Computer-Based Models to Identify High-Risk Adults with Asthma: Is the Glass Half Empty or Half Full?

Abstract

This study developed and evaluated the performance of prediction models for asthma-related adverse outcomes based on the computerized hospital, clinic, and pharmacy utilization databases of a large health maintenance organization. Prediction models identified patients at three- to four-fold increased risk of hospitalization and emergency department visits, and were valid for test samples from the same population. A model that identified 19% of patients as high risk had a sensitivity of 49%, a specificity of 84%, and a positive predictive value of 19%. We conclude that prediction models that are based on computerized utilization data can identify adults with asthma at elevated risk, but may have limited sensitivity and specificity in actual populations.