Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Sociology

Content Description

1 online resource (vii, 157 pages) : illustrations (some color), color maps.

Dissertation/Thesis Chair

Tse-Chuan Yang

Committee Members

Kate W. Strully, Glenn Deane


county, opioid crisis, rurality, spatiotemporal analysis, Opioid abuse, Rural health, Rural population, Spatial analysis (Statistics)

Subject Categories

Demography, Population, and Ecology | Sociology


Previous studies on the opioid crisis in the United States at the subnational level rarely consider temporal and spatial dependency simultaneously. Under the theoretical framework of spatial inequality, this dissertation aims to understand how the uneven development of the opioid crisis across the rural-urban spectrum has been constructed and evolved from 2006 to 2018. Based on a new continuous measure of rurality, a spatiotemporal dataset is assembled based on the Multiple-Cause-of-Deaths File and other federal data sources, such as the U.S. Opioid Dispensing Map and the American Community Survey. The methods include geographical mapping, fixed effects modeling, and spatial panel fixed effects modeling. The first two empirical studies examine the explanatory/confounding variables concerning the relationship between rurality and two opioid outcomes (i.e., opioid prescribing and opioid-involved mortality) while the third examines the within-rural and within-urban heterogeneity in opioid mortality. The findings show significant temporal and spatial structures of opioid outcomes. Increasing rurality elevates opioid prescribing stemming from the increasing percentages of whites and the unemployed, decreasing availability of nurse practitioners and physician assistants, and increasing numbers of opioid prescribers per 10,000. However, in contrast to rural counties, urban counties may have so-called “drug environments” (e.g., illicit opioid markets) and suffer from higher opioid mortality rates related to their lower opioid prescribing rates, lower unemployment rates, and wider availability of primary healthcare services. Demographic composition, socioeconomic environment, opioid prescribing, and the availability of healthcare services can explain the within-rural and within-urban variations while concentrations of disadvantaged populations negatively affect rural counties more than they do urban counties. The findings highlight the mismatch between opioid prescribing and opioid mortality and the necessity of utilizing the healthcare system to improve the equity between rural and urban areas to eventually mitigate the crisis.