The Capacity of Self-Reported Health Measures to Predict High-Need Medicaid Enrollees
SHARE Issue Brief
Key Findings
Key Findings:
- Self-reported HNA data can be used successfully by Medicaid agencies to prospectively classify individuals by risk of high health care utilization.
- Self-reported HNA data are particularly useful in the context of building predictive models for new and returning Medicaid populations about whom the program lacks recent medical records.
Medicaid programs are increasingly adopting initiatives such as targeted case management and risk adjustment of performance benchmarks that require the prospective stratification of patients into clinically distinct subgroups. This brief discusses the potential for a short, self-reported Health Needs Assessment (HNA) screener to perform this stratification when other more comprehensive data, such as billing records culled from medical claims databases, are unavailable. Using administrative data from Wisconsin paired with nationally representative survey data, we find that HNAs meet established statistical thresholds for predictive modeling and can be used to prospectively identify Medicaid-eligible low-income adults with elevated health care needs.
More like this from Mathematica
Featured Publication
Evaluation of the Maryland Total Cost of Care Model: Progress Report
How do you apply evidence?
Take our quick four-question survey to help us curate evidence and insights that serve you.
Take our survey