Graduation Year

2020

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Information Systems and Decision Sciences

Major Professor

He Zhang, Ph.D.

Co-Major Professor

Shivendu Shivendu, Ph.D.

Committee Member

Balaji Padmanabhan, Ph.D.

Committee Member

Jared Williams, Ph.D.

Keywords

data quality, EHR, human AI collaboration, length of stay, task complexity, workflow inefficiency

Abstract

This dissertation focuses on three key aspects in health IT management: (1) Complexities in the collection of health data in electronic health record (EHR) systems and the use of EHR data in research, (2) Complexities of collaboration between physicians and AI for improving healthcare delivery, and (3) Complexities of workflows and collaborations between healthcare organization (HCO) staff during the delivery of care. The first dissertation essay (Chapter 1) examines the key data quality issues that arise in recorded health information in EHR systems, provides quality thresholds that the data needs to meet for mitigating errors and increasing reproducibility of downstream research. The essay finally provides remedial actions that can be taken to enhance quality of EHR data. The second dissertation essay (Chapter 2), examines the challenge of collaboration between human experts and AI in order to increase the quality of care provided. The essay looks at this problem from the lens of task complexity where it first maps the various elements that can impact the complexity of an information processing task in HCOs. The essay then builds a framework the helps reduce the identified elements of task complexity to increase performance in such tasks by using complementary strengths of human-experts and AIs. The third essay (Chapter 3) studies complex and dynamic workflows in HCOs and investigates the impacts of workload and collaborations between HCO staff on inefficiencies and delays in healthcare delivery. The essay further investigates the impacts of these inefficiencies and delays on the quality of care (QoC) provided by the HCO to its patients. The essay develops a new methodology for studying dynamic and complex workflows in HCOs, provide insights into the factors that play a role in inefficient workflows and their impacts key QoC metrics.

Share

COinS