Original Contribution

Aspirin administration in ED patients who presented with undifferentiated chest pain: age, race, and sex effects^?

Kevin M. Takakuwa MD ^a,?, Frances S. Shofer PhD ^b, Judd E. Hollander MD ^b

^aDepartment of Emergency Medicine, Thomas Jefferson University Hospital, Philadelphia, PA 19107-5004, USA

^bDepartment of Emergency Medicine, University of Pennsylvania Health System, Philadelphia, PA 19104, USA

Received 30 January 2008; revised 20 December 2008; accepted 20 December 2008

Abstract

Study objectives: The study aimed to determine whether Aspirin therapy was differentially administered according to race, sex, or age in patients with undifferentiated chest pain who presented to an urban academic emergency department.

Methods: This was a prospective observational cohort study of patients older than 24 years who presented with chest pain between July 1999 and March 2002. Patients were grouped according to 30-day final diagnosis: acute myocardial infarction AMI, unstable angina USA, and non-acute coronary syndrome (ACS) chest pain. Data were analyzed using Fisher exact test and relative risk regression using the Gaussian estimating equation.

Results: There were 4478 patient visits, of which 4470 (99.8%) had complete information. Mean age was

52.2 +- 15.8 years. Blacks were 70.1% (n = 3135), whites 26.3% (n = 1175), and other 3.6% (n = 159). Women comprised 59.0% (n = 2639) of the patients. Aspirin therapy differed by race, sex, age, and final diagnosis. Patients who received aspirin were more likely to be white (60% vs 54%, P = .0009) or have an ACS diagnosis (82% vs 50%, P b .0001). By final diagnosis, there were no race, sex, or age differences for AMI or USA (P N .05). There were significant sex and age differences for non-ACS chest pain patients: men (53% vs 48% women, P = .0009) and older patients (N55 years, 60% vs 44% younger, P b .0001) had higher aspirin therapy due to administration to the patients with non-ACS chest pain. Conclusion: For patients with undifferentiated chest pain, overall race, sex, and age differences were explained by higher rates of aspirin administered to older men with non-ACS chest pain.

(C) 2010

^? An abstract of this article was presented at the ACEP Annual Meeting in San Francisco on October 18, 2004, and was published in abstract form as: Takakuwa KM, Shofer FS, Hollander JE. Aspirin therapy in patients who presented to an academic emergency department with chest pain: age, race and gender effects [abstract]. Ann Emerg Med 2004;44: S101.

* Corresponding author.

E-mail address: [email protected] (K.M. Takakuwa).

Introduction

Background

Race [1-5] and sex [1-3,5-6] disparities in cardiac care are known to exist starting in the emergency department (ED) and continuing through to in-hospital care. Aspirin is known to protect against myocardial infarction and death [7] and is efficacious in suspected myocardial infarction [8]. The

0735-6757/$ - see front matter (C) 2010 doi:10.1016/j.ajem.2008.12.021

American College of Cardiology and the American Heart Association Guidelines recommend that aspirin should be administered “in the ED as soon as the diagnosis of ACS is made or suspected” [9,10]. According to the Centers for Medicare and Medicaid Services, aspirin should be adminis- tered within 24 hours before or after hospital arrival for patients with acute myocardial infarction [11].

Some national database studies have reported differences in aspirin administration based on race and sex [5,12], but these were studies of patients based on final diagnosis rather than on an undifferentiated chest pain population. In practice, there should be no race, sex, or age differences in aspirin administration for patients with known or suspected ACS. However, in an undifferentiated chest pain population in which a diagnosis of ACS may not be clear during the ED stay, there may be discretionary decisions made on who gets aspirin based on clinical suspicion. We focused on aspirin and not on other therapies such as ?-blockers and antithrombins because aspirin is inexpensive, has a proven benefit, and is a Centers for Medicare and Medicaid Services measure for AMI.

Importance

We are unaware of any ED-based study to date that has quantified aspirin therapy in an undifferentiated chest pain population according to race, sex, and age. Determining whether such disparities in administering this low-cost proven therapy in patients with USA and AMI (see definition in section 2.9 below) is important in evaluating performance of patient care. Uncovering why differences may exist is the next step before enacting strategies in an attempt to mitigate any differences that might be found.

Goals of this investigation

We wanted to determine if there were race, sex, or age differences in aspirin administration for those with time- sensitive ACS.

Methods

Study design

This is a prospective observational cohort study of consecutively enrolled patients who presented to our ED with chest pain. We tested the null hypothesis that race, sex, and age would not affect aspirin administration. The study was approved by the University Committee on Research Involving Human Subjects.

Setting

This study was conducted at the Hospital of the University of Pennsylvania, an urban tertiary care hospital

with an ED that has an annual census of approximately 49 000 adult visits. About 2000 of the yearly ED census are patients with chest pain.

Participants

Patients 24 years or older who presented to the ED between July 6, 1999, and March 25, 2002, with a complaint of chest pain prompting an electrocardiogram (ECG) were included. Chest pain did not have to be the chief complaint. Patients younger than 24 years were included if they reported cocaine use. Patients were excluded only if they were younger than the stated ages, did not have chest pain, or did not have an ECG obtained. We only included patients with chest pain so we could more clearly define our study population and allow generalizability of the data. We did not specify any Clinical impression of cardiac or noncardiac chest pain and included a broad range of ages to enhance generalizablity.

Study protocol

During the study period, trained research assistants were present in the ED 16 hours per day from 8:00 AM to midnight, 7 days per week to identify and enroll patients. Patients were treated in the ED by a full-time board- certified or board-eligible ED physician working alone or in conjunction with housestaff.

Measurements

Patients underwent a structured history and physical examination either at or sometime after ED room or hallway placement. Patient information collected prospectively in the ED included demographic information, vital signs, cardiac risk factors, characteristics of chest pain, associated symptoms, medications, treatment, and disposition. Admitted patients were followed daily throughout their hospital course. Follow-up was obtained 30 days after presentation by calling patients or their proxies. We assumed that if patients or their proxies reported no further problems, then they did not have ACS.

Race and sex information

Race and sex were recorded by trained research assistants and classified based on appearance. When it was not obvious, patients were questioned. This protocol was designed to mimic the way clinicians assess this information in patients.

Aspirin administration

Aspirin was considered given if it had been administered in the ED. Because the American College of Cardiology/ American Heart Association guidelines recommend giving a

range of aspirin (162-325 mg) and some individuals may be prescribed a lower dose because of problems such as hypercoagulability, aspirin was considered either as given or not regardless of dose.

Cardiac biomarker assays

For admitted patients, venous blood samples were collected on presentation to the ED in phlebotomy tubes containing no anticoagulant or preservative. Cardiac tropo- nin I, creatine kinase, and creatine kinase-MB (CK-MB) assay were routinely performed.

Definition of acute coronary syndrome

Acute coronary syndrome (ACS) was defined as a diagnosis of acute myocardial infarction or unstable angina in accordance with the American College of Cardiology/ European Society of Cardiology criteria [13]. A diagnosis of AMI was made if the patient had an elevation of cardiac troponin I >=2 ng/mL or CK-MB enzyme >=10 ng/mL. USA was considered to occur if there was documented reversible ischemia on stress test, coronary artery occlusion

>=70% in at least one vessel as seen during cardiac catheterization, or elevations of cardiac enzymes above laboratory normal but below levels necessary for diagnosis of AMI (troponin I >=0.4 ng/mL but b2 ng/mL; CK-MB

>=5 ng/mL but b10 ng/mL) in accordance with the standardized guidelines [14].

Patients who did not meet the definitions of AMI or USA were classified as having noncardiac chest pain. Included in this group were some patients who received an ECG but not cardiac markers. A comparison of patients without markers

Table 1 Chest pain characteristics and medical history by race

Nonwhites n

Whites

n

P

%

%

to those with markers that were negative showed that the patients without markers were a lower-risk group. We therefore felt it was reasonable to group these patients together under noncardiac chest pain.

Data analysis

For analysis purposes, race was divided into 2 groups: nonwhite and whites. Age was divided into 2 groups: younger than 55 years, and 55 years and older. To determine differences in aspirin treatment between men and women, whites and nonwhite, and age group, Fisher exact test was used. To assess the independent contribution of each of the above grouping variables while simultaneously adjusting for confounders (cardiac risk factors and prior history of Cardiac conditions), we performed Relative risk regression using the Gaussian estimating equation [15,16]. Data for these analyses are presented as Relative risks with 95% confidence intervals. All analyses were performed using SAS statistical software (Version 9.1, SAS Institute, Cary, NC). A P value of less than .05 was considered statistically significant.

Results

Of 4478 total patient visits, 4470 (99.8%) had complete information and were included in this study. The mean age was 52.2 +- 15.8. Of these, blacks comprised 70.1% (n = 3,135), whites were 26.3% (n = 1175), and 3.6% (n = 159) were other ethnicities. Women comprised 59.0% (n = 2,639) and men 41.0% (n = 1831) of the study. A total of 2498 patients (56%) received aspirin therapy. Three hundred nineteen patients (7%) had an AMI, and 544 patients (12%)

Chest pain characteristics
Chest pain chief complaint	1943	60	696	61	.53
Pain location left chest	1317	41	435	38	.16
Pain location mid chest	1237	38	464	11	.13
Quality: pressure ?	1345	42	519	46	.02
Radiates to neck/left arm	912	28	347	61	.15
Medical history
Hypertension ?	1685	52	444	39	b.0001
Diabetes ?	608	19	168	15	.002
Tobacco ?	1240	38	398	35	.05
Family history ?	622	19	254	22	.03
Increased cholesterol ?	481	15	286	25	b.0001
Prior AMI	354	11	134	12	.38
Known coronary artery disease ?	552	17	304	27	b.0001
Known angina ?	385	12	172	15	.006
Congestive heart failure ?	342	11	96	8	.04
* P <= .05, significant difference.

had USA. Most, 3607 (81%), had a non-ACS chest pain diagnosis. Chest pain characteristics, medical history, and cardiac history by race are shown in Table 1 and by sex in Table 2.

Race

Overall, whites were significantly more likely to receive aspirin (60%) compared to nonwhites (54%, P b

.005). However, this difference was not seen based on final diagnosis category. There were no differences between whites and nonwhites with a final diagnosis of AMI or USA. Eighty-eight percent of whites received aspirin compared to 83% of nonwhites with a final diagnosis of AMI (P = .27). For USA, 77% of whites received aspirin vs 81% of nonwhites (P = .37). For patients with a final diagnosis of noncardiac chest pain, whites did not receive aspirin more than nonwhites (53% compared with 49%, P = .06, Fig. 1).

Sex

Men were significantly more likely to receive aspirin (61.2%) than women (52.2%, P b .0001). This difference depended on final diagnosis category. For time-sensitive ACS, there were no significant time differences in aspirin administration between women and men in the AMI or USA groups. Of women, 86.4% received aspirin compared with 83.5% of men with a final diagnosis of AMI (P = .53) and 77.5% compared with 81.7%, respectively, with a final diagnosis of USA (P = .24). There was a significant sex difference for the patients with noncardiac chest pain. Of the

Fig. 1 Aspirin use for AMI, USA, and non-ACS chest pain by sex, age, and race.

women, 47.6% received aspirin compared with 53.4% of men (P b .001, Fig. 1).

Table 2 Chest pain characteristics and medical history by sex

Women Men P

N % N %

Age

Chest pain characteristics
Chest pain chief complaint ?	1502	59	1137	63	.007
Pain location left chest ?	958	38	794	44	b.0001
Pain location mid chest ?	1055	41	646	36	b.0001
Quality: pressure	1099	43	765	42	.69
Radiates to neck/left arm	720	28	539	30	.25
Medical history
Hypertension	1278	50	851	47	.06
Diabetes	469	18	307	17	.24
Tobacco ?	843	33	795	44	b.0001
Family history	511	20	365	20	.88
Increased cholesterol ?	414	16	353	20	.005
Prior AMI ?	228	9	260	15	b.0001
Known coronary artery disease ?	433	17	423	23	b.0001
Known angina ?	294	12	263	15	.003
Congestive heart failure	249	10	189	10	.44
* P <= .05, significant difference.

There were overall age effects in which older patients

received aspirin more than younger patients (P b .0001). These effects differed by final diagnosis. For AMI and USA,

there were no age differences. However, for noncardiac chest pain, patients aged 55 years and older were significantly more likely to receive aspirin (60.3%) compared with those aged 30 to 54 years old (48.6%) or younger than 30 years

(12.9%, P b .0001, Fig. 1).

Multivariate analysis

After stratifying on final diagnosis (ACS compared to non-ACS), adjusted risk ratios indicated that men (RR = 1.10) and patients 55 years or older (RR = 1.13) were more likely to receive aspirin therapy in the non-ACS group (Fig. 2). No differences were found for these factors

Fig. 2 Relative risks comparing ACS with non-ACS diagnosis.

in the ACS group. In addition, these factors were unaffected by other classic risk factors as listed on Tables 1 and 2 (hypertension, diabetes, smoking history, family history, hypercholesterolemia, prior AMI, known coronary artery disease, known angina, or congestive heart failure).

Discussion

Race

Our results show that as a group, whites received aspirin more than nonwhites but not in the AMI or USA groups, which is considered a time-sensitive diagnosis because there are proven therapies available that could help the patient during these acute events. The difference is because of the tendency of whites to get aspirin administered more frequently in the noncardiac chest pain group. It is reassuring to see no race differences in the groups that

had time-dependent heart disease in which aspirin is known to be of benefit.

However, it is suboptimal to see that there were overall race differences with whites getting aspirin more than nonwhites. There are several possible explanations for this finding. One is that the white patients may have already been taking aspirin at higher rates so they appear to be given aspirin more frequently. Aspirin was considered as having been administered if it was taken within the past 24 hours. It is possible that in lower-risk patients, the treating physicians did not generally administer aspirin as a therapy but, because they were already taking it routinely, it appeared as if it had been prescribed in the ED more frequently.

Another explanation is related to pretest probability. It is possible that the treating physicians had noted from their experience that the white patients had more ACS than nonwhite patients. In our ED, 25.8% of white patients with chest pain had ACS in comparison to 17.0% of nonwhite patients so the physicians may have administered aspirin more frequently to white patients. This phenomenon might be related to the concept of “overtreatment” in which certain patients without disease are treated more aggressively for a presumed disease entity.

Also related to pretest probability are factors such as medical history. The white group had higher hypercholes- terolemia and a higher rate of family history of coronary artery disease than the nonwhite group, which could have swayed giving aspirin. However, the nonwhite group had more diabetes, hypertension, and smoking history, which should have been a counterbalancing force, especially because diabetes is well known to increase the risk of coronary artery disease [17-19].

Finally, another possible explanation for the race differences could be because of the chest pain characteristics. White patients were more likely to describe “pressure” than nonwhites. Pressure is considered “probable high risk” for ACS [20]. It is also possible that blacks, who comprised 95.2% of nonwhite patients in this study, might present with cardiac symptoms that are more atypical [21-24]. Combining these factors could have influenced physician decision to give aspirin, especially in what would become the non-ACS chest pain group.

Our race results for ACS appear to be consistent with 2 national database studies. The CRUSADE database is a National Quality Improvement Initiative of more than

400 US hospitals that collects data on non-ST segment elevation (NSTE) NSTE ACS. From this database, Sonel et al [12] reported no significant differences between whites and blacks in aspirin given. Within 24 hours, 90.9% of whites received aspirin compared with 92.0% of blacks for patients with USA. The National Registry of Myocardial Infarction is a national study that has enrolled over a million patients with myocardial infarction since 1990 in different collection phases. Using this database, Vaccarino et al [5] reported that 84.4% of white men received aspirin compared with 78.7% of white women, 83.7% of black men, and

78.4% of black women. Although they did not report for whites compared to blacks, there do not appear to be significant racial differences in aspirin administered for their patients with AMI.

Sex

As a group, men received aspirin more than women but this did not occur in the time-sensitive AMI or unstable angina groups, in which there were no differences. Our results differ from the large database studies. The NRMI trial showed that women received aspirin less than men with AMIs [5]. For the NSTE ACS patients in CRUSADE, women both received less aspirin than men within 24 hours of treatment (89.6% compared with 91.6%, respectively) and were less likely to be prescribed aspirin at discharge from the hospital (87.5% compared with 90.4%, respectively) [25].

The main group difference occurred because women received aspirin less in the noncardiac chest pain group. We are again reassured that the group with cardiac disease received the appropriate medication, but we are uncertain why women received less aspirin in the noncardiac chest pain group. As for race, it is possible that, in general, women at baseline are not prescribed aspirin as much as men, which could account for the noncardiac chest pain group of women seeming to get aspirin less than men.

It is also possible that men are overtreated with aspirin compared to women, possibly due to patient or physician preference or a higher pretest probability of men having heart disease compared to women. Men are demographically at a higher risk for coronary artery disease and AMI than women [26]. In addition, women may have received aspirin less frequently because they could present with coronary artery disease more atypically than men [27], although this belief has been challenged [28].

Age

The age results were what would be expected. Overall, there was a bias to give older patients aspirin in comparison to younger patients. For patients with AMI and USA, there were no significant age differences. However, for noncardiac chest pain, older patients were treated more with aspirin than younger patients. Given the natural progression of coronary artery disease with increasing age, it would be expected for older patients to be treated more aggressively than younger patients, even in the noncardiac chest pain group.

Limitations

Our study has several limitations. It was conducted at a single urban academic institution. The results may be applicable to an inner-city population that has a predomi- nantly black population but are not generalizable to suburban

or rural populations or hospitals that treat a different race and sex mix. Our data cannot be generalized to patients with “anginal equivalents” in the absence of chest pain.

We only collected data on aspirin administered in the ED. Some patients may have been given aspirin en route by paramedics or used aspirin at home, which may have affected results. However, of the 24% of patients on home aspirin, 69% received aspirin in the ED. When we recalculated the relative risk including patients who had taken aspirin at home as part of the outcome of interest, the same variables were statistically significant. Thus, it does not appear that patients’ home aspirin usage affected the physicians’ decision on whether to give aspirin in the ED.

We also note that race and sex were recorded by trained research assistants and classified based on appearance to mimic the way clinicians assess this information in patients. It is possible that this information was incorrect. However, because the self-reported demographics of our patient population support the fact that most patients seen in our ED are either white or black, it is unlikely that race was often misclassified. Thus, our study is limited to the perception of race as opposed to actual race. Insurance information is not routinely available to clinicians during the initial evaluation and thus would be unlikely to be a confounding variable. Our definition of AMI was based on definitions during the time of the study. Patients who were discharged without further problems at 30 days were assumed not to have ACS. Finally, there was no research assistant during midnight to 8:00 AM.

Conclusions

In an undifferentiated chest pain population, we found that there were overall differences in aspirin administration by race, sex, and age. Nonwhites, women, and younger patients appear to receive less aspirin compared to whites, men, and older patients. However, based on final diagnosis, there were no race, sex, or age differences for patients who had time-sensitive ACS. The overall differences for sex and age can be explained by the overtreatment of male and older noncardiac chest pain patients receiving aspirin.

References

1. Stone PH, Thompson B, Anderson HV, et al. Influence of race, sex, and age on management of unstable angina and non-Q-wave myocardial infarction: the TIMI III registry. JAMA 1996;275:1104-12.
2. Weitzman S, Cooper L, Chambless L, et al. Gender, racial, and geographic differences in the performance of cardiac Diagnostic and therapeutic procedures for hospitalized acute myocardial infarction in four states. Am J Cardiol 1997;79:722-6.
3. Rathore SS, Berger AK, Weinfurt KP, et al. Race, sex, poverty, and the medical treatment of acute myocardial infarction in the elderly. Circulation 2000;102:642-8.
4. Venkat A, Hoekstra J, Lindsell C, et al. The impact of race on the acute management of chest pain. Acad Emerg Med 2003;10:1199-208.
5. Vaccarino V, Rathore SS, Wenger N, et al. Sex and racial differences in the management of acute myocardial infarction, 1994 through 2002. N Engl J Med 2005;353:671-82.
6. Barakat K, Wilkinson P, Suliman A, Ranjadayalan K, Timmis A. Acute myocardial infarction in women: contribution of treatment variables to adverse outcome. Am Heart J 2000;140:740-6.
7. Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of Antiplatelet therapy, I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. [erratum appears in BMJ 1994;308: 1540]. BMJ 1994;308:81-106.
8. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction. Lancet 1988;2:349-60.
9. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non ST-Elevation Myocardial Infarction). Circulation 2007;116:e148-e304.
10. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of Patients with Acute Myocardial Infarction). Circulation 2004;110: e82-e292.
11. Specifications manual for national Hospital quality measures, version

2.2. Centers for Medicare and Medicaid Services , Joint Commission on Accreditation of Healthcare organizations (JCAHO); 2007.

1. Sonel AF, Good CB, Mulgund J, et al. Racial variations in treatment and outcomes of black and white patients with high-risk non-ST- elevation acute coronary syndromes: insights from CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the ACC/AHA Guidelines?). Circulation 2005;111:1225-32.
2. Thygesen K, Alpert JS, Garson A, et al. Myocardial infarction redefined: a consensus document of the Joint European Society of Cardiology/American College of Cardiology committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36: 959-69.
3. Hollander JE, Blomkalns AL, Brogan GX, et al. Multidisciplinary Standardized reporting Criteria Task Force. Standardized reporting guidelines for studies evaluating risk stratification of ED patients with

potential acute coronary syndromes. Acad Emerg Med 2004;11: 1331-40.

1. Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702-6.
2. McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 2003;157:940-3.
3. Pan WH, Cedres LB, Liu K, et al. Relationship of clinical diabetes and asymptomatic hyperglycemia to risk of coronary heart disease mortality in men and women. Am J Epidemiol 1986;123:504-16.
4. Kleinman JC, Donahue RP, Harris MI, et al. Mortality among diabetics in a national sample. Am J Epidemiol 1988;128:389-401.
5. Raza JA, Movahed A. Current concepts of cardiovascular disease in diabetes mellitus. Int J Cardiol 2003;89:123-34.
6. Swap CJ, Nagurney JT. Value and limitations of chest pain history in the evaluation of patients with Suspected acute coronary syndromes. JAMA 2005;294:2623-9.
7. Raczynski JM, Taylor H, Cutter G, Hardin M, Rappaport N, Oberman A. Diagnoses, symptoms, and attribution of symptoms among black and white inpatients admitted for coronary heart disease. Am J Public Health 1994;84:951-6.
8. Taylor Jr HA, Canto JG, Sanderson B, Rogers WJ, Hilbe J. Management and outcomes for black patients with acute myocardial infarction in the reperfusion era. Am J Cardiol 1998;82:1019-23.
9. Borzak S, Joseph C, Havstad S, et al. Lower thrombolytic use for African Americans with myocardial infarction: an influence of clinical presentation? Am Heart J 1999;137:338-45.
10. Manhapra A, Canto JG, Barron HV, et al. Underutilization of reperfusion therapy in eligible African Americans with acute myocardial infarction: role of presentation and evaluation characte- ristics. Am Heart J 2001;142:604-10.
11. Blomkalns AL, Chen AY, Hochman JS, et al. gender disparities in the diagnosis and treatment of Non-ST-segment elevation acute coronary syndromes: large-scale observations from the CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines) National Quality Improvement Initiative. J Am Coll Cardiol 2005;45:832-7.
12. American Heart Association. Heart and Stroke Facts. 1992-2003. Available at: http://www.americanheart.org/downloadable/ heart/1056719919740HSFacts2003text.pdf.
13. Goldberg RJ, O’Donnell C, Yarzebski J, et al. sex differences in symptom presentation associated with AMI: a population-based perspective. Am Heart J 1998;136:189-95.
14. Milner KA, Funk M, Arnold A, Vaccarino V. Typical symptoms are predictive of acute coronary syndromes in women. Am Heart J 2002; 143:283-8.