Date of Award


Document Type

Campus Access Dissertation


Epidemiology and Biostatistics



First Advisor

Wilfried Karmaus

Second Advisor

Erik Svendsen


Introduction: Environmental Public Health Disasters (EPHD) generate increased demand for primary care services after initial surge (i.e., "secondary surges") that overwhelm the local health system. There is a current gap in the literature examining a disaster's impact on long-term health care access for medically vulnerable populations. The goal of this quasi-experimental study was to quantify the long-term impact of the 2005 chlorine spill in Graniteville, South Carolina on a health care system serving the medically vulnerable by demonstrating unmet primary care need.

Objectives: 1) Develop a framework for understanding secondary surge by suggesting population and health system variables that act separately and together to affect the delivery of health care for medically vulnerable populations post-disaster. 2) Perform an interrupted time-series analysis by fitting autoregressive integrated moving average (ARIMA) models to 1) determine the impact of an EPHD on health care access among directly affected Medicaid beneficiaries and 2) forecast monthly patient volume of ACSC in medically vulnerable population recovering from an EPHD. 3) Perform multivariable log-linear regression analysis using repeated measures to examine whether visits for ACSCs among Medicaid beneficiaries are sensitive to the impact of an EPHD. Methods: We conducted a retrospective analysis of preventable hospital and ED discharges for ACSCs before and after the EPHD. We obtained data on ACSCs for South Carolina (SC) Medicaid recipients age 18-64 living in or near the Graniteville community (i.e., direct group) and compared to two neighboring SC MU communities (i.e., primary and secondary control groups) from 2002 to 2007, with 2005 representing the year of the chlorine accident.

Public Health Implications: Our research addresses traditionally overlooked inequities in access to timely and effective healthcare among medically vulnerable subgroups following an EPHD. Reduced health system performance creates a new vulnerable population that is likely to experience long-term access to primary care barriers during recovery. Preventable visits for ACSCs may be used in post-EPHD medical surveillance as a population-based indicator to track pre- and post- changes in primary care access for medically vulnerable populations in disaster recovery. By applying quantitative research methods to the systematic analysis and evaluation of health services following a disaster, we provide evidence-based information to aid policymakers, community leaders, public health practitioners, and medical providers to make better health care decisions that improve health care delivery for the medically vulnerable during disaster recovery.