Abstract P270: Automating Processes to Expedite Medical Record Review in Research

Abstract

Problem. Heart failure (HF) clinical trials often use medical record review data to determine HF related hospital events, such as emergency room visits and readmissions. However, reviewing medical records for each hospital event can be daunting. The purpose of this project was to expedite medical record review processes using an automated method for identifying potential HF related hospital events. Methods. A stepwise procedure was developed for obtaining electronic information of all consented HF clinical trial participants' hospital events. First, an electronic list of all study participants was generated from a study data base and electronically sent to the Medical Records department each month. Next, a list of all emergency room visits and hospitalizations, including admission and discharge dates, discharge disposition, and diagnosis and procedure codes was electronically generated by the Medical Records department. A trained cardiovascular research nurse reviewed the abstracted coding and billing data for each hospital event and identified specific codes potentially related to HF. Last, hospital events identified as meeting specific criteria were included in the medical record review. Results. Using the automated system, a total of 294 hospital events for 125 subjects were obtained from the Medical Records department during the initial 2 years of the clinical trial. Of these, 85 (29%) hospital events were identified as needing to undergo chart review. Thirty-three (39%) of the 85 charts that underwent blinded review were identified as being HF related hospital events. Conclusions. Automating procedures for obtaining hospital event information expedited both the systematic data review and chart review processes. In this study, combined use of automated processes for obtaining hospital event data and nurse review reduced the number of charts requiring blinded review by nearly 60%.