Skip to main content
NCBI home page
Neurology logoLink to Neurology
. 2023 Aug 1;101(5):e464–e474. doi: 10.1212/WNL.0000000000207446

Twenty-Year Disparity Trends in United States Stroke Death Rate by Age, Race/Ethnicity, Geography, and Socioeconomic Status

Adam de Havenon 1,✉, Lily W Zhou 1, Karen C Johnston 1, Neha S Dangayach 1, John Ney 1, Shadi Yaghi 1, Richa Sharma 1, Mehdi Abbasi 1, Alen Delic 1, Jennifer Juhl Majersik 1, Mohammad Anadani 1, David L Tirschwell 1, Kevin Navin Sheth 1
PMCID: PMC10401675  PMID: 37258298

Abstract

Background and Objectives

In 2017, the Centers for Disease Control and Prevention (CDC) issued an alert that, after decades of consistent decline, the stroke death rate levelled off in 2013, particularly in younger individuals and without clear origin. The objective of this analysis was to understand whether social determinants of health have influenced trends in stroke mortality.

Methods

We performed a longitudinal analysis of county-level ischemic and hemorrhagic stroke death rate per 100,000 adults from 1999 to 2018 using a Bayesian spatiotemporally smoothed CDC dataset stratified by age (35–64 years [younger] and 65 years or older [older]) and then by county-level social determinants of health. We reported stroke death rate by county and the percentage change in stroke death rate during 2014–2018 compared with that during 2009–2013.

Results

We included data from 3,082 counties for younger individuals and 3,019 counties for older individuals. The stroke death rate began to increase for younger individuals in 2013 (p < 0.001), and the slope of the decrease in stroke death rate tapered for older individuals (p < 0.001). During the 20-year period of our study, counties with a high social deprivation index and ≥10% Black residents consistently had the highest rates of stroke death in both age groups. Comparing stroke death rate during 2014–2018 with that during 2009–2013, larger increases in younger individuals' stroke death rate were seen in counties with ≥90% (vs <90%) non-Hispanic White individuals (3.2% mean death rate change vs 1.7%, p < 0.001), rural (vs urban) populations (2.6% vs 2.0%, p = 0.019), low (vs high) proportion of medical insurance coverage (2.9% vs 1.9%, p = 0.002), and high (vs low) substance abuse and suicide mortality (2.8 vs 1.9%, p = 0.008; 3.3% vs 1.5%, p < 0.001). In contrast to the younger individuals, in older individuals, the associations with increased death rates were with more traditional social determinants of health such as the social deprivation index, urban location, unemployment rate, and proportion of Black race and Hispanic ethnicity residents.

Discussion

Improvements in the stroke death rate in the United States are slowing and even reversing in younger individuals and many US counties. County-level increases in stroke death rate were associated with distinct social determinants of health for younger vs older individuals. These findings may inform targeted public health strategies.

In 2017, the Centers for Disease Control and Prevention (CDC) reported that “after more than 4 decades of decline, stroke death rates in the United States have declined more slowly, stalled, or reversed among some subpopulations in recent years.”1 In their analysis, which was reported in the Morbidity and Mortality Weekly Report, the inflection point occurred in 2013, and no clear cause was identified. Additional research supported by the CDC confirmed that the stroke death rate was increasing in younger individuals (ages 35–64 years) in 2016 vs 2010, which accounted for the stagnation of the stroke death rate across all individuals.2 To an increasing degree, social determinants of health are recognized as a basis for disparities in cardiovascular outcomes.3 Traditional social determinants of health have included economic stability, educational status, social support, physical environment, and healthcare access, but in the context of cardiovascular disease, race/ethnicity is perhaps the most important social determinant of health.4

Because national data are not linked to individual patient characteristics and state-level data are too broad, county-level data provide quantitative measures that can clarify the potential role of social determinants of health in trends for stroke mortality. Using a unique CDC methodology that interrogates the National Vital Statistics System (NVSS)5 and smooths the county-level stroke death rate over space, time, and age with a stratification by ages 35–64 years (younger) and 65 years or older (older), we explored the association of county-level social determinants of health with stroke death rate over a 20-year period from 1999 to 2018. When considering national trends in death rates in the United States, the primary data source is the NVSS, which reviews approximately 2.8 million death records every year to inform public health efforts.5 At the national or state level, these data are accurate for relatively common causes of death, including stroke, but when considering smaller geographical units, such as counties, the data become unreliable because of low cell counts.6 This creates difficulties when using the standard NVSS data through CDC WONDER, which is a collection of online databases that uses a rich ad hoc query system for the analysis of public health data.7

For example, CDC WONDER data for 2018 generate results for 3,148 counties but do not provide stroke death rate for 58.7% of these counties. The Bayesian dataset used in this analysis permitted continuous data from 1999 to 2018 in 98.2% of counties in younger individuals and 96.2% in older individuals. Hierarchical Bayesian spatiotemporal models such as this are a common and robust approach to minimizing bias from low cell count and allow exploration of death rates in smaller geographical units such as counties.8,9 We were able to leverage this unique methodology to provide preliminary insights into the dynamic changes in stroke death in the United States.

Methods

Data Source

We performed a post hoc analysis of a publicly available dataset produced by the CDC's Division for Heart Disease and Stroke Prevention entitled “Rates and Trends in Coronary Heart Disease and Stroke Mortality Data Among US Adults (35+) by County – 1999–2018.”10 This deidentified dataset did not require IRB approval for use. There were data available in 3,082 counties for younger individuals and 3,019 counties for older individuals. The unequal number of counties reflects the availability of CDC data, with suppression of data in counties with missing data. The number of counties in the United States has fluctuated over time, but remained approximately 3,140 during the period of this analysis,11 suggesting that we captured data on 98.2% and 96.2% of counties in younger individuals and older individuals, respectively.

Outcomes

The primary outcome was stroke death rate per 100,000 in US counties. Stroke death is defined by the NVSS as a primary cause of death with the ICD-10-CM codes I60-69, encompassing ischemic stroke, intracranial hemorrhagic stroke, and extra-axial hemorrhage.8 The CDC reported county-specific stroke mortality rate for 2 groups, those aged 35–64 years (younger) and those aged 65 years or older (older), using a Bayesian spatiotemporal model that is age standardized within each group. The secondary outcome was the percent change in stroke death rate for 2014–2018 compared with that during 2009–2013. The tertiary outcomes were as follows: (1) the binary outcome of a ≥1% increase in stroke death rate for the period 2014–2018 compared with that during 2009–2013 and (2) the percentage change in sequential 5-year periods (2004–2008 vs 1999–2003, 2009–2013 vs 2004–2008, and 2014–2018 vs 2009–2013).

Stratifications

Apart from the primary stratification by age, we also secondarily stratified by county-level social determinants of health including race/ethnicity, which is a social determinant of health that is self-reported for the US Census (2010). For the other secondary stratifications, we used data from the Area Health Resources File (AHRF) issued by the National Center for Health Workforce Analysis, Bureau of Health Workforce, Health Resources and Services Administration.12 We used the social deprivation index (SDI) from 2015, originally developed by Butler et al. in 2012, which is updated annually by the Robert Graham Center.13,14 The SDI is a validated composite measure of area-level deprivation widely used to quantify socioeconomic variation in health outcomes at the community level.15 The SDI is calculated using factor analysis methods for 7 demographic characteristics from the US Census Bureau's American Community Survey16 including percent living in poverty, percent with less than 12 years of education, percent of single-parent households, percent living in a rented housing unit, percent living in an overcrowded housing unit, percent of households without a car, and percent of unemployed adults younger than 65 years. The age-standardized mortality rate for substance abuse and self-harm (suicide) was derived from National Center for Health Statistics data in 2014.14,17 We picked time points for the data used in these stratifications that were closest to 2014 because they correspond to the inflection in the prior CDC analyses.1

The secondary stratifications of counties include census regions (Northeast vs Midwest vs South vs West), tertiles of the SDI (2015, Robert Graham Center), urban vs rural (2013, ERS Department of Agriculture), high poverty (20% or more of county residents in poverty, as measured by the American Community Survey 5-year estimates for 2008–2012), tertiles of unemployment rate (2013, Bureau of Labor Statistics), low education (20% or more of county residents aged 25–64 years without a high school diploma or equivalent between 2008 and 2012), tertiles of health insurance coverage (2013, US Census Bureau Small Area Health Insurance Estimates, percentage without health insurance younger than 65 years), proportion of county population that is non-Hispanic Black (2010 US Census, <10% vs ≥ 10%), proportion of county population that is Hispanic (any race) (2010 US Census, <10% vs ≥10%), proportion of county population that is American Indian/Alaskan Native (2010 US Census, <1% vs ≥1%), proportion of county population that is non-Hispanic White (2010 US Census, <90% vs ≥90%), tertiles of substance abuse mortality (2014, National Center for Health Statistics), and tertiles of self-harm mortality (2014, National Center for Health Statistics).

Analytic Approach

We reported the mean county-level point estimate for the stroke death rate from 1999 to 2018 after the primary stratification of age group. We showed the results graphically with 95% CIs (seen in Figure 1). After visually confirming the prior CDC report that the trend of decline in our primary outcome changed in 2013,1 we fit an interrupted time-series model that treated 2013 as the interruption. This model tests whether the trajectory of our primary outcome after 2013 was significantly different than before 2013, which was determined by fitting 2 ordinary least-squares regression-based models with Newey-West standard errors and comparing the difference-in-differences of slopes.18 This method was selected over a generalized least-squares method after confirming the absence of autocorrelation in the error distribution using the Cumby-Huizinga test.19

Figure 1. County-Level Stroke Death Rate per 100,000 Shown for Younger (A) and Older (B) Individuals From 1999 to 2018, With the Mean Value and 95% CI (A.a and B.a) and the Interrupted Time-Series Analysis (A.b and B.b).

Figure 1

Open in a new tab

To contrast temporal trends in our primary outcome, we computed mean rates for 5-year periods (1999–2003, 2004–2008, 2009–2013, and 2014–2018) and compared sequential 5-year periods to generate a percent change (e.g., 2004–2008 vs 1999–2003). These results are shown graphically for the continental United States, superimposed on a county-level map (seen in Figure 2). We then used the Student t test, ANOVA, and the χ2 test to calculate whether, based on the secondary stratifications, there were differences in the following: (1) the secondary outcome of percent change in 2014–2018 stroke death rate compared with that during 2009–2013, (2) the mean rate of the primary outcome for 2014–2018, and (3) the proportion of counties with a ≥1% increase in the stroke death rate in 2014–2018 compared with that during 2009–2013 and percentage change in sequential 5-year periods from 1999 to 2018. Because the prevalues (2009–2013) and postvalues (2014–2018) used in the calculation of the secondary outcome were highly correlated (ICC = 0.98) and the resulting percentage values had a standard distribution (eFigure 1, links.lww.com/WNL/C855), we did not use the ANCOVA methodology to baseline adjust the percentage values.20

Figure 2. County-Level 5-Year Comparisons in Mean Stroke Death Rate per 100,000 Shown for Younger (A) and Older (B) Individuals in the Continental United States.

Figure 2

Open in a new tab

The rows in the figure represent a percentage change between two 5-year periods, with blue colors representing decreases in the stroke death rate over that period and orange/red representing an increase in stroke death rate.

As a sensitivity analysis, we derived 1,000 bootstrapped values of the point estimate for the primary outcome sampled randomly from within the bounds of the 95% CI of the initial CDC Bayesian estimate. We used these data to confirm the validity of the CDC Bayesian point estimate by fitting an ANOVA model to compare the bootstrapped point estimate with the CDC point estimate.

Standard Protocol Approvals, Registrations, and Patient Consents

We used publicly available data for this study; therefore, IRB approval and patient consent were not required.

Data Availability

Data not published within this article will be made available by request from any qualified investigator.

Results

The county-level trend of stroke death rate from 1999 to 2018 per 100,000 is shown in Figure 1. For younger individuals, the rate in 1999 was 20.7 per 100,000 (95% CI 20.4–21.0) and that in 2018 was 17.1 per 100,000 (95% CI 16.8–17.3) (p < 0.001). For older individuals, the rate in 1999 was 478.8 per 100,000 (95% CI 475.4–482.2) and that in 2018 was 269.3 per 100,000 (95% CI 267.5–271.0) (p < 0.001). In the sensitivity analysis where we derived 1,000 bootstrapped values of the point estimate for the primary outcome sampled randomly from within the bounds of the 95% CI of the initial CDC Bayesian estimate point estimates, we found no difference between the 2 distributions of values (eTable 1, links.lww.com/WNL/C855).

Despite the reduction in stroke death rate over time, in 2013, there was a change in the slope of the curve for both age groups (Figure 1). In the interrupted time-series model, the change in slope was significant for both younger and older individuals (p < 0.001, p < 0.001). For the period 1999–2013, the yearly change in county-level stroke death rate per 100,000 for younger individuals was −4.8/year and for older individuals, it was −15.6/year. However, for the period 2013–2018, for younger individuals, there was an increase in the rate (0.1/year), and for older individuals, there was a decrease in the slope (−1.9/year).

The geographical spread of increased stroke death rate over time in US counties is seen graphically in Figure 2, highlighting how this phenomenon increasingly involves a larger proportion of counties in the United States for both younger and older individuals. For example, in younger individuals, in 2004–2008, 2009–2013, and 2014–2018, the percentage of counties with a ≥1% increase in stroke death rate compared with that in the immediate prior 5-year period was 9.4%, 13.0%, and 58.3%, respectively. For older individuals, the percentage of counties with a ≥1% increase in stroke death rate during the same time intervals was 0.2%, 0.3%, and 15.0%, respectively.

The number of counties after the secondary stratifications is summarized in eTable 2 (links.lww.com/WNL/C855). We saw significant differences in county-level stroke death rate for 2014–2018 in all the secondary stratifications (eTable 3). The highest rate of stroke death by US census region for younger individuals was the South (21.8, 95% CI 21.6–22.1), which was almost twice as high as the Northeast (11.0, 95% CI 10.4–11.7) (p < 0.001). The stroke death rate in younger individuals was higher in counties with high poverty (23.3, 95% CI 22.9–23.7) compared with low-poverty counties (15.1, 95% CI 14.9–15.4) (p < 0.001), in counties with a proportion of Black residents ≥10% (23.5, 95% CI 23.1–23.9), and in counties with a high SDI (22.6, 95% CI 22.2–22.9). Similar findings were present in older individuals, although of a smaller magnitude (eTable 3). The trends of these disparities are seen over time for younger individuals, presented in Figure 3 and eFigure 2, and older individuals, presented in Figure 4 and eFigure 3. These figures also present, during the 20-year period of our study, counties with a high SDI, and ≥10% Black residents consistently have the highest rates of stroke death in both age groups.

Figure 3. County-Level Stroke Mortality per 100,000 in Younger Individuals (Aged 35–64 Years) Shown After Non-Race/Ethnicity Secondary Stratifications, for the Years 1999–2018.

Figure 3

Open in a new tab

The race/ethnicity stratification figures are seen in eFigure 2 (links.lww.com/WNL/C855).

Figure 4. County-Level Stroke Mortality per 100,000 in Older Individuals (65 Years or Older) Shown After Non-Race/Ethnicity Secondary Stratifications, for the Years 1999–2018.

Figure 4

Open in a new tab

The race/ethnicity stratification figures are seen in eFigure 3 (links.lww.com/WNL/C855).

The secondary outcome of the percentage increase in stroke death rate for 2014–2018 compared with that during 2009–2013 was associated with distinct social determinants of health for younger vs older individuals (Table 1). For younger individuals, there were larger increases in counties with ≥90% (vs <90%) non-Hispanic White residents (3.2% vs 1.7%, p < 0.001), rural (vs urban) counties (2.6% vs 2.0%, p = 0.019), low (vs high) proportion of medical insurance coverage (2.9% vs 1.9%, p = 0.002), and high (vs low) substance abuse and self-harm (suicide) mortality (2.8 vs 1.9%, p = 0.008; 3.3% vs 1.5%, p < 0.001). In older individuals, associations were seen for more traditional social determinants of health including SDI (highest tertile vs lowest, −4.6% vs −7.5%, p < 0.001), urban (vs rural) location (−4.0% vs −7.7%, p < 0.001), unemployment (high vs low, −5.2% vs −8.4%, p < 0.001), and proportion of Black race (≥10% vs <10%, −3.0% vs −7.4%, p < 0.001) and Hispanic ethnicity (≥10% vs <10%, −4.6% vs −6.7%, <0.001) residents.

Table 1.

Percentage Change in County-Level Mean Stroke Mortality per 100,000 Shown by the Secondary Stratifications, Comparing 2014–2018 With 2009–2013

graphic file with name WNL-2023-000279t1.jpg

Open in a new tab

The tertiary outcome of a ≥1% percentage increase in county-level stroke death rate in 2014–2018 compared with that during 2009–2013 is summarized in eTable 3 (links.lww.com/WNL/C855) in the Supplement. Younger individuals in rural (vs urban) counties had a higher rate of this outcome (34.7% vs 30.9%, p = 0.028). Counties with ≥90% non-Hispanic White residents had the highest rate of the tertiary outcome (38.6%) in younger individuals. However, counties with higher proportions of non-Hispanic Black and Hispanic residents had lower rates of this outcome, and there was no association with the SDI. For older individuals, more traditional social determinants of health at the county level were associated with the tertiary outcome of a ≥1% percentage increase in county-level stroke death rate in 2014–2018 compared with that during 2009–2013, including urban location, high unemployment, and density of Black race and Hispanic ethnicity (eTable 4).

Discussion

In a nationally representative dataset of county-level stroke death rates from 1999 to 2018, we found that individuals aged 35–64 years had a significant increase in stroke death rate beginning in 2013 in contrast to individuals aged 65 years and older who are still experiencing a decline in stroke death rate, albeit at a significantly slower pace. Comparing 2014–2018 with 2009–2013, in individuals aged 35–64 years, 1,796 (58.3%) counties had a ≥1% increase in mean stroke death rate per 100,000, while in individuals aged 65 years and older, there were 452 (15.0%) counties with a ≥1% increase stroke death rate. These findings are similar to a prior analysis of CDC county-level stroke death rate data obtained in 2010–20162 and to trends in the cardiovascular death rate, which increased in recent years for subpopulations of individuals younger than 65 years.21 However, unlike prior analyses that focused on social determinants of health and stroke mortality,22,23 we were able to broaden the scope of our analysis to a national level using a relevant unit of measure—the county.

Similar to prior analyses, we found that social determinants of health remain associated with the absolute stroke death rate in US counties, which remains highest overall in counties with a high proportion of non-Hispanic Black residents in addition to the associated and overlapping factors of counties with a high SDI and counties in the Southern United States.24 During all periods of our study and specifically within the last 5 years (2014–2018), the absolute stroke death rate was highest in counties with more Black individuals. This is consistent with findings from many longitudinal cohorts showing race-based disparities in the United States with a higher stroke incidence in Black individuals compared with that in White individuals, which is most pronounced in patients younger than 65 years.25-27 The high rate of stroke death in Black communities is among the most persistent and alarming injustices in cardiovascular disease.28 Black communities have less access to resources and opportunities, more exposure to vascular risk, and experience discrimination and racism, all of which have pervasive effects on cardiovascular outcomes.3 While further exploration of the root causes and potential solutions are beyond the scope of this article, we acknowledge that this health disparity is a direct result of long-standing policies, practices, and attitudes toward racial equality and social justice in the United States.

Contrary to these long-standing health disparities, we found that the more recent increases in stroke mortality in younger individuals were associated with rural counties where ≥90% of residents were White and counties with higher levels of substance abuse mortality and suicide. The underlying causes for these findings require more investigation but mirror trends in life expectancy in the United States where recent increases in overall mortality have been seen in White individuals during midlife but not among other races and ethnicities at midlife nor in those older than 65 years of any race/ethnicity.29 The inflection point for stroke death rate in 2013 also occurs coincident with sharp increases in deaths due to drug overdose starting in the United States from opiates, psychoactive stimulants (predominantly methamphetamine), and cocaine, principally among White individuals.30 Survey studies in the United States in recent decades also show higher illicit drug use and binge alcohol drinking during adolescence31 and young adulthood32,33 among White respondents. Combining higher levels of substance abuse mortality with suicide mortality, which together are called “deaths of despair,” the associated increase in stroke death seems to be potentially influenced by overall trends in the United States.34,35 Deaths of despair is a unique phenomenon among middle-aged non-Hispanic White individuals in the United States first described in 2015 and is highly correlated with other socioeconomic factors that could influence stroke mortality in this demographic, including declining income, social stability, and mental health.36 Of course, there may be other shared upstream social determinants of health that disproportionately affect younger individuals, which are more challenging to measure.37

In addition, there are other factors that likely contribute to this trend. A previous population-based study using 20% of all US hospital discharges from 2003 to 2012 showed increases in stroke hospitalizations among patients of all race/ethnicities aged 35–54 years and among White and Hispanic patients aged 18–34 years.38 Within these young patients, the prevalence of ≥3 traditional vascular risk factors roughly doubled during the study period.38 Data from the Greater Cincinnati and Northern Kentucky regions similarly showed increasing incidence of stroke in those aged 20–54 years and increasing prevalence of dyslipidemia and heart disease.39 An increased burden of disease from nontraditional risk factors, such as secondary hypertension, vasculopathy, vasospasm, and endocarditis, among the younger individuals may also partially explain a recent finding of stagnating 90-day mortality among ischemic stroke patients aged 70 years or younger but not older patients from 2012 to 2018.40

Our study has multiple limitations. First, the publicly available dataset produced by the CDC does not distinguish between stroke subtypes (ischemic stroke vs intraparenchymal hemorrhage vs extra-axial hemorrhages), which are associated with different risk factors and case fatalities.41 Because of this, we were unable to further analyze temporal trends by stroke subtypes. Second, stroke mortality with stratification by sex or gender was not available to allow for examination of sex-related or gender-related differences in temporal trends. Previously described sex differences in stroke mortality show modification by age with lower stroke mortality in midlife (aged 45–84 years) and higher in later life (older than 85 years) for women compared with men.42 The impact that sex has in differential exposure to intermediary determinants of health such as material circumstances (e.g., housing, food security) and psychosocial factors (e.g., childhood adversity, community violence) is increasingly better characterized.43 With access to more granular data, the examination of both sex and gender differences in temporal trends in stroke mortality may yield insights on persistent causes of disparities, generational effects, and potential targets for interventions. In addition, the lack of data granularity restricted the ability to look at the impact of stroke risk factors in this dataset. Last, both mortality rates and covariates used for stratification were aggregated at the county level. Because of this, heterogeneity in individuals living within the same county could not be examined, and insights from our analysis revolved around both an individual's social determinants of health and those of their neighbors.

Most state or county-level policy interventions to improve stroke outcomes in the United States have been focused on systems of care optimization to provide acute stroke treatments to more individuals.44-47 While these developments have been important and largely successful, they do not address the difficulty of improving population-level premorbid risk factors for stroke and stroke mortality, which evidence suggests are increasing over time.38,39,48-50 Ultimately, reversing these trends will require dedicated and consistent public health interventions targeting different goals for unique communities. The first step is understanding what is associated with communities most affected by either high stroke mortality or increasing stroke mortality, so appropriate interventions can be developed.

The trend of consistently decreasing stroke death rate in the United States stalled in 2013 due to a spreading county-level increase in stroke death among those younger than 65 years. The increase in stroke mortality in younger individuals was associated with poor rural counties with more non-Hispanic White residents and higher rates of substance abuse and suicide mortality. In older individuals, changes in stroke death rate were associated with more traditional social determinants of health. These findings could inform public health strategies that are targeted to age groups.

Acknowledgment

This article was prepared using the Centers for Disease Control and Prevention's Division for Heart Disease and Stroke Prevention dataset “Rates and Trends in Coronary Heart Disease and Stroke Mortality Data Among US Adults (35+) by County – 1999–2018” and does not necessarily reflect the opinions or views of the CDC.

Glossary

AHRF

Area Health Resources File

CDC

Centers for Disease Control and Prevention

NVSS

National Vital Statistics System

SDI

Social Deprivation Index

Appendix. Authors

Appendix.

Open in a new tab

Footnotes

Editorial, page 195

Study Funding

Dr. de Havenon reports NIH/NINDS funding (K23NS105924). Dr. Zhou reports CIHR funding (RN387091 - 420683). Dr. Sheth reports funding by NIH-NINDS U01NS106513, R01NS11072, R01NR018335, R01EB031114, R01MD016178, R03NS112859, U24NS107215, U24NS107136, and American Heart Association 17CSA33550004. Dr. Johnston reports NIH/NCATS funding (UL1TR003015 and KL2TR003016) and NIH/NINDS funding (R44 NS120798 and U01 NS086872). Dr. Majersik reports funding by NIH/NINDS U24NS107228 and U24NS107156.

Disclosure

A. de Havenon has received investigator-initiated clinical research funding from the AAN, has received consultant fees from Integra and Novo Nordisk, royalty fees from UpToDate, and has equity in TitinKM and Certus; K.N. Sheth reports compensation from Sense and Zoll for data and safety monitoring services, compensation from Cerevasc for consultant services, compensation from Rhaeos for consultant services, compensation from Certus for consultant services, and equity in Alva Health; K.C. Johnston is a consultant for Biogen as chair of multiple Independent Data Monitoring Committees; N.S. Dangayach has received investigator-initiated clinical research funding from AAN, The Aneurysm and AVM Foundation, and DiyaHealth; J.J. Majersik is an Associate Editor for Stroke and reports reviewer fees for UpToDate; The rest of authors report to no disclosures. Go to Neurology.org/N for full disclosures.

References

  • 1.Yang Q. Vital signs: recent trends in stroke death rates—United States, 2000-2015. MMWR Morb Mortal Wkly Rep. 2017;66(35):933-939. doi: 10.15585/mmwr.mm6635e1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Hall EW, Vaughan AS, Ritchey MD, Schieb L, Casper M. Stagnating national declines in stroke mortality mask widespread county-level increases, 2010-2016. Stroke. 2019;50(12):3355-3359. doi: 10.1161/STROKEAHA.119.026695 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Powell-Wiley TM, Baumer Y, Baah FO, et al. Social determinants of cardiovascular disease. Circ Res. 2022;130(5):782-799. doi: 10.1161/CIRCRESAHA.121.319811 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Javed Z, Haisum Maqsood M, Yahya T, et al. Race, racism, and cardiovascular health: applying a social determinants of health framework to racial/ethnic disparities in cardiovascular disease. Circ Cardiovasc Qual Outcomes. 2022;15(1):e007917. doi: 10.1161/CIRCOUTCOMES.121.007917 [DOI] [PubMed] [Google Scholar]
  • 5.NVSS - mortality data. Published March 21, 2022. Accessed March 28, 2022. cdc.gov/nchs/nvss/deaths.htm
  • 6.Khana D, Rossen LM, Hedegaard H, Warner M. A bayesian spatial and temporal modeling approach to mapping geographic variation in mortality rates for subnational areas with R-INLA. J Data Sci. 2018;16(1):147-182. [PMC free article] [PubMed] [Google Scholar]
  • 7.CDC WONDER. Accessed April 3, 2022. wonder.cdc.gov/
  • 8.Quick H. Estimating county-level mortality rates using highly censored data from CDC WONDER. Prev Chronic Dis. 2019;16. doi: 10.5888/pcd16.180441 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Rouamba T, Samadoulougou S, Tinto H, Alegana VA, Kirakoya-Samadoulougou F. Bayesian spatiotemporal modeling of routinely collected data to assess the effect of health programs in malaria incidence during pregnancy in Burkina Faso. Sci Rep. 2020;10(1):2618. doi: 10.1038/s41598-020-58899-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Rates and trends in coronary heart disease and stroke mortality data among US adults (35+) by county—1999-2018 | chronic disease and health promotion data & indicators. Accessed March 28, 2022. chronicdata.cdc.gov/Heart-Disease-Stroke-Prevention/Rates-and-Trends-in-Coronary-Heart-Disease-and-Str/9cr5-2tt7
  • 11.U.S. Census Bureau. Substantial changes to counties and county equivalent entities: 1970-present. Accessed July 14, 2022. census.gov/programs-surveys/geography/technical-documentation/county-changes.html
  • 12.Area health resources files. Accessed March 22, 2022. data.hrsa.gov/topics/health-workforce/ahrf
  • 13.Butler DC, Petterson S, Phillips RL, Bazemore AW. Measures of social deprivation that predict health care access and need within a rational area of primary care service delivery. Health Serv Res. 2013;48(2pt1):539-559. doi: 10.1111/j.1475-6773.2012.01449.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Social deprivation index (SDI). Accessed April 1, 2022. graham-center.org/content/brand/rgc/maps-data-tools/social-deprivation-index.html
  • 15.White-Williams C, Rossi LP, Bittner VA, et al. Addressing social determinants of health in the care of patients with heart failure: a scientific statement from the American Heart Association. Circulation. 2020;141(22):e841-e863. doi: 10.1161/CIR.0000000000000767 [DOI] [PubMed] [Google Scholar]
  • 16.U.S. Census Bureau. American Community Survey (ACS). Accessed January 15, 2023. census.gov/programs-surveys/acs [Google Scholar]
  • 17.Dwyer-Lindgren L, Bertozzi-Villa A, Stubbs RW, et al. Trends and patterns of geographic variation in mortality from substance use disorders and intentional injuries among US counties, 1980-2014. JAMA. 2018;319(10):1013-1023. doi: 10.1001/jama.2018.0900 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Linden A. Conducting interrupted time-series analysis for single- and multiple-group comparisons. Stata J Promot Commun Stat Stata. 2015;15(2):480-500. doi: 10.1177/1536867X1501500208 [DOI] [Google Scholar]
  • 19.Cumby R, Huizinga J. Testing the autocorrelation structure of disturbances in ordinary least squares and instrumental variables regressions. Econometrica. 1992;60(1):185-195. [Google Scholar]
  • 20.Vickers AJ. The use of percentage change from baseline as an outcome in a controlled trial is statistically inefficient: a simulation study. BMC Med Res Methodol. 2001;1(1):6. doi: 10.1186/1471-2288-1-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Cross SH, Mehra MR, Bhatt DL, et al. Rural-Urban differences in cardiovascular mortality in the US, 1999-2017. JAMA. 2020;323(18):1852-1854. doi: 10.1001/jama.2020.2047 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Pedigo A, Aldrich T, Odoi A. Neighborhood disparities in stroke and myocardial infarction mortality: a GIS and spatial scan statistics approach. BMC Public Health. 2011;11:644. doi: 10.1186/1471-2458-11-644 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hanchate AD, Schwamm LH, Huang W, Hylek EM. Comparison of ischemic stroke outcomes and patient and hospital characteristics by race/ethnicity and socioeconomic status. Stroke. 2013;44(2):469-476. doi: 10.1161/STROKEAHA.112.669341 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Yadav RS, Chaudhary D, Avula V, et al. Social determinants of stroke hospitalization and mortality in United States' Counties. J Clin Med. 2022;11(14):4101. doi: 10.3390/jcm11144101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Howard VJ, Kleindorfer DO, Judd SE, et al. Disparities in stroke incidence contributing to disparities in stroke mortality. Ann Neurol. 2011;69(4):619-627. doi: 10.1002/ana.22385 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kissela B, Schneider A, Kleindorfer D, et al. Stroke in a biracial population: the excess burden of stroke among blacks. Stroke. 2004;35(2):426-431. doi: 10.1161/01.STR.0000110982.74967.39 [DOI] [PubMed] [Google Scholar]
  • 27.Gardener H, Sacco RL, Rundek T, Battistella V, Cheung YK, Elkind MSV. Race and ethnic disparities in stroke incidence in the Northern Manhattan Study. Stroke. 2020;51(4):1064-1069. doi: 10.1161/STROKEAHA.119.028806 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Churchwell K, Elkind MSV, Benjamin RM, et al. Call to action: structural racism as a fundamental driver of health disparities: a presidential advisory from the American Heart Association. Circulation. 2020;142(24):e454-e468. doi: 10.1161/CIR.0000000000000936 [DOI] [PubMed] [Google Scholar]
  • 29.Case A, Deaton A. Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century. Proc Natl Acad Sci U S A. 2015;112(49):15078-15083. doi: 10.1073/pnas.1518393112 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.National Institute on Drug Abuse. Overdose Death Rates. Published January 20, 2022. Accessed April 25, 2022. nida.nih.gov/drug-topics/trends-statistics/overdose-death-rates [Google Scholar]
  • 31.Wu LT, Woody GE, Yang C, Pan JJ, Blazer DG. Racial/ethnic variations in substance-related disorders among adolescents in the United States. Arch Gen Psychiatry. 2011;68(11):1176-1185. doi: 10.1001/archgenpsychiatry.2011.120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Vaughn MG, Salas-Wright CP, Cordova D, Nelson EJ, Jaegers L. Racial and ethnic trends in illicit drug use and binge drinking among adolescent and young adult offenders in the United States. J Crim Justice. 2018;59:71-80. doi: 10.1016/j.jcrimjus.2017.05.009 [DOI] [Google Scholar]
  • 33.McCabe SE, Morales M, Cranford JA, Delva J, McPherson MD, Boyd CJ. Race/ethnicity and gender differences in drug use and abuse among college students. J Ethn Subst Abuse. 2007;6(2):75-95. doi: 10.1300/J233v06n02_06 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Schmid CH. Increased mortality for white middle-aged Americans not fully explained by causes suggested. Proc Natl Acad Sci U S A. 2016;113(7). doi: 10.1073/pnas.1522239113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Meara E, Skinner J. Losing ground at midlife in America. Proc Natl Acad Sci U S A. 2015;112(49):15006-15007. doi: 10.1073/pnas.1519763112 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Beseran E, Pericàs JM, Cash-Gibson L, Ventura-Cots M, Porter KMP, Benach J. Deaths of despair: a scoping review on the social determinants of drug overdose, alcohol-related liver disease and suicide. Int J Environ Res Public Health. 2022;19(19):12395. doi: 10.3390/ijerph191912395 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Sterling P, Platt ML. Why deaths of despair are increasing in the US and not other industrial nations-insights from neuroscience and anthropology. JAMA Psychiatry. 2022;79(4):368-374. doi: 10.1001/jamapsychiatry.2021.4209 [DOI] [PubMed] [Google Scholar]
  • 38.George MG, Tong X, Bowman BA. Prevalence of cardiovascular risk factors and strokes in younger adults. JAMA Neurol. 2017;74(6):695. doi: 10.1001/jamaneurol.2017.0020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Kissela BM, Khoury JC, Alwell K, et al. Age at stroke: temporal trends in stroke incidence in a large, biracial population. Neurology. 2012;79(17):1781-1787. doi: 10.1212/WNL.0b013e318270401d [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Anand SK, Benjamin WJ, Adapa AR, et al. Trends in acute ischemic stroke treatments and mortality in the United States from 2012 to 2018. Neurosurg Focus. 2021;51(1):E2. doi: 10.3171/2021.4.FOCUS21117 [DOI] [PubMed] [Google Scholar]
  • 41.Andersen KK, Olsen TS, Dehlendorff C, Kammersgaard LP. Hemorrhagic and ischemic strokes compared: stroke severity, mortality, and risk factors. Stroke. 2009;40(6):2068-2072. doi: 10.1161/STROKEAHA.108.540112 [DOI] [PubMed] [Google Scholar]
  • 42.Reeves MJ, Bushnell CD, Howard G, et al. Sex differences in stroke: epidemiology, clinical presentation, medical care, and outcomes. Lancet Neurol. 2008;7(10):915-926. doi: 10.1016/S1474-4422(08)70193-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Miani C, Wandschneider L, Niemann J, Batram-Zantvoort S, Razum O. Measurement of gender as a social determinant of health in epidemiology—a scoping review. PLoS One. 2021;16(11):e0259223. doi: 10.1371/journal.pone.0259223 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Adeoye O, Nyström KV, Yavagal DR, et al. Recommendations for the establishment of stroke systems of care: a 2019 update. Stroke. 2019;50(7):e187-e210. doi: 10.1161/STR.0000000000000173 [DOI] [PubMed] [Google Scholar]
  • 45.Reduce stroke deaths — HDS-03 - healthy people 2030. Accessed January 17, 2023. health.gov/healthypeople/objectives-and-data/browse-objectives/heart-disease-and-stroke/reduce-stroke-deaths-hds-03
  • 46.Rudd AG, Williams LS. Advances in health policy and outcomes. Stroke. 2009;40(5):e301-e304. doi: 10.1161/STROKEAHA.108.544551 [DOI] [PubMed] [Google Scholar]
  • 47.Centers for Disease Control and Prevention. Stroke Systems of Care: State Policy Interventions by Evidence Level. Published May 13, 2021. Accessed January 17, 2023. cdc.gov/dhdsp/policy_resources/stroke_systems_of_care/evidence_interventions.htm [Google Scholar]
  • 48.Tsao CW, Aday AW, Almarzooq ZI, et al. Heart Disease and stroke statistics—2022 update: a report from the American Heart Association. Circulation. 2022;145(8):e153-e639. doi: 10.1161/CIR.0000000000001052 [DOI] [PubMed] [Google Scholar]
  • 49.Mohebi R, Chen C, Ibrahim NE, et al. Cardiovascular disease projections in the United States Based on the 2020 census estimates. J Am Coll Cardiol. 2022;80(6):565-578. doi: 10.1016/j.jacc.2022.05.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.de Havenon A, Prabhakaran S, Stulberg E, et al. Blood pressure and low‐density lipoprotein cholesterol in stroke survivors versus controls in the United States from 1999 to 2018. Stroke Vasc Interv Neurol. 2022;2(3):e000286. doi: 10.1161/SVIN.121.000286 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data not published within this article will be made available by request from any qualified investigator.

Articles from Neurology are provided here courtesy of American Academy of Neurology

RESOURCES