Search tips
Search criteria 


Logo of neurologyNeurologyAmerican Academy of Neurology
Neurology. 2008 September 2; 71(10): 731–735.
PMCID: PMC2557022

Gender and ethnic differences in subarachnoid hemorrhage

S V. Eden, MD, W J. Meurer, MD, B N. Sánchez, PhD, L D. Lisabeth, PhD, MPH, M A. Smith, DrPH, MPH, D L. Brown, MD, MS, and L B. Morgenstern, MD



Mexican Americans (MAs) comprise the largest component of the largest minority group within the United States. The purpose of this study was to examine ethnic and gender differences in the epidemiology, presentation, and outcomes after subarachnoid hemorrhage (SAH) in a representative United States community. Targeted public health interventions are dependent on accurate assessments of groups at highest disease risk.


All patients with nontraumatic SAH older than 44 years were prospectively identified from January 1, 2000, to December 31, 2006, as part of the Brain Attack Surveillance In Corpus Christi project, an urban population-based study in southeast Texas. Risk ratios for cumulative SAH incidence comparing MAs with non Hispanic whites (NHWs) and women with men were calculated. Descriptive statistics for other clinical and demographic variables were computed overall, by gender, and by ethnicity.


A total of 107 patients had a SAH during the time period (7-year cumulative incidence: 11/10,000); of these, 43 were NHW (40% of cases vs 53% of the population) and 64 were MA (60% of cases vs 48% of the population). The overall age-adjusted risk ratio for SAH in MAs compared with NHWs was 1.67 (95% CI: 1.13, 2.47), and in women compared to men was 1.74 (95% CI 1.16, 2.62). Overall in-hospital mortality was 32.2%. No ethnic difference was observed for discharge disability or in-hospital mortality.


Subarachnoid hemorrhage disproportionately affects Mexican Americans and women. Public health interventions should target these groups to reduce the impact of this severe disease.


= Brain Attack Surveillance in Corpus Christi;
= Glasgow Coma Scale;
= International Classification of Diseases;
= intracerebral hemorrhage;
= Mexican American;
= modified Rankin Scale;
= non-Hispanic white;
= subarachnoid hemorrhage.

Subarachnoid hemorrhage (SAH) has received little attention relative to other stroke types with respect to racial and ethnic differences. In a study of predominantly Caribbean Hispanics in Northern Manhattan there was a trend suggesting higher SAH rates compared with non Hispanic whites (NHWs) but just 43 cases were observed.1 Similarly, Hispanics in New Mexico were found to have higher SAH rates compared with NHWs in a study that involved 47 cases.2

SAH, secondary to ruptured intracranial aneurysm, represents a devastating event that is associated with a high rate of morbidity and mortality.3 We sought to examine and describe ethnic differences in the epidemiology and outcomes of SAH in a representative bi-ethnic United States community devoid of an academic referral center. Given that Mexican Americans (MAs) now comprise the largest component of the largest and fastest-growing minority group within the United States, it is prudent to examine the burden of SAH in MAs. There is a suggestion in some studies that SAH disproportionately affects women compared with men.4 A secondary goal of this population-based work was to compare SAH by gender.



The Brain Attack Surveillance in Corpus Christi (BASIC) Project is a population-based stroke surveillance study in Nueces County, a bi-ethnic community in southeast Texas. Nueces County is predominantly an urban population with 90% of county residents residing within the city of Corpus Christi. According to the United States Census,5 the Nueces County population was 313,645 in 2000 and 321,457 in 2006, an increase of just 2.49% in 6 years. Nueces County is not an immigrant community, there is little influx or efflux of residents, and we have previously shown that the MAs are predominantly second and third generation United States citizens and have resided in Nueces County longer than the NHW population.6 In a population-based random digit dialed telephone survey performed in this community, no difference was found between MAs and NHWs with respect to hypertension, current tobacco use, or elevated lipid levels. Diabetes was more common in MAs (15%) compared with NHWs (9%).6

Surveillance methodology.

An academic medical center is not present in this community. The closest referral centers are in Houston and San Antonio (approximately 150 miles away). This geographic isolation allows for complete case capture of initial cerebrovascular events at the seven hospitals in the county. The methods employed in the BASIC project have previously been described in detail.7 Briefly, active and passive surveillance were used to identify all cases of nontraumatic SAH from January 1, 2000, to December 31, 2006, in patients older than 44 years. Cases were identified from emergency department and inpatient documentation by manually searching emergency department and inpatient logs for a set of previously validated screening diagnostic criteria that included a comprehensive list of focal and generalized neurologic symptoms of SAH.8 This active surveillance was supplemented by a review of hospital discharge records, searching for ICD-9 (International Classification of Diseases, ninth revision)9 codes for stroke, as previously described.7 The county coroner was contacted to determine if there were cases of sudden death due to SAH that never made it to the hospital. Several years ago, we reported preliminary SAH findings in this population from 2000 to 2002 as part of a larger study of all stroke types.10 Planned additional time has allowed sufficient case capture to more fully examine SAH in MAs compared with NHWs and these data are presented here.

Demographics (including race-ethnicity), length of stay, time to presentation, and risk factors were obtained from the medical records by trained abstractors. Among the Hispanics in Nueces County included in the 2000 US Census, only 1.2% identified an ancestral origin from a country other than Mexico (1% identified Spain and 0.2% identified Central or South America). We have previously demonstrated high agreement between ethnicity recorded in the medical record and self-reported ethnicity in this population (kappa = 0.94).7 Further, agreement between the chart and self-reported risk factors was high for hypertension as well (kappa = 0.62). We prospectively decided to include risk factors that previously were shown to affect SAH incidence.11 Risk factors included documented history of hypertension, current smoking, excessive alcohol consumption (≥4 alcoholic drinks per day), and any health insurance. Time from symptom onset to presentation was calculated from the medical record. If specific onset time was not recorded, a previously published, standardized estimation method was used.12 If no reference to time of onset was listed in the medical record, time was recorded as unknown. Discharge modified Rankin (mRS) and Glasgow Coma Scales (GCS) were retrospectively calculated from the medical record by the abstractors.

A board certified neurologist validated all cases of SAH using source documentation including emergency department and hospital admission records, blinded to ethnicity and age. SAH in this study was defined as neuroimaging evidence of the presence of subarachnoid blood. Additionally, cases with CSF results that had a red blood cell count of >1,000 (without a 25% decrease in red blood cells from first or last tubes) or evidence of xanthochromia were validated as SAH.13 Cases that had both intracerebral hemorrhage (ICH) and SAH are classified as SAH if an aneurysmal source of bleeding was documented on radiologic reports or the report suggested a subarachnoid origin of the bleeding. Cases were classified as ICH if a parenchymal source of bleeding was thought more likely. SAH occurring in the setting of a coagulopathy without the identification of an arteriovenous malformation or aneurysm were excluded.

The Institutional Review Boards of the local hospitals and the University of Michigan provided approval for the BASIC project. All individual patient identifiers were removed from the source documents.

Statistical methods.

This was a primary, prospective, prespecified analysis of BASIC. Patients’ demographic and clinical characteristics were expressed as mean and standard deviations for continuous variables, as medians and interquartile ranges for skewed and ordinal variables, and as frequencies for nominal variables.

Cumulative incidence of SAH over the 7-year study period was calculated by race/ethnicity, by gender, and by age, categorized into groups (45–59, 60–74, and ≥75). Denominators for the incidence calculations were derived from the 2000 US Census.5 Population data for 2000 were chosen to represent the population at risk for stroke as our case ascertainment for stroke began in this year. Risk ratios and 95% CIs were calculated to compare MAs and NWHs and women and men overall and within age groups. The Breslow Day test of homogeneity was used to test for ethnic and gender differences in the SAH risk ratios by age.

A Poisson regression model that simultaneously adjusted for age, ethnicity, and gender was used to evaluate the independent contribution of these factors on the incidence of SAH. The outcome was the number of cases in strata defined by gender, ethnicity, and age category (45–90, 60–74, and ≥75), and the offset was the population in the respective strata. We tested for gender by ethnicity interaction after adjustment for age. Because adjustment by age in only three categories may render the analysis susceptible to residual confounding, adjustment for age in 5-year intervals was also considered.


From January 1, 2000, to December 31, 2006, cases screened by abstractors totaled 29,907. The screened population was 38% NHW, 55% MA, 6% African American, and 1% other race/ethnic groups. The median age was 70 and 60% were women. Of these cases, 6,550 met criteria for cerebrovascular events (TIA, ischemic stroke, ICH, and SAH) and were referred to a neurologist for validation. There were a total of 5,540 validated cerebrovascular events during the study time period. A total of 117 cases were validated as SAH by study neurologists. Ten cases were excluded due to small numbers in the race-ethnic group (8 African Americans and 2 Asian Americans) leaving 107 SAH cases for analysis. Each case represents one individual. No one had more than one SAH during the study course.

Of the 107 patients included in the analysis, 43 (40.0%) were NHW and 64 (60.0%) were MA (table 1), although 52% of the population 45 years and older is NHW, and 48% are MA. Moreover, 67.3% of the patients with SAH in this study were women, although only 53.5% of the population are women. There were no significant racial/ethnic differences in prevalence of hypertension, excessive alcohol use, tobacco use, hospital length of stay, in-hospital mortality, initial and discharge GCS, and discharge mRS. Similarly, there were no gender differences in any of the variables listed in table 1 except that men had greater excessive alcohol use (p = 0.03).

Table thumbnail
Table 1 Risk factors, demographics, and clinical characteristics, 2000–2006, Nueces County, Texas

The 7-year cumulative incidence for SAH was higher in MAs compared with NHWs (table 2). The overall age-adjusted risk ratio for SAH in MAs compared with NHWs was 1.67 (95% CI: 1.13, 2.47). We also found gender differences in SAH incidence (table 2). The Breslow-Day test for homogeneity of gender differences across age was p = 0.08. Given the marginal results, and since this test is underpowered, both age-adjusted gender risk ratio and stratified age/gender risk ratios were computed. The overall age-adjusted risk ratio for SAH in women compared to men was 1.74 (95% CI: 1.16, 2.62); age-specific gender risk ratios are given in table 3.

Table thumbnail
Table 2 Seven-year cumulative incidence and cumulative incidence risk ratios by age, ethnicity, and gender, 2000–2006, Nueces County, Texas
Table thumbnail
Table 3 Age- and sex-specific 7-year cumulative incidence and risk ratios, 2000–2006, Nueces County, Texas

The Poisson regression model that simultaneously adjusted for age, ethnicity, and gender confirmed that gender and ethnicity were independent risk factors for SAH incidence, and showed a marginal increasing incidence of SAH with increasing age (table 4, p value for trend = 0.08). Although MA women have the highest incidence, a formal test for the interaction of gender and ethnicity for SAH incidence was not conclusive (p = 0.16). Age adjustment in finer age categories did not change the estimated ethnicity and gender risk ratios.

Table thumbnail
Table 4 Poisson regression model, adjusted for age, ethnicity, and gender, of 7-year cumulative incidence, 2000–2006, Nueces County, Texas


This study found that SAH is 67% more common in MAs compared to NHWs adjusted for age. Prior population-based studies have also documented racial/ethnic differences in SAH incidence. In Northern Manhattan, the age- and sex-adjusted incidence of SAH was 30% greater for Caribbean Hispanics compared with NHWs, although a significant difference was not observed perhaps due to the small number of cases observed.1 Another population-based study that used passive surveillance in Bernallilo County, NM, documented a two and a half fold higher incidence of spontaneous SAH in Hispanic Americans compared to NHWs.2

The reasons for the higher risk of SAH in MAs compared with NHWs are likely multifactorial. Genetic heterogeneities resulting in a higher prevalence of intracranial aneurysms or other vascular malformations and a greater tendency to bleed from these vascular structures among MAs have been reported.2 Racial/ethnic differences in risk factors including hypertension, heavy alcohol consumption, and cigarette smoking have also been suggested as possible explanations for racial/ethnic differences in SAH incidence.2 Although differences in SAH risk factors, including hypertension and alcohol and tobacco use, did not differ among the cases in the current study, these factors may still explain some of the observed race-ethnic differences in SAH. Still, previously reported survey data from the community studied in the current work did not demonstrate differences in cigarette smoking and hypertension among MAs and NHWs.6 Nevertheless, differences in the amount and duration of tobacco use and underdiagnosis or undertreatment of hypertension may play a role in ethnic differences in SAH incidence. Prior studies have demonstrated lower treatment rates of hypertension in Hispanics compared with NHWs as well as more poorly controlled hypertension despite medical treatment in racial/ethnic minorities.14,15 Further studies are warranted to examine these factors as well as other factors such as socioeconomic status and access to care that may mediate racial/ethnic differences in SAH risk.

We found no ethnic difference in outcome at discharge as measured by the modified Rankin score. Because of the relatively small sample size we had limited power to detect such a difference.

Another interesting finding was the gender difference in the incidence of SAH. Women comprised 67.3% of the patients affected by SAH in this study. MA women were disproportionately affected by SAH (table 2). Prior studies have documented higher rates of SAH in women compared to men.4 Differences in hypertension, smoking, and alcohol use as well as the involvement of estrogen have been suggested as possible explanations of the gender disparity in SAH.16 Further studies are warranted to examine the gender disparities and potential gender/ethnicity interaction in SAH.

An advantage of this study is the rigorous methods used for SAH case identification. We capture events through a combination of active and passive surveillance using rigorous validation and screening mechanisms in this community. This study has several important limitations. We were limited by the overall number of cases, and were therefore unable to comment further on racial/ethnic differences in aneurysm size, location, or treatments. Only those >44 were included in the study. Prior studies have found higher SAH incidence in young men compared with young women.17 If younger ages were included it is possible that an interaction between gender and age would be found, making age adjustment inappropriate and leading to different gender-specific ratios at different ages. However, SAH incidence increases with age and is much less common before age 45, with half the incidence in those 35–45 compared to those 45–55.17

Given our study design, we were unable to explore potential reasons for the observed ethnic differences aside from demographic characteristics. In the current study, it would have been useful to directly compare risk factor profiles to subjects in the community who did not have a SAH. These comparisons are the necessary next step in identifying modifiable risk factors to lessen the SAH burden in women and MAs. This study is also limited to one geographic area, which is unique in its demographic composition, so caution is advised when generalizing to other populations.

This study provides the epidemiology, risk factors, and outcome of subarachnoid hemorrhage in a bi-ethnic community devoid of a tertiary referral center in the United States. We found a substantially elevated SAH risk in MAs compared with NHWs as well as a substantial gender disparity in SAH incidence. Further studies are indicated to fully appreciate the etiology of these differences. Analysis of larger populations may be helpful as the current study also suggests a trend that women may contribute disproportionately to the increased burden of SAH in MAs. Public health planners may consider women and MAs high-risk groups for risk-factor reduction efforts.


Address correspondence and reprint requests to Dr. Lewis B. Morgenstern, Stroke Program, University of Michigan, Cardiovascular Center Room 3194, SPC 5855, Ann Arbor, MI 48109-5855 ude.hcimu@snegroML

e-Pub ahead of print on June 11, 2008, at

Funded by NIH R01 NS38916.

Disclosure: The authors report no disclosures.

Received December 15, 2007. Accepted in final form May 1, 2008.


1. Labovitz DL Halim AX Brent B Boden-Albala B Hauser WA Sacco RL Subarachnoid hemorrhage incidence among Whites, Blacks and Caribbean Hispanics: the Northern Manhattan Study. Neuroepidemiology. 2006;26:147–150. [PubMed]
2. Bruno A Carter S Qualls C Nolte KB Incidence of spontaneous subarachnoid hemorrhage among Hispanics and non-Hispanic whites in New Mexico. Ethn Dis. 1997;7:27–33. [PubMed]
3. Feigin VL Findlay M Advances in subarachnoid hemorrhage. Stroke. 2006;37:305–308. [PubMed]
4. Kongable GL Lanzino G Germanson TP et al. Gender-related differences in aneurysmal subarachnoid hemorrhage. J Neurosurg. 1996;84:43–48. [PubMed]
5. United States Census 2000 [online]. Available at: Accessed February 2008.
6. Morgenstern LB Steffen-Batey L Smith MA Moye LA Barriers to acute stroke therapy and stroke prevention in Mexican Americans. Stroke. 2001;32:1360–1364. [PubMed]
7. Smith MA Risser JM Moye LA et al. Designing multi-ethnic stroke studies: the Brain Attack Surveillance in Corpus Christi (BASIC) project. Ethn Dis. 2004;14:520–526. [PubMed]
8. Morgenstern LB Wein TH Smith MA Moye LA Pandey DK Labarthe DR Comparison of stroke hospitalization rates among Mexican-Americans and non-Hispanic whites. Neurology. 2000;54:2000–2002. [PubMed]
9. ICD-9. International Classification of Disease, 9th Revision, Clinical Modification. Los Angeles: Practice Management Information Corporation, 1995.
10. Morgenstern LB Smith MA Lisabeth LD et al. Excess stroke in Mexican Americans compared with non-Hispanic Whites: the Brain Attack Surveillance in Corpus Christi Project. Am J Epidemiol. 2004;160:376–383. [PMC free article] [PubMed]
11. Feigin VL Rinkel GJ Lawes CM et al. Risk factors for subarachnoid hemorrhage: an updated systematic review of epidemiological studies. Stroke. 2005;36:2773–2780. [PubMed]
12. Menon SC Pandey DK Morgenstern LB Critical factors determining access to acute stroke care. Neurology. 1998;51:427–432. [PubMed]
13. Morgenstern LB Luna-Gonzales H Huber JCJr et al. Worst headache and subarachnoid hemorrhage: prospective, modern computed tomography and spinal fluid analysis. Ann Emerg Med. 1998;32:297–304. [PubMed]
14. Sudano JJJr Baker DW Antihypertensive medication use in Hispanic adults: a comparison with black adults and white adults. Med Care. 2001;39:575–587. [PubMed]
15. Cooper-DeHoff RM Handberg EM Cohen J et al. Characteristics of contemporary patients with hypertension and coronary artery disease. Clin Cardiol. 2004;27:571–576. [PubMed]
16. Horiuchi T, Tanaka Y, Hongo K. Sex-related differences in patients treated surgically for aneurysmal subarachnoid hemorrhage. Neurol Med Chir (Tokyo) 2006;46:328–332; discussion 332. [PubMed]
17. de Rooij NK Linn FH van der Plas JA Algra A Rinkel GJ Incidence of subarachnoid haemorrhage: a systematic review with emphasis on region, age, gender and time trends. J Neurol Neurosurg Psychiatry. 2007;78:1365–1372. [PMC free article] [PubMed]

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