Identifying individuals at risk for weight gain using machine learning in electronic medical records from the United States
Authors' conclusion
The current study demonstrates an ML approach applied to EMR to identify those at risk for weight gain over 12 months. Providers may use this risk stratification to prioritise prevention strategies or earlier obesity intervention.
Study context
- Population
- Adults in Optum's de-identified Market Clarity Data (US Electronic Medical Records (EMR) database)
- Comparison
- XGBoost vs random forest, simple logistic regression, Elastic Net and LASSO
- Outcome
- ≥10% increase in index BMI; model performance by AUC
- Follow-up
- 12 months
Methodological detailselectronic medical records, US database
- Electronic medical records
- US database
Performance
As reported in the abstract
Prediction of ≥10% increase in index BMI · 12 months post-index
| Test or model | AUC |
|---|---|
| XGBoost | 0.75 |
XGBoost
AUC0.75
Further findings
Top predictors of ≥10% increase in index body mass index (BMI)
- Lower baseline BMI, any ED visit during study period, no DM, no lipid disorders and younger age were among top predictors
Original publication
Identifying individuals at risk for weight gain using machine learning in electronic medical records from the United States. Diabetes Obes Metab. Published 11 March 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →