Retrospective Chart Review Research: Get It Right to Get It Published

Retrospective Chart Review Research: Get It Right to Get It Published (Research Committee and Evidenced-Based Healthcare & Implementation (EBHI) Interest Group Sponsored)
Presenters:
Dan Mayer, MD
Carly Eastin, MD
Andrew C. Meltzer, MD, MS
Martin P. Wegman, MD, PhD
Michael Gottlieb, MD
Peter C. Hou, MD
Learning Objectives:
Upon completion of this session, participants should be able to:
-Describe the pros and cons of performing a medical records review study.
-Describe the steps in creating a medical records review study.
-Review three medical record review studies and decide how they can be improved
Chart or medical records review studies rely on existing patient medical records to link cause and effect. Relatively easy to perform with plentiful clinical research material data readily available in the patient’s chart, they are a popular type of research platform. When done properly, they are relatively easy to get published but there are many potential pitfalls in performing these studies. Researchers must better understand the processes of performing these studies and getting their research published. We will show how a well-done medical record review study should be done to insure timely publication. Medical record data may have poor quality due to subjective descriptions made by a variety of clinicians. Reviewers may use subjective standards to determine the presence of the risk factor and outcome. An implicit medical record review study does not have a clear-cut protocol for reducing these sources of bias. Explicit medical record review studies use clearly defined protocols and objective measures in reviewing medical charts as described by Gilbert and Lowenstein (Annals of Emergency Medicine, March 1996) and Kaji et. al. (Annals of Emergency Medicine, April 2014). This includes review done in a blinded manner with clearly determined descriptors and reviewer training and measurement of inter-rater reliability. We will ask the small groups to give three items that need to be modified in each three hypothetical medical record review studies for them to become acceptable for publication. These will include one that is publishable in its current form, a second that can be improved and could then be published and a third study that is clearly not publishable and cannot be fixed for publication.