Date of Award

7-13-2016

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Information Science and Systems Engineering

First Advisor

Daniel Berleant

Second Advisor

James Bailey

Abstract

Expenditures on health information technology (HIT) for healthcare organizations are growing exponentially and the value of it is the subject of criticism and skepticism. Because HIT is viewed as capable of improving major health care indicators, the government offers incentives to health care providers and organizations to implement solutions. However HIT implementation is still subject to numerous technical, financial and organizational barriers. Health Information Exchange (HIE) networks are examples of HIT implementations that have been shown to assist healthcare providers in exchanging data and facilitating patient care through the circulation of critical clinical information between health care organizations statewide and nationwide. However, the real utility of these networks in terms of improving health outcomes is still unclear. The objective of this research illustrates how data quality issues impact health information and impede efforts to spread effective HIT solutions. The main focus of this research investigates data quality issues and other health information problems for patients with congestive heart failure (CHF) whose information was shared within a regional HIE network, the MidSouth eHealth Alliance (MSeHA) in Tennessee. Congestive Heart Failure is a major health problem and contributor to healthcare expenditures in the US. First an empirical data quality assessment method identified and evaluated the quality dimensions of the dataset. Secondly, prediction models using GEE models and decision tress were built to determine the factors influencing major health care outcomes such as the number of encounters, number of laboratory testing, number of procedures, number of subsequent visits and admissions, realization of specific procedure like catheterization, CABG, echocardiogram, CT-angiogram or other cardiac procedures. All models included administrative and biological parameters from data extracted from the HIE as well as HIE usage data in order to determine the impact of the HIE use these health outcomes. Both analytical methods produced models that showed that use of HIE in the Emergency Department for patients with CHF was associated with somewhat improved health outcomes. However, it also showed the limitation of the HIE in terms of data quality. Although this study provided evidence supporting the use of HIT and quality information exchange through the HIE, more integrated clinical solutions are now needed.

Download

Share

COinS