Abstract
Objectives. To examine rural-urban disparities in overall mortality and leading causes of death across Hispanic (any race) and non-Hispanic White, Black, American Indian/Alaska Native (AI/AN), and Asian/Pacific Islander populations.
Methods. We performed a retrospective analysis of age-adjusted death rates for all-cause mortality and 5 leading causes of death (cardiovascular, cancer, unintentional injuries, chronic lower respiratory disease, and stroke) by rural versus urban county of residence in the United States and race/ethnicity for the period 2013 to 2017.
Results. Rural populations, across all racial/ethnic groups, had higher all-cause mortality rates than did their urban counterparts. Comparisons within causes of death documented rural disparities for all conditions except cancer and stroke among Hispanic individuals; Hispanic rural residents had death rates similar to or lower than urban residents. Rural Black populations experienced the highest mortality for cardiovascular disease, cancer, and stroke. Unintentional injury and chronic lower respiratory disease mortality were highest in rural AI/AN and rural non-Hispanic White populations, respectively.
Conclusions. Investigating rural-urban disparities without also considering race/ethnicity leaves minority health disparities unexamined and thus unaddressed. Further research is needed to clarify local factors associated with these disparities and to test appropriate interventions.
In 2017, the Centers for Disease Control and Prevention (CDC) published a series of studies examining differences in health behaviors and outcomes between metropolitan (urban) and nonmetropolitan (rural) populations in the United States.1 These reports have helped refocus discussion regarding the health needs of rural residents and potential policies and programs to mitigate rural-urban disparities. However, apart from rural-urban differences in suicide deaths,2 the potential dual disparities experienced by rural minority residents have not received the same level of attention. Recent work has documented disparities in age-adjusted mortality rates across racial/ethnic groups and by rurality separately, without showing rural-urban disparities in mortality rates within each racial/ethnic group.3,4 Other studies have addressed a range of populations but not American Indian/Alaska Native (AI/AN) rural residents.5 In this article, we quantify the absolute and relative mortality differences in the 5 leading causes of death across rural and urban populations, stratified by race/ethnicity.
METHODS
We examined US age-adjusted death rates for both overall and cause-specific mortality for the period 2013 to 2017. Data were drawn from the National Vital Statistics System underlying-cause-of-death files, accessed through the Wide-ranging Online Data for Epidemiologic Research (WONDER) Web site of the CDC.6 Death certificates are part of the National Vital Statistics System; they are required by all states and include specification of county of residence. Because some of the racial/ethnic groups studied have relatively low representation in the rural population, we used a 5-year period to ensure adequate numbers of observations within specific causes of death for all racial/ethnic categories studied. Annual mortality rates were age adjusted using the 2000 US standard population.
Defining “rural” can be challenging; there are multiple levels of geography at which the rural-urban distinction can be made and multiple coding schemes and levels of rurality.7 Following Moy et al.,8 residence was dichotomized based on county of residence into rural (noncore and micropolitan) versus urban (metropolitan) county, using the 2013 National Center for Health Statistics definition.9 Although this dichotomous approach does not examine the distinction between rural counties with an urbanized area of 10 000 to 49 999 residents (micropolitan) and those lacking this level of urbanization (noncore), it has 2 advantages. First, it allows for sufficient numbers of observations within causes of death and racial/ethnic groups. Second, it allows comparability with previous work, such as the analysis by Moy et al.8 cited earlier.
Race/ethnicity is examined as a social differentiation that may expose an individual to interpersonal and structural disadvantage.10 Race/ethnicity was categorized as Hispanic (any race) and non-Hispanic White, Black, AI/AN, and Asian/Pacific Islander (API). Records for which Hispanic ethnicity was not coded as either “yes” or “no,” 0.33% of the records studied, were not included in the analysis.
Leading causes of death were identified and defined by using previous research.8 For overall mortality and for each leading cause of death, we compared the age-adjusted mortality rates between rural and urban populations by race/ethnicity; P values for differences were calculated with t statistics.
RESULTS
Overall age-adjusted mortality rates estimated across the 5-year period were highest among rural Black (981.3 deaths per 100 000) and rural AI/AN (970.0 deaths per 100 000) populations and lowest among urban API residents (394.3 deaths per 100 000; Table 1). Rural disparity, the degree to which rural death rates exceeded urban rates within the same period, was highest among AI/AN populations (+42%) and varied between +11% and +18% for all other groups. No group had all-cause mortality rates that were lower among rural than among urban residents. Within each racial/ethnic category, overall mortality disparities for rural residents have remained at the levels shown in Table 1, with only minor year-to-year variation, for the past five years (Table A, available as a supplement to the online version of this article at http://www.ajph.org).
TABLE 1—
Age-Adjusted Mortality per 100 000 US Residents, All-Cause Mortality, and Selected Leading Causes of Death, by Residence and Race/Ethnicity: 2013–2017
| Age-Adjusted Rate (95% CI) |
|||
| Cause of Death | Rural | Urban | Rural Disparitya (% Difference) |
| All-cause mortality | |||
| AI/AN | 970.0 (960.5, 979.5) | 684.5 (677.9, 691.1) | +42 |
| API | 466.5 (458.1, 474.9) | 394.3 (392.9, 395.7) | +18 |
| Black | 981.3 (976.7, 985.9) | 867.3 (865.8, 868.8) | +13 |
| Hispanicb | 580.7 (576.1, 585.3) | 522.7 (521.5, 523.9) | +11 |
| White | 837.7 (836.5, 838.8) | 728.8 (728.3, 729.3) | +15 |
| Cancer | |||
| AI/AN | 164.7 (160.9, 168.6) | 123.4 (120.6, 126.2) | +33 |
| API | 107.7 (103.7, 111.7) | 101.0 (100.3, 101.7) | +7 |
| Black | 203.1 (201.1, 205.2) | 188.3 (187.7, 189.0) | +8 |
| Hispanicb | 112.0 (110.0, 114.0) | 113.7 (113.2, 114.3) | −1 |
| White | 181.0 (180.5, 181.5) | 164.6 (164.4, 164.9) | +10 |
| Cardiovascular disease | |||
| AI/AN | 180.3 (176.1, 184.5) | 135.1 (132.0, 138.1) | +33 |
| API | 105.6 (101.6, 109.6) | 86.6 (85.9, 87.3) | +22 |
| Black | 240.0 (237.7, 242.3) | 207.6 (206.9, 208.4) | +16 |
| Hispanicb | 126.6 (124.4, 128.9) | 115.9 (115.3, 116.5) | +9 |
| White | 193.7 (193.2, 194.3) | 164.7 (164.5, 165.0) | +18 |
| Unintentional injury | |||
| AI/AN | 101.9 (99.0, 104.8) | 63.6 (61.7, 65.4) | +60 |
| API | 22.7 (21.0, 24.5) | 15.8 (15.5, 16.0) | +44 |
| Black | 47.1 (46.1, 48.1) | 39.1 (38.8, 39.4) | +20 |
| Hispanicb | 40.7 (39.7, 41.8) | 28.5 (28.3, 28.8) | +43 |
| White | 58.6 (58.3, 59.0) | 47.9 (47.7, 48.0) | +22 |
| Chronic lower respiratory disease | |||
| AI/AN | 44.9 (42.8, 47.0) | 36.0 (34.5, 37.6) | +25 |
| API | 13.5 (12.0, 15.0) | 12.3 (12.0, 12.5) | +10 |
| Black | 33.0 (32.2, 33.8) | 29.4 (29.2, 29.7) | +12 |
| Hispanicb | 20.3 (19.4, 21.2) | 17.4 (17.2, 17.6) | +17 |
| White | 56.8 (56.6, 57.1) | 43.8 (43.6, 43.9) | +30 |
| Stroke | |||
| AI/AN | 36.6 (34.7, 38.5) | 29.1 (27.7, 30.5) | +26 |
| API | 36.8 (34.4, 39.2) | 29.7 (29.4, 30.1) | +24 |
| Black | 60.1 (58.9, 61.2) | 50.6 (50.3, 51.0) | +19 |
| Hispanicb,c | 31.3 (30.2, 32.4) | 31.3 (31.0, 31.6) | 0 |
| White | 40.1 (39.9, 40.4) | 34.8 (34.7, 34.9) | +15 |
Note. AI/AN = American Indian/Alaska Native; API = Asian/Pacific Islander; CI = confidence interval.
Rural disparity is calculated by dividing the rural mortality rate for each line by the urban mortality rate.
Hispanic includes persons identified on the death certificate as Hispanic, regardless of race. All other race categories include only individuals recorded as non-Hispanic.
Rural–urban rates for stroke among Hispanic residents are not significantly different: All other rural–urban comparisons are significant at P < .001.
Within causes of death, rural Black populations had the highest mortality rates across combinations of race/ethnicity and rurality for cardiovascular disease (240.0 deaths per 100 000), cancer (203.1 deaths per 100 000), and stroke (60.1 deaths per 100 000). Unintentional death rates were highest among rural AI/AN residents (101.9 deaths per 100 000), whereas the burden of chronic lower respiratory disease was highest among rural White residents (56.8 deaths per 100 000).
Patterns of rural-urban disparity varied across both race/ethnicity and the leading causes of death. As noted in the previous paragraph, rural White residents had the highest age-adjusted death rates for chronic lower respiratory disease; disparities in rates of death also were highest for this condition (+30% of urban rate; Table 1). Among all other racial/ethnic groups, disparities were greatest for unintentional injury. Rural AI/AN residents had unintentional injury mortality rates 60% higher than did those in their urban peers. Although cancer death rates were either the first or the second leading cause of death among all groups, the magnitude of rural-urban disparities were lower for cancer than for other leading causes of death among all groups except AI/AN populations.
Hispanic residents were the only subpopulation within which death rates from any causes studied were equal to or less than those among urban populations. Rural Hispanic population had slightly lower cancer mortality rates than did their urban peers and did not differ from urban Hispanic residents in stroke mortality. Despite this, overall mortality remained higher for rural than for urban Hispanic populations.
DISCUSSION
Overall mortality among rural residents was higher than that among their urban peers across all racial/ethnic categories. Rural Black and AI/AN populations were particularly disadvantaged. The pattern of rural disparity reflects multiple structural factors, all of which are exacerbated in counties with high minority representation. Rural Black residents, for example, are concentrated in the southern United States, and nearly all of these states failed to implement Medicaid expansion under the Affordable Care Act.11 Rural residents, particularly rural minority populations, disproportionately possess characteristics such as low educational attainment and poverty that place them at risk for increased mortality,1 and these characteristics are associated with lack of resources in rural communities.12 A full discussion of rural gaps in all public health and direct health services is beyond the scope of this brief article.
This study had multiple limitations, because it was based on nationally aggregated death rates. We used the county-level metropolitan-nonmetropolitan distinction available in the CDC WONDER data set; we did not have the data to examine other possible metrics for rurality, such as census tract–based rural-urban commuting area codes. We did not adjust results for community-level factors such as provider availability or for individual comorbidities and socioeconomic characteristics. Nonetheless, our analysis illustrates within-group differences between rural and urban populations of diverse race/ethnicity and thus may prompt additional research with more robust analytic approaches.
PUBLIC HEALTH IMPLICATIONS
Unseen disparities cannot be addressed. Therefore, we need more research examining the intersection of rural residence and race/ethnicity, particularly in identifying gaps in the availability of services and preventive interventions that may be linked to these disparities. Interventions need to be developed and tested with rural minority populations rather than with convenient minority populations. Cultures and health care systems are each local; they cannot be understood at a distance.
To the extent possible, future research should employ commonly used definitions of levels of rurality, to allow comparison across studies. Rural minority residents, whose outcomes are masked by the larger White population, will continue to experience disparities unless public health surveillance, policies, and interventions are improved.
CONFLICTS OF INTEREST
The authors have no conflicts of interest to disclose.
HUMAN PARTICIPANT PROTECTION
This study was deemed exempt (no human participants) by the institutional review board of the University of South Carolina.
Footnotes
REFERENCES
- 1.Centers for Disease Control and Prevention. MMWR Rural Health Series. Available at: https://www.cdc.gov/mmwr/rural_health_series.html. Accessed October 29, 2019.
- 2.Ivey-Stephenson AZ, Crosby AE, Jack SP, Haileyesus T, Kresnow-Sedacca M. Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death — United States, 2001–2015. MMWR Surveill Summ. 2017;66(18, No. SS-18):1–16. doi: 10.15585/mmwr.ss6618a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Woolf SH, Schoomaker H. Life expectancy and mortality rates in the United States, 1959-2017. JAMA. 2019;322(20):1996–2016. doi: 10.1001/jama.2019.16932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hall JE, Moonesinghe R, Bouye K, Penman-Aguilar A. Racial/ethnic disparities in mortality: contributions and variations by rurality in the United States, 2012–2015. Int J Environ Res Public Health. 2019;16(3):436. doi: 10.3390/ijerph16030436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Yaemsiri S, Alfier JM, Moy E et al. Healthy People 2020: rural areas lag in achieving targets for major causes of death. Health Aff (Millwood) 2019;38(12):2027–2031. doi: 10.1377/hlthaff.2019.00915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Centers for Disease Control and Prevention. CDC WONDER: about underlying cause of death, 1999–2018. Atlanta, GA: Centers for Disease Control and Prevention; 2019. Available at: https://wonder.cdc.gov. Accessed November 15, 2019.
- 7.Bennett KJ, Borders TF, Holmes GM, Kozhimannil KB, Ziller E. What is rural? Challenges and implications of definitions that inadequately encompass rural people and places. Health Aff (Millwood) 2019;38(12):1985–1992. doi: 10.1377/hlthaff.2019.00910. [DOI] [PubMed] [Google Scholar]
- 8.Moy E, Garcia MC, Bastian B et al. Leading causes of death in nonmetropolitan and metropolitan areas- United States, 1999-2014. MMWR Surveill Summ. 2017;66(1):1–8. doi: 10.15585/mmwr.ss6601a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ingram DD, Franco SJ. 2013 NCHS urban-rural classification scheme for counties. Vital Health Stat 2. 2014;(166):1–73. [PubMed] [Google Scholar]
- 10.Whaley AL. Ethnicity/race, ethics, and epidemiology. J Natl Med Assoc. 2003;95(8):736–742. [PMC free article] [PubMed] [Google Scholar]
- 11.Tolbert J, Orgera K, Singer N, Damico A. Key facts about the uninsured population. December 13, 2019. Available at: https://www.kff.org/uninsured/issue-brief/key-facts-about-the-uninsured-population. Accessed March 25, 2020.
- 12.Lombardo C. Why White school districts have so much more money. February 26, 2019. Available at: https://www.npr.org/2019/02/26/696794821/why-white-school-districts-have-so-much-more-money. Accessed March 23, 2020.