Accuracy of mild traumatic brain injury case ascertainment using ICD-9 codes

Objectives: To determine the accuracy of mild traumatic brain injury (TBI) case ascertainment using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes proposed by the Centers for Disease Control and Prevention (CDC) in a 2003 Report to Congress. Methods: This was a prospective cohort study of all patients presenting to an urban academic emergency department (ED) over six months in 2003. A real-time clinical assessment of mild TBI was compared with the ICD-9 codes assigned after ED or hospital discharge for a determination of sensitivity and specificity. Results: Of the 35,096 patients presenting to the ED, 516 had clinically defined mild TBI and 1,000 were assigned one or more of the mild TBI ICD-9 codes proposed by the CDC. The sensitivity of these codes was 45.9% (95% confidence interval [95%, CI] = 41.3%, to 50.2%) with a specificity of 97.8% (95% CI = 97.6% to 97.9%). Conclusions: The identification of mild TBI patients using retrospectively assigned ICD-9 codes appears to be inaccurate. These codes are associated with a significant number of raise-positive and false-negative code assignments. Mild TBI incidence and prevalence estimates using these codes should be interpreted with caution. ICD-9 codes should not replace a clinical assessment for mild TBI when accurate case ascertainment is required.