Date of Award

1-1-2013

Document Type

Open Access Dissertation

Department

Health Services and Policy Management

First Advisor

Janice C Probst

Abstract

Background. The U.S. healthcare movement to improve quality and patient outcomes has prompted investigations into tools that can assist in these aims. Electronic health records (EHRs) are one tool proposed by the Institute of Medicine (IOM). The objective of this original dissertation research was to examine the relationship between electronic health record functionalities and two outcomes of care as proxies for quality: risk-adjusted mortality for abdominal aortic aneurysm (AAA) repair, coronary artery bypass grafting (CABG), and percutaneous coronary intervention (PCI) and log estimated cost-per-discharge. Methods. This study used 2009-2010 hospital inpatient administrative discharge data from the Nationwide Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Research and Quality merged with data from the 2009-2010 American Hospital Association Information Technology Supplement. A pooled cross-sectional design was used, at the hospital-level, to determine if advanced levels of select CD, CPOE, and CDS functionalities implementation were associated with two outcomes of interest.

Results. Regression results revealed a significant positive relationship between level of CDS implementation and risk-adjusted mortality for AAA repair and PCI, controlling for patient-mix and hospital characteristics. Bivariate analyses revealed significant relationships for risk-adjusted mortality across levels of CDS implementation for

hospitals performing AAA repair (drug-allergy alerts and drug-drug interaction alerts) and PCI (drug-allergy alerts and drug-dosing support). The multivariate regression models for all three procedures modeled individually failed to detect a relationship among average level of CD, CPOE, and CDS implementation and log-transformed estimated cost per discharge, controlling for patient and hospital characteristics.

Conclusion. Despite knowing the exact ways in which EHR functionalities of interest are implemented and used across the inpatient setting, this study aimed to provide a foundation for future research on such relationships. While no significant relationship was detected between level of EHR functionalities implementation and log-transformed estimated cost per discharge, risk-adjusted mortality for AAA repair and PCI were found to be positively associated with increased implementation of select CDS functionalities. This study answers the 2012 call from the IOM for researchers to report any findings of the potential unintended consequences of EHR use.

Rights

© 2013, Deshia Ann Leonhirth

Share

COinS