Date of Award

January 2022

Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Reza Yaesoubi


Context: Medicaid managed care network adequacy standards vary widely across states and are not typically informed by scientific evidence. The comparative efficacy of these standards for protecting the health of the Medicaid population has not yet been comprehensively researched.Objective: The aims of this study are to construct simulation modeling methods to approach this policy problem and to determine which numeric values for network adequacy standards are most effective for producing favorable health outcomes for Medicaid recipients who develop CVD. Design and Setting: A continuous-time Markov model was used to simulate the natural history of cardiovascular disease, using a cohort that is representative of the Medicaid population over 40, under different provider appointment wait times and CVD emergency travel time delays. Input and Output Measures: Medicaid claims data from Tennessee in 2019, Social Security life expectancy data, the Centers for Disease Control and Prevention's Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) database, and values pulled from existing literature were used to inform input parameters. Survival time, age at death, number of lifetime hospital visits, number of lifetime provider visits, time spent in recovery, time spent waiting for an appointment, lifetime healthcare costs, and lifetime healthcare costs attributable to hospitalization were collected as simulation outputs. Results: It was found that the strategy with a 45-day appointment wait time and 0-minute emergency travel time delay yielded the most favorable health outcomes for individuals with CVD: mean age at death of 83.79 (83.10, 84.47) and mean survival time of 32.08 (31.09, 33.07). When the strategies hypothesized to be the “best” (7-day wait times and no travel delay) and “worst” (90-day wait times and 90-min travel delay for emergencies) were run in comparison to one another, statistically significant differences were found for time spent in recovery, time spent waiting for an appointment, provider visit quantity, and healthcare system cost burden. Statistically significant differences were not found for life expectancy, hospital visit quantity, and costs attributable to hospitalization. Conclusion: There is not enough evidence of robustness in these results to recommend that policy decisions should be made using them; further complexities and calibration should be incorporated into the model before doing so.


This is an Open Access Thesis.

Open Access

This Article is Open Access