Sign in Subscribe
Medical research

Machine Learning Models Predict Cardiovascular Events in Older Adults with Type 2 Diabetes

Strategic Insights

— 1 min read

Researchers from Brigham and Women's Hospital and Harvard Medical School have developed and validated machine learning models to predict cardiovascular events in older adults with type 2 diabetes. The study used linked Medicare claims and electronic health records data from 2013 to 2020 to develop and test the models. The researchers found that the models accurately predicted risk in both patients with and without existing heart disease, and that claims-only models can be useful for researchers studying treatments in large healthcare databases.

Key Takeaways:

  • Researchers developed and validated machine learning models to predict cardiovascular events in older adults with type 2 diabetes using linked Medicare claims and electronic health records data.
  • The study included 14,776 patients with baseline CVD and 10,679 without baseline CVD, with a mean age of 77 and 74 years, respectively.
  • Claims-only models achieved a c-statistic of 0.75 and a Brier score of 0.09 in patients with baseline CVD, while in those without baseline CVD, the c-statistic was 0.73 and the Brier score was 0.01.
  • Adding EHR data did not improve performance compared to using claims data alone.
  • The models predicted one-year cardiovascular outcomes with good discrimination and accuracy, independently of CVD history.
  • The study concluded that these models can help identify people at higher risk, guide research on diabetes medications, and support better planning for healthcare resources.

Statistics:

  • 14,776 patients with baseline CVD
  • 10,679 patients without baseline CVD
  • Mean age of 77 years for patients with baseline CVD
  • Mean age of 74 years for patients without baseline CVD
  • C-statistic of 0.75 for claims-only models in patients with baseline CVD
  • C-statistic of 0.73 for claims-only models in patients without baseline CVD
  • Brier score of 0.09 for claims-only models in patients with baseline CVD
  • Brier score of 0.01 for claims-only models in patients without baseline CVD

Sources:

  • Machine Learning For Predicting Cardiovascular Events In Older Adults With Type 2 Diabetes Using Medicare Claims And Electronic Health Records. Journal of Clinical Epidemiology, 2025:112001.
  • Managed Care Weekly Digest, October 13, 2025; p 131.

Read more