Predicting high-cost privately insured patients based on self-reported health and utilization data

Area of Science:

• Health Services Research
• Health Economics
• Predictive Analytics

Background:

• Innovative care delivery models aim to lower healthcare costs and improve quality through intensive patient management.
• Accurate identification of high-cost patients is essential for the success of these models.
• Predicting future healthcare expenditures is crucial for resource allocation and intervention strategies.

Purpose of the Study:

• To evaluate the predictive accuracy of self-reported health data for future high healthcare costs.
• To compare the predictive power of self-reported measures against claims-based data.
• To determine the utility of patient-reported information in identifying high-cost individuals within a privately insured population.

Main Methods:

• A 2012 mail survey of 3983 privately insured autoworkers was linked to their 2012-2013 health insurance claims data.
• High healthcare costs were defined as expenditures at or above the 75th percentile.
• Predictive models incorporating claims-based and self-reported measures were compared using the C statistic.

Main Results:

• Claims-based measures of prior healthcare costs and comorbidity were strong predictors of future high costs (C statistic = 0.78).
• Self-reported data on chronic conditions, health status, health behaviors, and hospital use also demonstrated predictive value.
• Models including self-reported measures were less accurate than claims-based models, with the most comprehensive model achieving a C statistic of 0.73.

Conclusions:

• While claims data provide superior prediction of future healthcare costs, patient-reported measures offer a valuable alternative when claims data are unavailable.
• Self-reported health status and prior healthcare utilization are reasonable predictors for identifying high-cost patients in privately insured populations.
• Patient-reported data can support targeted care management strategies to control costs and enhance quality.