UPMC is at the forefront of an emerging field called predictive health analytics, which aims to improve patients’ health care outcomes and contain costs. The Pittsburgh health plan has developed prediction models that analyze data such as patient claims, prescriptions and census records to determine which members are most likely to use the most emergency and urgent care, which can be expensive.
UPMC recently bolstered its forecasting models with details including members’ household incomes, education levels, marital status and number of children at home. While the plan has not yet acted on the correlations it found in the household data, it segments its members into different market baskets based on analysis of more traditional data sets. Then it assigns care coordinators to certain members flagged as high risk because of chronic conditions that are not being properly treated.