Abstract 218: Impact of Electronic Health Record Components on Hospitals’ Performance in Heart Failure and Myocardial Infarction Mortality and Readmission Rates

Abstract

Listen

Translate article

Background: Cross-sectional studies have identified an association between the use of electronic health records (EHRs) and the quality of cardiovascular care; however, there is little longitudinal data to support this association. The objective of this study was to assess the impact of EHR implementation on heart failure (HF) and myocardial infarction (MI) 30-day mortality and 30-day readmission rates in the following two years. Methods: To measure hospitals’ implementation of EHR and its specific components, we used national data from the American Hospital Association (AHA) Health IT survey representing data for years 2009 and 2010. Hospitals which did not use EHRs in 2009 but did in 2010 were considered newly implemented. Hospitals which did not use an EHR system in either 2009 or 2010 were considered as not yet implemented and hospitals which were using EHRs in both 2009 and 2010 had already implemented EHRs. We examined the use of specific EHR components including clinical decision support tools, electronic clinical documentation, computerized provider order entry, the ability to generate quality data and the ability to electronically share patient data with outside hospitals. To assess hospital performance in HF and MI 30-day mortality/readmission rates, we used data from CMS Hospital Quality Initiative, Hospital Compare. We used baseline performance in 2010 and tracked change to 2012. We used a difference-of-difference statistical analysis to compare the change over time between hospitals with EHR components newly implemented to ones who had not yet implemented EHRs. We adjusted for hospital characteristics including bed size, region, profit-status and their baseline performance. Results: This study included 3300 hospitals that responded to both 2009 and 2010 surveys. In general, large, not-for-profit, and southern hospitals were significantly more likely to have implemented the different EHR components. For HF, implementation of two EHR components, ‘Clinical Discharge Summaries’ and ‘Electronic Exchange of Patient Data’ were significantly associated with slower increase in 30-day mortality as compared to hospitals who had not yet implemented them (p=0.01 and 0.002 respectively). Only the EHR component ‘Computerized Provider Order Entry of Lab Tests’ was significantly associated with a faster decrease in HF readmissions (p=0.03). For MI, none of the EHR components’ implementation was associated with changes in 30-day mortality. ‘Clinical Provider Order Entry for Radiology and Lab Tests’ were significantly associated with a slower increase in 30-day MI readmissions (p=0.03). Conclusion: Select components of EHRs have positive impact on heart failure 30-day mortality and myocardial infarction 30-day readmission rates. As hospitals and physician practices continue to adopt and expand their EHR systems these components and their potential benefits should be considered.