Measurements of Rurality and Their Effect on Mental Illness and Substance Use

Abstract

According to the US government, nearly 1 in 5 Americans live in rural areas. In general, rural Americans have poor health outcomes, including higher rates of chronic disease, mental illness and certain types of substance use. A variety of different methods are used to assess rurality in health services research, making it challenging to precisely quantify the prevalence of mental illness and substance use in this population, as well as compare study conclusions. As policymakers become increasingly interested in addressing health disparities between urban and rural populations, it is important to assess and evaluate the different methods used to define rurality itself and determine how those methods affect estimates of depression and substance use, so that true disparities can be accurately captured and addressed. This dissertation will identify current definitions and methods used to measure rurality among published studies and then employ various identified methods to quantify the effect of measurement choice on prevalence of mental illness and substance use in rural populations. The dissertation will follow a three publishable paper model that will include a literature review and two empirical studies using secondary data as described below. For Paper 1, I identify peer-reviewed studies from HSR journals that use any method to measure rurality in their analysis. I analyze whether geographic units and methods used to classify rurality differ by focus area including costs, quality, and access to care. For paper 2, I quantify the impact of different measurements of rurality have on estimates for hospitalizations for depression and substance use. Using 5 different measurements of rurality, I calculate the levels of agreement as well as examine how characteristics of patients with depression or substance use disorder differ based on the definition of rurality used. In paper 3, I examine differences in the relationship between unmet mental health need and subsequent drug use in individuals with a history of depression. Using the National Survey on Drug Use and Health and a pooled crosssectional study design, I examine drug use by alcohol, marijuana, and prescription opioid use. Additionally, I compare self-medication and substance use in individuals by geographic location.