Using clinical analysis to classify 30-day trauma readmissions.

Abstract

Unplanned hospital readmissions are common across specialties. Descriptive readmission studies commonly query large administrative databases, which some speculate lack clinical granularity. This article provides the framework for a process improvement initiative aimed at identifying clinically meaningful reasons for trauma readmission. Our study hypothesizes an expected difference between the clinically abstracted reasons for readmission and those documented by the data processing staff in the trauma registry and that those differences will be the starting point to target performance improvement. This is a retrospective, cohort study from 2014 to 2016 involving 18,998 trauma evaluations at a Level I trauma center. The systematic categorization of trauma readmissions was completed via clinical chart review. Readmissions were categorized following an organizational flowchart. The chart reviews ultimately resulted in two readmission categories: primary and secondary reasons for 30-day trauma readmission. There were 413 readmissions, an overall readmission rate of 2.7%. The highest rate of readmission, by mechanism of injury, was gunshot wounds (11%). Secondary reasons for readmission predominated (76.1%). Complications led (41%), followed by observation (8.8%) and pain (8.6%). Following readmission chart review and categorization, the trauma registry data were queried and categorized via the same method. When the two methods of data collection were compared, there was a significant difference (p < 0.0001). The granular dissection of readmission charts proved to assist in isolating clinically significant readmission variables, providing clarity into the reasons behind trauma readmission. If determined solely by the trauma registry data, our performance and quality improvement initiatives would be misguided. We recommend clinical oversight of databases, with clinical review of key areas in order to guide performance improvement.