278 Effect of Electronic Medical Record Transition on Emergency Department Length of Stay

Abstract

Regulatory changes within the health care market have driven adoption of electronic medical records (EMRs) in the emergency department (ED) with anticipated benefits of improved documentation, compliance, hand-offs, and efficiency. Although EMRs in the ED are now nearly ubiquitous, there is still relatively little published experience regarding the operational effects of their implementation. We report the impact of a transition from one EMR (IDX Systems, General Electric) to another (Cerner Millennium, Cerner Corporation) on ED length of stay (LOS). Design: Retrospective analysis of routinely acquired operational data. Setting: Tertiary 23-bed ED without an emergency medicine training program. Type of Participants: Visits in the one year (September 5, 2009 through September 4, 2010) prior to EMR transition and the year (September 5, 2010 through September 4, 2011) after EMR transition. We categorized post-transition visits as early (first 180 days; September 5 through March 4) versus sustained (days 181-365; March 5 through September 4) and compared them to a similar time cohort from the previous year. In the primary analysis, we performed a simple pre-post comparison of LOS. In the secondary analysis, we evaluated the effects of both the transition and confounders (patient age, ED daily volume, nursing staffing, physician staffing, and effective hospital occupancy) using univariable and multivariable linear regression models. Statistical analyses were performed using version 9.3 of the SAS software package (SAS Institute, Cary, NC). All tests were two-sided and P < .05 were considered statistically significant. There were 24,640 visits post-transition (12,325 early and 12,315 sustained) and 23,348 visits pre- transition (12,058 corresponding to early and 11,290 corresponding to sustained). In the primary analysis, mean (standard deviation, SD) LOS in the early phase post-transition was 264 (133) minutes, and in the corresponding dates pre-transition was 241 (131) minutes; the difference was 23 minutes (P < .001). Mean (SD) LOS in the sustained phase post-transition was 248 (129) minutes, and in the corresponding dates pre-transition was 233 (130) minutes; the difference was 15 minutes (P < .001). In the secondary analysis, after adjusting for covariates, the mean LOS increased post-transition by 28 minutes in the early phase (P < .001) and by 20 minutes in the sustained phase (P < .001). In a single-facility study, transition from one EMR to another was associated with an increase in LOS, even when adjusting for covariates. This increase in LOS was seen in both the early phase after adoption and in the sustained phase as well. While this work is consistent with several previous descriptions of the effects of EMR transition, it is inconsistent with others. It is not clear if our observed association was related to the impact of EMR transition in general, the use of this EMR in particular, local site factors, or other factors yet to be considered. This work adds to the growing knowledge of the effect of EMRs on ED operations.