Reliability and validity of a system for coding asthma outcomes from medical records

To evaluate the reliability and validity of a standardized asthma outcome coding system, we obtained medical records for 182 asthmatic children. Records were coded by trained staff using explicit and detailed criteria. Outcome variables coded included number of corticosteroid bursts, asthma-related physician contacts, emergency room visits, hospitalizations, and number of asthma episodes. Interrater reliability was excellent. Patterns of associations between the coded variables and other independently obtained outcome measures supported concurrent and construct validity. Given the intense scrutiny of health outcomes in the current managed-care marketplace, use of this system may foster further clinical research examining asthma outcomes.