MACHINE LEARNING APPLIED TO MEDICARE CLAIMS DATA TO IDENTIFY OLDER ADULTS WITH UNDIAGNOSED DEMENTIA
2024

Using Machine Learning to Find Older Adults with Undiagnosed Dementia

Sample size: 837440 publication Evidence: moderate

Author Information

Author(s): Barnes Deborah, Benjamin Cynthia, Mannen Delia, Boscardin John

Primary Institution: University of California San Francisco

Hypothesis

Can machine learning methods effectively identify older adults with undiagnosed dementia using Medicare claims data?

Conclusion

Machine learning can accurately identify older adults at risk of undiagnosed dementia using Medicare claims data.

Supporting Evidence

  • Approximately half of people living with Alzheimer’s disease and related dementias are undiagnosed.
  • On average, 24,410 (2.9%) of Medicare beneficiaries were diagnosed with incident dementia each year.
  • Model accuracy for detecting undiagnosed dementia ranged from 0.77 to 0.81.

Takeaway

Researchers used computer programs to look at health records and found ways to spot older people who might have dementia but don't know it.

Methodology

A retrospective cohort study using the 5% Medicare claims sample to identify predictors of undiagnosed dementia.

Limitations

The study is retrospective and may not account for all variables affecting dementia diagnosis.

Participant Demographics

Beneficiaries aged 65 years or older.

Digital Object Identifier (DOI)

10.1093/geroni/igae098.2276

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication