An Industrial Engineering-Based Approach to Designing and Evaluating Healthcare Systems to Improve Veteran Access to Care
VanDeusen, Adam
2021
Abstract
Access to healthcare is a critical public health issue in the United States, especially for veterans. Veterans are older on average than the general U.S. population and are thus at higher risk for chronic disease. Further, veterans report more delays when seeking healthcare. The Veterans Affairs (VA) Healthcare System continuously works to develop policies and technologies that aim to improve veteran access to care. Industrial engineering methods can be effective in analyzing the impact of such policies, as well as designing or modifying systems to better align veteran patients’ needs with providers and resources. This dissertation demonstrates how industrial engineering tools can guide policy decisions to improve healthcare access by connecting veterans with the most appropriate healthcare resources, while highlighting the trade-offs inherent in such decisions. This work comprises four stages: (1) using optimization methods to design a healthcare network when introducing new provider options for chronic disease screening, (2) developing simulation tools to model how access to care is impacted when scheduling policies accommodate patient preferences, and (3) simulating triage strategies for non-emergency care during COVID-19, and (4) evaluating how treatment decisions impact patient access when guided by risk-based prediction models compared to current practice. In the first stage, we consider veteran access to chronic eye disease screening. Ophthalmologists in the VA have developed a platform in which ophthalmic technicians screen patients for major chronic eye diseases during primary care visits. We use mixed-integer programming-based facility location models to understand how the VA can determine which clinics should offer eye screenings, which provider type(s) should staff those clinics, and how to distribute patients among clinics. The results of this work show how the VA can achieve various objectives including minimizing the cost or maximizing the number of patients receiving care. In the second stage, we simulate patients seeking care for gastroesophageal reflux disease with primary care and gastrointestinal providers. This simulation incorporates policies about how to schedule patients for visits in various modalities, including face-to-face and telehealth, and also considers uncertainty in key factors like patient arrivals and demographics. Results of these models can help us understand how scheduling based on these preferences impacts access, including time to first appointment and number of patients seen. Such metrics can guide healthcare administrators as new technologies are introduced that offer options for how patients interact with their providers. In the third stage, we simulate patients seeking non-emergency outpatient care under reduced appointment capacity due to the COVID-19 pandemic. We demonstrate this using endoscopy visits as a central example. We use our simulation model to understand how various strategies for adjusting patient triage and/or clinic operations can mitigate patient backlog and reduce patient waiting times. In the fourth stage, we integrate multiple industrial engineering methods to examine how access is impacted among chronic liver disease patients when predictive modeling is introduced into treatment planning. We developed a simulation model to help clinical decision-makers better understand how using a predictive model may change the care pathway for a specific patient and also impact system decisions, such as required staffing levels and clinical data acquired at specific patient visits. The model also helps clinicians understand the value of specific clinical data (lab values, vitals, etc.) by demonstrating how better or worse inputs to the predictive models have larger system impacts to patient access.Deep Blue DOI
Subjects
operations research healthcare
Types
Thesis
Metadata
Show full item recordCollections
Remediation of Harmful Language
The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.
Accessibility
If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.