between genders regarding other risk factors
- Exercise and BMI significant for males, but not females
- Waist to Height Ratio (WhtR) models performed better than BMI models
- Importance of considering waist circumference relative to height as an indicator of risk
- Findings can inform targeted healthcare interventions and public health strategies

## Summary

Type 2 diabetes mellitus is a significant global health concern and understanding its risk factors is crucial for effective prevention and management. In this study conducted in Kolkata, West Bengal, India, researchers aimed to investigate the gender-specific differences in the impact of age, lifestyle, BMI (Body Mass Index), and Waist to Height Ratio (WhtR) on the risk of Type 2 diabetes mellitus. The sample consisted of 428 patients from the outpatient department of Belle Vue Clinic. Various machine learning models including Logistic Regression, Random Forest, and Support Vector Classifier were employed to predict the risk of diabetes based on different predictors. The performance of these models was compared to determine their effectiveness in assessing the risk. The findings revealed a significant age-related increase in the risk of diabetes for both males and females. However, there were notable differences between genders regarding other risk factors. While exercising and BMI were found to have a significant impact on the risk of Type 2 diabetes in males, they turned out to be statistically insignificant for females. Interestingly, predictive models based on WhtR demonstrated superior performance in assessing the risk compared to those based on BMI for both males and females. This highlights the importance of considering waist circumference relative to height as a more accurate indicator of diabetes risk. Overall, this study provides valuable insights into gender-specific differences in Type 2 diabetes risk factors. These findings can inform targeted healthcare interventions and public health strategies aimed at preventing and managing this prevalent disease.