Date of Award

4-17-2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Engineering Technology

First Advisor

Melody Greer

Second Advisor

John Talburt

Abstract

A 2016 survey conducted by the Office of the National Coordinator for Health Information Technology (ONC) found that 96 percent of United States hospitals currently utilize certified Electronic Health Record (EHR) technology and meet meaningful use requirements promoted by the Centers for Medicare and Medicaid Services. With the increasing use of EHR, early studies on data quality within electronic health records revealed a long list of concerns within EHR workflows. The work in this dissertation applied Rule-based data quality assessment to Adverse Events (AEs) within the All Payers Claim Database (APCD). An adverse event is any patient harm resulting from medical care. Claims databases contain large integrated collections of medical insurance claims data including medical claims, pharmacy claims, dental claims, and eligibility and provider files and may help research into adverse events. Earlier studies of applying rules were done on clinical data and in this approach, it’s applied on Claims data for data errors. The main approach is broadly categorized into two questions to pursue: 1) adverse events detection, cleaning methodologies and their impact on the data quality of APCD data, and 2) evaluate the process of rules applied to APCD and adverse events data and the use proportions to make some general statements about International Classification of Diseases (ICD) codes and AEs that are novel. The diagnosis codes are standardized in the International Classification of Diseases (ICD) codes list by the World Health Organization. The number of rules applied to claims and AE data is 9 out of 22 possible, and the final analysis gave results of invalid member numbers. The first question talks about data cleansing, policies, standards in Extract Transform Load (ETL), and transformation rules that extract the AE data. Applying the rules is the second question. For example, in the Gender_Dx rule, the number of members with invalid gender and diagnosis is 213, these members have ICD codes that are for females but the gender is 'Male' in claims data. Based on the accessibility of claims data there is significant additional work to be done in this area of rules-based data quality assessment and monitoring in healthcare facilities and APCD data. Claim denials by the insurance company based on data quality errors and error types, comparing Arkansas Vs National Coverage, etc are some of the few additional topics.

Download

Share

COinS