Rural-Urban Differences in Mortality From Ischemic Heart Disease, Heart Failure, and Stroke in the United States

Rural areas experience higher age-adjusted mortality rates (AAMR) from cardiovascular disease than urban areas.¹ However, trends for major cardiovascular disease subgroups remain unknown. Therefore, we analyzed trends in AAMR by urban-rural status for major cardiovascular disease subtypes: ischemic heart disease (IHD), heart failure (HF), and stroke.

We queried the publicly available, nationally representative deidentified CDC Wide-ranging Online Data for Epidemiological Research database for deaths secondary to IHD, HF, and stroke from 1999 to 2017. The analysis was considered exempt from Institutional Review Board approval. Annual AAMR estimates of deaths per 100 000 population from IHD, HF, and stroke were obtained using International Classification of Diseases codes² for large metropolitan (≥1 million), small-medium metropolitan (50 000–999 999), and nonmetropolitan rural (<50 000) areas.³ We used the Poisson regression with log link and robust standard errors to estimate annual percentage change (APC) in AAMR across 1999 to 2011 and 2011 to 2017 as the year 2011 has been previously used as an inflection point to assess recent changes in cardiovascular disease AAMR.¹ The coefficients for the year obtained from the regression model were used to calculate the APC estimates [APC=100×(exp[coefficient]−1)]. Post hoc Bonferroni correction was used to ensure consistency of our results. The Pearson χ² test was performed for categorical variables. A 2-tailed P value of <0.05 was considered statistically significant.

A total of 12 364 036 deaths (50% in large metropolitan, 30.3% in small-medium metropolitan, and 19.7% in rural areas) occurred from 1999 to 2017. The combined AAMR was largest in the rural areas followed by small metropolitan areas and large metropolitan areas.

IHD AAMR showed a sharp decline from 1999 to 2011 (large metropolitan APC −5.2%, 201.3–107.4; small metropolitan APC −4.7%, 182.1–105.5; and rural APC −4.2%, 198.2–122.1) followed by a slower but consistent decline from 2011 to 2017 (large metropolitan APC −3%, 107.4–88.8; small metropolitan APC −2.4%, 105.5–91.7; and rural APC −1.8%, 122.1–109.4) with a relatively large reduction in large metropolitan areas followed by small metropolitan and rural areas uniformly across all ages, race, and ethnic groups (Table).

Table.

Trends in AAMR Across Decedents Characteristics for Ischemic Heart Disease, Heart Failure, and Stroke (1999–2017)

HF AAMR showed significant downtrend from 1999 to 2011 (large metropolitan APC −2.1%, 27.6–22.1; small metropolitan APC −2.2%, 31–24.2; and rural APC −2.2%, 35.8–28) across all subgroups. However, from 2011 to 2017, across all areas, HF AAMR increased (large metropolitan APC +2.6%, 22.1–24.6; small metropolitan APC +1.9%, 24.2–26.3; and rural APC +1.7%, 28–30.2) predominantly more in large metropolitan areas compared with rural and small metropolitan areas.

Stroke-related AAMR declined consistently from 1999 to 2011 (large metropolitan APC −4.6%, 58–35.1; small metropolitan APC −4.5%, 63.6–39.3; and rural APC −3.8%, 67.8–43.7) with a relatively higher reduction in large metropolitan areas compared with rural areas across all ages, race, and ethnic groups. However, from 2011 to 2017, stroke-related AAMR slightly increased in large metropolitan areas (APC +0.9%, 35.1–36.3) while it showed a slow but consistent downtrend in rural (APC −1.2%, 43.7–40.3) and a flattened nonsignificant decline in small metropolitan areas (APC 0.2%, 39.3–38.5) areas. This increasing AAMR from stroke was specifically noted among older adults (APC +1.1%), Black people (APC +1.0%), and Hispanic people (APC +2.3%) in large metropolitan areas and younger adults in rural (APC +2.5%) and small metropolitan areas (+3.4%).

Rural areas experience higher mortality from IHD, HF, and stroke and have experienced fewer reductions in AAMR. Rural areas are home to a fifth of Americans and poor cardiovascular outcomes among rural areas could be because of poorer cardiovascular risk factor profile and access to health care. However, these trends could also be reflective of broader economic and demographic changes. Rural areas have seen significant economic decline and population loss. Rural hospital closures have been accelerating, hastened by the COVID-19 pandemic. For racial and ethnic minorities, structural inequities are magnified further in rural areas. Limitations of our analysis include data errors because of dependence on nonadjudicated death certificates to identify cause of death.

HF-related AAMR increase since 2011, across urban and rural areas, with notable spikes in racial minorities, consistent with previous reports.⁴ While rural areas had both the highest HF-related AAMR and greatest absolute increase in AAMR, AAMR APC was largest in large metropolitan areas. These findings highlight a need to identify factors driving this disturbing trend, particularly amongst vulnerable groups.

Last, our study also found a small increase in stroke-related AAMR among older adults, Hispanic people, and Black people in large metropolitan areas and younger adults in nonurban areas. These findings in Hispanic individuals are at odds with otherwise lower risk of cardiovascular mortality in that group.⁵ Studies on stroke-related mortality in Hispanic populations are limited, however, the higher AAMR could be from comparatively poor socioeconomic status and barriers to timely healthcare access.⁵ It is, therefore, critical to focus on vulnerable rural communities to further reduce cardiovascular mortality in the United States.

Sources of Funding

None.

Disclosures

Dr Mehra reports receiving travel support and consulting fees, paid to Brigham and Women’s Hospital, from Abbott; fees for serving on a steering committee from Medtronic and Janssen (Johnson and Johnson), fees for serving on a data and safety monitoring board from Mesoblast, consulting fees from Portola, Bayer and Triple Gene, and fees for serving as a scientific board member from NuPulseCV, Leviticus, and FineHeart. The other authors report no conflicts.