Analysis of Diagnostic Errors in Emergency Medicine Using a Descriptive Medical Error Registry

Author: Nnaemeka G. Okafor, MD

Primary Advisor: James P. Turley, PhD, RN

Committee Members: Jiajie Zhang, PhD

Masters thesis, The University of Texas School of Biomedical Informatics at Houston.


Emergency departments (ED) are the sites of 51 – 82% of preventable medical errors, most of which are diagnostic errors. A diagnostic error is a missed or an unintentionally delayed diagnosis. Diagnostic errors result in increased healthcare costs as well as significant patient morbidity or death. Research into diagnostic errors is limited. Incident reporting systems have been suggested as a mechanism to identify active and latent errors. This project utilizes an incident reporting system and a medical error registry to study diagnostic errors. The goal of this project was to identify patient encounter elements commonly associated with diagnostic errors in the ED and analyze those diagnostic errors to detect trends & relationships of the contributing factors. A total of 84 patient encounters were identified with 90 mutually exclusive medical errors. Diagnostic errors accounted for 55% of the errors. The most commonly noted patient-encounter elements were: adult patients, patient arrival time interval of 3pm – 6pm, day of arrival: Tuesday, Friday, or Sunday and a discharge or ICU admission disposition. The identified contributing factors were grouped as system/process, cognitive or non-remediable factors. The predominant system/process contributing factors were high workloads, test or treatment processing delay, ineffective patient handoffs and lack of departmental policies regarding condition evaluation and management. The predominant cognitive contributing factors were faulty information processing and information verification while the non-remediable contributing factors were limited patient historians and atypical patient presentations. Though this project was subject to hindsight, recall and reporting biases, it demonstrated that the combination of a clinical incident reporting system, a revised incident review process and a descriptive medical error registry was a feasible method to monitor the frequency and trends of diagnostic errors in the ED and may suggest future areas to focus to patient safety interventions and research.