eTable 1. Definitions of Risk Factors (ICD-9-CM).
eTable 2. Ischemic Stroke Hospitalization Rates/10,000 hospitalizations by Age and Sex, 1995-96 – 2011-12.
eTable 3. Discharge Destination for Acute Ischemic Stroke Admissions, National Inpatient Sample (NIS) 2003-2012.
eTable 4. Discharge Destination for Intracerebral Hemorrhage Admissions, National Inpatient Sample (NIS) 2003-2012.
eTable 5. Discharge Destination for Subarachnoid Hemorrhage Admissions, National Inpatient Sample (NIS) 2003-2012.
eFigure. Acute Ischemic Stroke hospitalization Rates/10,000 Hospitalizations by Age and Sex, 1995-1996—2011-2012.
Customize your JAMA Network experience by selecting one or more topics from the list below.
George MG, Tong X, Bowman BA. Prevalence of Cardiovascular Risk Factors and Strokes in Younger Adults. JAMA Neurol. 2017;74(6):695–703. doi:10.1001/jamaneurol.2017.0020
Are stroke hospitalization rates for younger adults continuing to increase, and is the prevalence of associated risk factors increasing among those hospitalized for acute stroke?
This analysis found that stroke hospitalization rates from 2003 to 2012 significantly increased for acute ischemic stroke hospitalization rates among men (41.5%) and women (30%) aged 35 to 44 years, with a near doubling of the prevalence of 3 or more of 5 common stroke risk factors among both men and women aged 18 to 64 years hospitalized for acute ischemic stroke.
Hospitalization rates for acute ischemic stroke in younger adults continued to increase since 1995-1996, coexistent with increasing prevalence of stroke risk factors.
While stroke mortality rates have decreased substantially in the past 2 decades, this trend has been primarily limited to older adults. Increasing trends in stroke incidence and hospitalizations have been noted among younger adults, but there has been concern that this reflected improved diagnosis through an increased use of imaging rather than representing a real increase.
To determine whether stroke hospitalization rates have continued to increase and to identify the prevalence of associated stroke risk factors among younger adults.
Design, Setting, and Participants
Hospitalization data from the National Inpatient Sample from 1995 through 2012 were used to analyze acute stroke hospitalization rates among adults aged 18 to 64 years. Hospitalization data from 2003 to 2012 were used to identify the prevalence of associated risk factors for acute stroke. Acute stroke hospitalizations were identified by the principal International Classification of Diseases, Ninth Revision, Clinical Modification code and associated risk factors were identified by secondary International Classification of Diseases, Ninth Revision, Clinical Modification codes for each hospitalization.
Main Outcomes and Measures
Trends in acute stroke hospitalization rates by stroke type, age, sex, and race/ethnicity, as well as the prevalence of associated risk factors by stroke type, age, and sex.
The 2003-2004 set included 362 339 hospitalizations and the 2011-2012 set included 421 815 hospitalizations. The major findings in this study are as follows: first, acute ischemic stroke hospitalization rates increased significantly for both men and women and for certain race/ethnic groups among younger adults aged 18 to 54 years; they have almost doubled for men aged 18 to 34 and 35 to 44 years since 1995-1996, with a 41.5% increase among men aged 35 to 44 years from 2003-2004 to 2011-2012. Second, the prevalence of stroke risk factors among those hospitalized for acute ischemic stroke continued to increase from 2003-2004 through 2011-2012 for both men and women aged 18 to 64 years (range of absolute increase: hypertension, 4%-11%; lipid disorders, 12%-21%; diabetes, 4%-7%; tobacco use, 5%-16%; and obesity, 4%-9%). Third, the prevalence of having 3 to 5 risk factors increased from 2003-2004 through 2011-2012 (men: from 9% to 16% at 18-34 years, 19% to 35% at 35-44 years, 24% to 44% at 45-54 years, and 26% to 46% at 55-64 years; women: 6% to 13% at 18-34 years, 15% to 32% at 35-44 years, 25% to 44% at 45-54 years, and 27% to 48% at 55-65 years; P for trend < .001). Finally, hospitalization rates for intracerebral hemorrhage and subarachnoid hemorrhage remained stable, with the exception of declines among men and non-Hispanic black patients aged 45 to 54 with subarachnoid hemorrhage (13.2/10 000 to 10.3/10 000 hospitalizations and 15.8/10 000 to 11.5/10 000 hospitalizations, respectively).
Conclusions and Relevance
The identification of increasing hospitalization rates for acute ischemic stroke in young adults coexistent with increasing prevalence of traditional stroke risk factors confirms the importance of focusing on prevention in younger adults.
Stroke was the third leading cause of death in the United States from 1938 to 2007, and dropped to the fifth leading cause of death in 2013.1 Yet, the relative rate of decline in mortality among those aged 65 years and older is approximately double the rate of decline among those aged 45 to 64 years.1 The overall decline in stroke mortality over the past 50 years reflects the improved control of hypertension through efforts such as the National Heart, Lung, and Blood Institute’s Joint National Commission guidelines on hypertension control, as well as increased treatment of atrial fibrillation with anticoagulants, increased use of aspirin for secondary prevention, improvements in stroke care through structured quality improvement programs, and the development of stroke systems of care that promote the use of evidence-based care.2
Against this backdrop of overall declining stroke mortality rates is suggestive evidence that acute stroke incidence rates among younger adults are increasing.3 In 2011, we reported that acute stroke hospitalization rates were increasing among young adults aged 15 to 44 years.4 Studies since then suggest that this increase is likely not due to increased use of imaging.3,5 Additionally, while rates of some traditional stroke risk factors have declined over the past decade in the overall adult population (tobacco smoking), many others have shown little change or have increased (overweight, obesity, diabetes, hypertension, and hypercholesterolemia).6,7 It is important to understand whether these risk factor patterns are similar among those experiencing an acute stroke. To identify whether previous trends in stroke hospitalization rates and the patterns of associated risk factors have continued, we examined the hospitalization rates for acute stroke by stroke type along with the prevalence of associated risk factors by sex and age group among younger adults (aged 18 to 64 years).
The National Inpatient Sample (NIS)8 is part of the Healthcare Cost and Utilization Project (HCUP),8 sponsored by the Agency for Healthcare Research and Quality. The NIS is a database of hospital inpatient stays derived from billing data submitted by hospitals to statewide data organizations across the United States. It is a stratified sample that represents discharges from approximately 20% of all community hospitals participating in HCUP; hospitals are selected based on 5 characteristics: rural/urban location, hospital size, region, teaching status, and ownership. There were 37 states that participated in 2003 and the number of states increased each year to 44 states in 2012.8 Because the data are publicly available and do not contain direct personal identifiers, this study was exempt from review by the institutional review board, per Agency for Healthcare Research and Quality guidelines.
International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes were used to identify hospitalizations for subarachnoid hemorrhage (430), intracerebral hemorrhage (431), and acute ischemic stroke (AIS) (433.01, 433.11, 433.21, 433.31, 433.81, 433.91, 434.01, 434.11, 434.91, and 436). For each hospitalization, only the principal diagnosis code was used to identify events, which were then assigned to 1 of 3 mutually exclusive categories: subarachnoid hemorrhage (SAH), intracerebral hemorrhage (ICH), and AIS.
The ICD-9-CM code 436 was included in ischemic stroke as recommended by HCUP when doing trend over time analysis across the time as coding instructions changed for this code. The use of this code underwent a change in fiscal year 2005, where coders were instructed to use code 434.91 rather than 436.
The unit of analysis was hospital discharge. In this analysis, hospital discharge records were included if the patient was 18 years or older at the time of the hospital admission. We combined NIS annual data to create 5 consecutive 2-year periods from 2003-2004 through 2011-2012. Four age groups were defined: 18 to 34 years, 35 to 44 years, 45 to 54 years, and 55 to 64 years. Separate stroke hospitalization rates for men and women were estimated for each stroke type and age group. Risk factor prevalence was estimated using secondary ICD-9-CM codes, taking into account changes in ICD-9-CM coding over time (eTable 1 in the Supplement). Common risk factors for stroke in this population were assessed for linear trends using orthogonal polynomial contrasts. We also looked at the combination of the following 5 traditional cardiovascular risk factors related to the stroke: hypertension, diabetes, lipid disorders, obesity, and tobacco use, which were identified as none, 1 to 2 risk factors, or 3 to 5 risk factors to address the burden of multiple stroke risk factors. Linear trends were not reported if the data were too sparse to report based on HCUP reporting guidelines. Further, trends in stroke hospitalizations for those aged 18 to 34, 35 to 44, 45 to 54, and 55 to 64 years by sex from 1995 to 2012 were assessed across 5 periods by using linear regression models.
National estimates were obtained by using individual discharge sampling weights. As a result of changes implemented in the redesign of 2012 NIS data, we used trend weights developed by the Agency for Healthcare Research and Quality to make estimates comparable for data prior to 2012. All statistical analyses were conducted using SAS version 9.3–callable SUDAAN (Research Triangle Institute) to account for the multistage, disproportionate stratified sampling design.
Among acute ischemic stroke hospitalizations, rates increased from 2003-2004 through 2011-2012 for all race/ethnic groups; men; and women in age groups 18 to 34, 35 to 44, and 45 to 54 years, except among non-Hispanic black individuals and other race groups for ages 18 to 34 years. Among those aged 55 to 64 years, only Hispanic individuals showed a significant increase in AIS hospitalization rates (from 181.8/10 000 to 199.7/10 000; P = .02), but there was no change in stroke hospitalization rates in this age group by sex or among other non-Hispanic race groups (Table 1).
The hospitalization rate for men aged 18 to 34 years increased from 11.2 to 18.0 per 10 000 hospitalizations; for women aged 18 to 34 years, it increased from 3.8 to 5.8 per 10 000 hospitalizations; for men aged 35 to 44 years, it increased from 37.7 to 68.2 per 10 000 hospitalizations; and for women aged 35 to 44, it increased from 24.8 to 35.8 per 10 000 hospitalizations from 1995-1996 through 2011-2012 (relative percentage increases were 74.8%, 65.7%, 91.0%, and 53.6%, respectively) (eFigure and eTable 2 in the Supplement).
Table 2 shows the prevalence of stroke-related risk factors among adults admitted for AIS by age and sex. Across all age groups and both sexes, increasing prevalence was seen in hypertension, lipid disorders, diabetes, tobacco use, obesity, and among multiple risk factors. The prevalence of 3 or more traditional risk factors (hypertension, diabetes, lipid disorders, obesity, and tobacco use) roughly doubled among all age groups for both men and women, while the prevalence of none of these risk factors decreased. Notably, during the study period of 2003-2004 through 2011-2012, hypertension increased from 34.0% to 41.3%, and lipid disorders increased from 14.6% to 29.1% among men aged 18 to 34 years. Smaller increases were seen among women aged 18 to 34 years, although in 2011-2012, 30.7% had hypertension, 21.7% had lipid disorders, 26.5% used tobacco, and 15.7% were obese. One in 7 men aged 18 to 34 years and 1 in 3 men aged 35 to 44 years had 3 to 5 traditional risk factors. There were small but significant increases in the prevalence of atrial fibrillation among the older 2 age groups for both men and women. Among those with AIS, the prevalence of ischemic heart disease declined among men aged 45 to 54 and 55 to 64 years, yet increased among women aged 18 to 34 years. There were no changes among the other sex/age groups (Table 2).
There were no changes in ICH hospitalization rates by age, race/ethnicity, or sex from 2003-2004 through 2011-2012 (Table 3). However, men had significantly higher hospitalization rates than women among all age groups. The relative difference between men and women decreased with increasing age group.
Among those with ICH, the prevalence of hypertension among men in 2011-2012 was 44.1% (aged 18-34 years), 73.8% (aged 35-44 years), and 81.4% (aged 45-54 and 55-64 years), and the prevalence of hypertension among men aged 18 to 34 and 35 to 44 years with ICH was higher than the prevalence among their counterparts with AIS. Rates of hypertension among women with ICH were comparable with those with AIS (Table 2 and Table 4). Tobacco use increased among all male age groups and among women aged 18 to 34, 45 to 54, and 55 to 64 years. There was a significant trend toward increasing prevalence of having multiple stroke risk factors among men and women across all age groups among those hospitalized for ICH (Table 4).
From 2003-2004 through 2011-2012, hospitalization rates for SAH declined among all age groups, but significantly only among men aged 45 to 54 years (13.2/10 000 to 10.3/10 000; P = .01), non-Hispanic white individuals (14.5/10 000 to 11.9/10 000; P = .045), and non-Hispanic black individuals (15.8/10 000 to 11.5/10 000; P = .006). Rates of SAH among Hispanic individuals aged 45 to 54 and 55 to 64 years were significantly higher compared with non-Hispanic white and non-Hispanic black individuals. The hospitalization rate for men aged 18 to 35 years were 3 times the rate of women, but women had significantly higher hospitalization rates for SAH among those aged 45 to 54 and 55 to 64 years (Table 3).
Significant increases in the prevalence of stroke risk factors were seen among hospitalizations for SAH across all age groups of men for hypertension, obesity, and tobacco use and for lipid disorders and diabetes among men aged 35 to 44, 45 to 54, and 55 to 64 years. Among women with acute SAH hospitalizations, the prevalence of hypertension increased among all age groups, and the prevalence of diabetes, obesity, and lipid disorders increased among age groups 35 to 44, 45 to 54, and 55 to 64 years. Women aged 45 to 54 years had an increased prevalence of alcohol abuse and tobacco use. The presence of multiple risk factors (inclusive of hypertension, diabetes, lipid disorders, obesity, and tobacco use) shifted toward a greater number of risk factors among both men and women across all age groups (Table 5).
For most age groups and stroke types, there were trends of decreasing in-hospital death and increasing trends of transfer to skilled nursing units/rehabilitation facilities/long-term care facilities (eTable 3, eTable 4, and eTable 5 in the Supplement). While this data set lacks important socioeconomic data, we found that among all age groups and stroke types, between 56% and 60% of stroke hospitalizations occurred among the lower 50th percentile of median household income of the patients’ zip code, with no change over time.
Overall, AIS hospitalizations among those aged 18 to 64 years increased from an average of 141 474 per year in 2003-2004 to 171 386 per year in 2011-2012. Hospitalizations for SAH declined from an average of 17 598 per year in 2003-2004 to 15 974 per year in 2011-2012. There was little change in the number of hospitalizations for ICH (22 097/year in 2003-2004 vs 23 545/year in 2011-2012).
The major findings in this study are as follows: first, AIS hospitalization rates increased for both men and women and for certain race/ethnic groups among younger adults aged 18 to 54, and they have almost doubled for men aged 18 to 34 and 35 to 44 years since 1995-1996, while rates for men and women aged 55 to 64 have not changed from 2003-2004. Second, the hospitalization rates for ICH did not change from 2003-2004 to 2011-2012 among those aged 18 to 64 years, while the only change in hospitalization rates for SAH was a decrease among men and non-Hispanic individuals aged 45 to 54 years. Third, the prevalence rates of multiple risk factors associated with people experiencing AIS and ICH hospitalizations have continued to increase for both men and women of all age groups studied over the most recent decade of available data, and they are consistent with increases seen from 1995-1996 and 2007-2008. Finally, the prevalence of having multiple traditional stroke risk factors among young adults experiencing an AIS has doubled over the decade from 2003-2004 through 2011-2012.
The increasing trends in AIS hospitalization rates among the 3 younger age groups (18-34 years: 27.3% increase; 35-44 years: 35.6% increase; and 45-54 years: 20.5% increase) are consistent with other studies,3,9 as well as earlier hospitalization trends.4 They are also associated with high and increasing trends in the prevalence of having multiple traditional cardiovascular risk factors.9-12 This is also consistent with a clinical study reported 25 years ago, which found that among young adults aged 15 to 45 years, AIS was due to atherosclerotic disease in one-third of patients.13 Both small- and large-vessel disease have been identified as an increasing cause of AIS beginning at age 30 years.14 It is unclear why hospitalization rates among non-Hispanic black individuals aged 18 to 34 years remained stable; however, Kissela et al3 noted a nonsignificant increase in incident stroke rates among black individuals aged 20 to 44 years from 1993 to 2005 and yet a significant increase among white individuals. The young adult population experiencing AIS has rates of traditional stroke risk factors that are also nearly double that of their peers in the general population or greater.7 Approximately 12% of the general population aged 45 to 64 years has diabetes,7 yet approximately 40% of those hospitalized with an AIS had diabetes. Tobacco use is high among men of all age groups experiencing an AIS, at rates nearly double the national rate among adults. In comparison with European studies on stroke in young adults, we found similar rates of lipid disorders, atrial fibrillation, and tobacco use, but much higher rates of hypertension, diabetes, and ischemic heart disease.14 We found a lower prevalence of obesity than other traditional risk factors and yet obesity has been identified as a major risk factor for stroke in young adults.15-17 This could be due to underreporting in an administrative data set. The shift to an increasing number of multiple traditional risk factors for stroke and the extremely high and increasing prevalence of hypertension among those hospitalized with all stroke types is of particular concern, as the hypertension control rate among men aged 18 to 39 years is only 27.9%.18
One in 3 acute stroke hospitalizations for all stroke, AIS, and ICH occurs among those aged 18 to 64 years, while 60% of SAH hospitalizations occur among adults aged 18 to 64 years, when adults are in their prime years for productive careers and busy family lives.19 Most age, race/ethnic, and sex groups had nonsignificant decreasing trends in hospitalization rates for SAH, with only modest, although significant, decreases among men aged 45 to 54 years. Hypertension, tobacco use, and alcohol use are among the most important risk factors for SAH.20 We saw significant increases in hypertension, tobacco use, and drug abuse only among men and women aged 55 to 64 years and among women aged 45 to 54 years, but no significant increase in alcohol abuse among those with SAH. With increasing prevalence of key risk factors for SAH, the trend toward stable or decreasing hospitalizations for SAH are difficult to explain. There are improved diagnostic modalities for other cerebral conditions that could be leading to increased incidental findings of aneurysms from imaging for other reasons. Management and prophylactic treatment prior to rupture when indicated could in part explain decreases in hospitalizations for SAH.21
Nearly 40% of ICH hospitalizations occur in those aged 18 to 64 years.19 Because intracerebral hemorrhage can result in significant morbidity and mortality, the relatively high prevalence of unhealthy lifestyle behaviors among men aged 18 to 34 years (drug abuse: 14.3%; tobacco use: 25.8%; and hypertension: 44.1% among men and 28.1% among women aged 18-34 years), suggests that many of these strokes are likely preventable, although we are not able to determine the cause of stroke in this study. The findings that Hispanic and non-Hispanic black individuals have ICH hospitalization rates 1.5 to 2 times the rates for white individuals is not inconsistent with the findings from Walsh et al,22 which found an increased risk for ICH among Hispanic and non-Hispanic black individuals with treated or untreated hypertension among younger age groups compared with white individuals. They concluded that untreated hypertension confers a greater ICH risk in black and Hispanic relative to white individuals.
These findings of increasing AIS hospitalization rates among those aged 18 to 54 years and no change among those aged 55 to 64 years reveals an inflection point in stroke hospitalization rates when compared with findings of decreasing AIS hospitalization rates among those aged 65 to 74 and 75 to 84 years, and a nonsignificant decline among those 85 years and older.23,24 We identified no change in the hospitalization rate for ICH among those aged 18 to 64 years, while others have seen decreasing rates among those with ICH aged 65 to 74, 75 to 84, and 85 years and older.23 It is possible that these findings represent better hypertension control over time among elderly people compared with younger adults. We only noted a decrease in SAH hospitalization rates among those aged 45 to 54 years, while others also noted no change in rates among the 3 oldest age groups.23
This study has several strengths in that the data are nationally representative of hospitalizations in the United States and provide the ability to report trends over time. The reported comorbidities and risk factors were documented in the medical record using ICD-9-CM codes. However, there are some limitations. First, a person who experiences more than 1 hospitalization for an acute stroke during a 2-year analytic period would be represented for each acute stroke hospitalization event. Second, while one can identify the prevalence of multiple comorbidities, we are unable to characterize stroke severity nor the cause of the stroke. Third, while we cannot rule out any influence of increased use of advanced imaging over time, Kissela et al3 found high rates of the use of computed tomography (87%) and magnetic resonance imaging (60%-70%) among younger adults by 2005, and the impact of magnetic resonance imaging findings does not appear to influence stroke incidence,5 making it unlikely that the trends in stroke hospitalizations among younger adults are due solely to increased imaging and hence increased detection. Finally, we are unable to account for several important risk factors such as family history or the use of estrogen-based medications.
The young adult population is experiencing an increase in a serious yet largely preventable disease. We note the trend of increasing AIS among Hispanic individuals (a small but relative increase of 41% among those aged 18-34 years) of all age groups as a particularly important demographic to assure access to adequate stroke prevention. The significant increases in ischemic stroke hospitalizations and associated traditional stroke risk factors from 2003-2012, which is consistent with rates from 1995-1996 and 2007-2008 among those experiencing an acute stroke4 should serve as a call to action to focus on improving the health of younger adults. Preventing and controlling stroke risk factors among young working-age adults can save lives, reduce disability, decrease societal health care costs, and improve the quality of life for hundreds of thousands of Americans and their families. Identifying the high and rising prevalence of stroke risk factors among younger adults presenting with acute stroke should prompt a sense of urgency among younger adults, public health practitioners, clinicians, and policy makers to engage adolescents and their families, as well as younger adults, to identify and treat stroke risk factors and promote opportunities that allow for healthy lifestyles to prevent the tragedy of stroke at such early ages.
Corresponding Author: Mary G. George, MD, MSPH, Division for Heart Disease and Stroke Prevention, Centers for Disease Control and Prevention, 4770 Buford Hwy, MS-F-77, Atlanta, GA 30341 (firstname.lastname@example.org).
Accepted for Publication: January 14, 2017.
Published Online: April 10, 2017. doi:10.1001/jamaneurol.2017.0020
Author Contributions: Dr Tong had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: George, Tong.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: George, Tong.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Tong.
Administrative, technical, or material support: Bowman.
Conflict of Interest Disclosures: None reported.
Funding/Support: The US Centers for Disease Control and Prevention supported this study as part of the official duties of the authors. No external funding was used.
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the US Centers for Disease Control and Prevention.