Community Characteristics Associated with Premature Mortality in the Appalachian Region

Background: A key component of population health status is premature mortality, defined as a death occurring before the age of 75 years. Of particular concern is premature mortality among vulnerable populations of Appalachia. However, it remains unclear what specific community characteristics are the leading predictors of premature mortality in the Appalachian population to be considered when developing high-priority interventions to improve premature mortality. Objective: We hypothesized that a limited number of community characteristics would account for most of premature deaths in the Appalachian population. The objective of our study was to identify these essential community characteristics. Methods: County Health Rankings and Roadmaps data were used to investigate premature mortality, as measured by the number of years of potential life lost (YPLL) before age 75 years per 100,000 population (age-adjusted), and associated community characteristics in the 13 Appalachian states. Counties’ rurality status was defined using the United States Departments of Agriculture Economic Research Service’s (USDA ERS) Rural-Urban Continuum Codes (RUCC). Descriptive and inferential statistical analyses, including multiple regression, were conducted using SAS and SPSS. A conventional Type I error (alpha) level of 0.05 (p<0.05) was used for all analyses. Results: Premature mortality rates, as measured by the number of years of potential life lost (YPLL) before age 75 years, in the 420 Appalachian counties were statistically significantly higher than in their counterparts – the 679 non-Appalachian counties of the 13 Appalachian states: 9,299.7 vs. 8,425.4 per 100,000 (p<0.001). Within the Appalachian counties, premature mortality rates were statistically significantly higher in rural areas than in urban areas: YPLL 10,010.9 vs. 8,546.6 per 100,000 (p<0.001). Multivariable analysis identified key factors predicting premature mortality in the Appalachian population that can be grouped into three major categories: health behavior, health care, and social environment factors. Conclusions: Premature mortality in rural Appalachian counties is statistically significantly higher than in urban Appalachian counties. The key predictors of premature mortality in the Appalachian Region can be grouped into three major categories: health behavior; health care; and social environment factors. This study furthers the body of knowledge necessary to develop targeted interventions to improve premature mortality in the Appalachian Region.


Introduction
A key component of population health status is premature mortality, defined as a death occurring before the age of 75 years [1,2]. Premature mortality is a major public health problem both in the United States and worldwide, and it is exacerbated by socioeconomic disparities [3][4][5]. Stretching along the Appalachian Region's population is vulnerable to socio-economic disparities and associated public health issues [7]. It has income level below the U.S. average and unemployment rates above the U.S. average rates, and only about 25% of the Appalachian adults hold a bachelor's degree, compared to nearly 33% nationwide [8]. Of particular concern is premature mortality among the vulnerable population of the Appalachian Region. However, it remains unclear what specific community characteristics are the leading predictors of premature mortality in the Appalachian population to be considered when developing high-priority interventions to improve premature mortality. We hypothesized that a limited number of community characteristics would account for most of premature deaths in the Appalachian population. The objective of our study was to identify these essential community characteristics. areas of >1 million population) to 9 (completely rural or <2,500 urban population, not adjacent to a metro area) [11]. We classified nonmetropolitan counties with population less than 20,000 as rural (RUCC codes 6-9) and other counties (RUCC codes 1-5) as urban. Of the total of 420 Appalachian counties (total population:

Statistical Analysis
The primary outcome of interest was premature mortality,

Bivariate Analysis
Within the 13 states containing Appalachian counties, ageadjusted premature mortality rates (YPLL) in the 420 Appalachian counties were statistically significantly higher than in their counterparts -the 679 non-Appalachian counties of the 13

Multivariable Analysis
To identify the key factors predicting age-adjusted premature mortality in the Appalachian Region, we conducted multivariable analysis with multiple linear regression, with the number of YPLL as the dependent (outcome) variable. Results of the multiple linear regression are presented in Table 3. Smoking (p<0.001), physical inactivity (p<0.001), unemployment (p<0.001), and percentage of children living in poverty (p<0.001) were statistically significantly directly associated with premature mortality: an increase in these factors was associated with an increase in the number of YPLL (Table 3). Mammography screening (p=0.019) and high school graduation rates (p=0.006) were statistically significantly inversely associated with premature mortality: an increase in these factors was associated with a decrease in the number of YPLL (Table 3).

Discussion
Our study found statistically significantly higher premature mortality rates in the Appalachian counties compared to their non-Appalachian counterparts as well as rural-urban disparities in premature mortality in the Appalachian Region. We also identified the key predictors of premature mortality in the Appalachian population which is vulnerable to socio-economic disparities [7].
Rural-urban premature mortality disparities can be attributed to several factors. Firstly, urban areas have better availability of healthcare services, with more healthcare facilities and health providers per capita. We found that the number of PCP per 100,000 residents is higher in urban Appalachian counties than in rural ones. Our findings of health services disparities are consistent with findings from other studies that reported rural-urban disparities in mortality [12][13][14]. Lack of healthcare providers in rural areas and issues with rural hospital closures were reported [15,16]. Secondly, limited availability is exacerbated by limited accessibility impeded by transportation barriers in rural areas. We found that rural Appalachian residents are less likely to undergo mammography screening and diabetes monitoring than urban residents. Other studies demonstrated that rural residence is associated with longer travel distances and longer lasting trips to health care resulting in unmet health care needs [17,18].
Thirdly, our results demonstrated a higher prevalence of unhealthy behaviors, such as smoking and physical inactivity, in rural Appalachian counties compared to their urban counterparts.
Our findings are consistent with results from other studies that demonstrated higher prevalence of non-smoking, aerobic leisure time physical activity and fruit consumption along with lower prevalence of obesity among the adults residing in urban counties than in non-urban ones [19,20]. Our study identified essential, statistically significant key predictors of premature mortality in the Appalachian Region: smoking, physical inactivity, unemployment and children in poverty (statistically significantly associated with a higher premature mortality) and those which are statistically significantly associated with a lower premature mortality (mammography screening and high school graduation). These key predictors can be grouped into three major categories: health behavior (smoking, physical inactivity), health care utilization (mammography screening), and social environment factors (unemployment, percentage of children in poverty, high school graduation rate). Our study is not free from limitations. Because only county-level data were available, the results cannot be extrapolated directly to the individual level. Nevertheless, the objective of our study was to investigate the big picture of premature mortality in the Appalachian Region and community characteristics associated with disparities in premature mortality in this population.

Conclusions
Premature mortality in rural Appalachian counties is statistically significantly higher than in urban Appalachian counties. The key predictors of premature mortality in the Appalachian Region can be grouped into three major categories: health behavior; health care; and social environment factors. This study furthers the body of knowledge necessary to develop targeted interventions to improve premature mortality in the Appalachian Region.