Racial disparities in cardiovascular health are well documented1 and account for approximately one third of the difference in life expectancy between African Americans/blacks (blacks) and whites.2 Although disparities affect both sexes, many studies3,4 indicate that women, especially black women, are most at risk for disparate coronary heart disease (CHD) outcomes. In fact, CHD is the leading cause of death and disability in women in the United States,4 but the disparities between black and white women are evident in CHD mortality rates per 100,000 of 125.1 for white women while 160.3 for black women.4
One means of improving outcomes is to decrease the delay from time of onset of acute myocardial infarction (AMI) symptoms to presentation for medical assistance. Although optimal benefit, in terms of morbidity and mortality, requires myocardial reperfusion within 1 hour of infarction,5 many patients delay seeking treatment for AMI symptoms. The national data on median delay times range from 1.5 to 6 hours, with approximately 50% of individuals delaying longer than 4 hours.6 This translates into more than a quarter of a million people forecasted to die within 1 hour after the onset of AMI symptoms and before they reach the hospital this year alone.4
Women are consistently found to delay longer than men.3,7,8 Goldberg et al8 reported that the National Registry of Myocardial Infarction data (N = 364,131) demonstrated a significant difference in median treatment-seeking time of 2.3 hours for women as compared with 2 hours for men. Research indicates that blacks delay seeking treatment longer than whites.6,9,10 Despite the excess CHD mortality among black women, few studies of treatment-seeking delay have included sufficient numbers of women, especially blacks, to guide practice. Increased understanding of differences in delay between black and white women will guide development of evidence-based interventions to improve women's outcomes. This secondary data analysis compares AMI treatment-seeking delay intervals and describes correlates of delay in a comparable sample of black and white women.
Individuals who correctly attribute their AMI symptoms to a cardiac cause seek treatment faster than individuals who incorrectly attribute their symptoms to other causes.11,12 One reason misattribution of symptoms to other causes occurs is because personal expectations of AMI symptoms are often dissimilar to actual symptom experiences.13-15 Lee16 suggests that blacks delay longer than whites, in part, because they have a lower incidence of chest pain, the most commonly recognized AMI symptom, and therefore are less likely to attribute their symptoms to cardiac problems. Klingler et al17 note that perceptions of symptom etiology often differ for blacks and whites, which may also contribute to different delay times.
Large studies of the association between delay and various comorbidities, AMI risk factors, and other characteristics3,7,8,18 indicate that a history of hypertension,3,8 diabetes,3,8,19 and/or angina20 increases delay time, whereas a history of CHD or a recent cardiac procedure/event decreases delay time.3,8 Older age, social, and situational factors are commonly linked with delay.6 Low education and income levels are consistently and inversely associated with delay, whereas the effect of insurance is often paradoxical.6 However, it is difficult to compare data from multiple studies because researchers often examined different variables, did not report analysis by sex and race, or examined small samples of women and non-whites. Therefore, despite an abundance of research focusing on treatment-seeking delay, information that specifically investigates women and stratifies by race is lacking.
The research questions addressed in this article are as follows: (1) How do treatment-seeking delay intervals differ between black and white women? and (2) What clinical and personal characteristics are associated with getting to treatment within 2 hours of onset of AMI symptoms?
The researchers conducted a secondary analysis of data from the first author's study of prodromal and acute symptoms of AMI in women. Methods for the original study are summarized below, followed by those for this secondary analysis. A study that presents the primary results of this study, women's prodromal and acute symptoms of AMI, is currently under review.
For the original retrospective telephone survey, the researchers recruited a nonprobability sample of women from 15 medical centers in 8 states across the United States from 1999 to 2005. An employee from each center developed a list of women with a discharge diagnosis of AMI and telephoned each to request permission to release information to the research team. Besides having experienced an AMI in the previous 4 to 6 months, inclusion criteria were that women be 21 years or older; cognitively intact; have access to a telephone; identify themselves as black, Hispanic, or white; and speak English or Spanish.
After receiving approval from each site's Institutional Review Board, the research assistants contacted potentially eligible women, explained the study, gained consent, determined eligibility, and conducted the interviews. Using a computer-assisted telephone interview technique, we administered the McSweeney Acute and Prodromal Myocardial Infarction Symptom Survey21 4 to 6 months after the most recent AMI. The McSweeney Acute and Prodromal Myocardial Infarction Symptom Survey contains questions about symptoms before and during AMI, attribution of initial symptoms (ie, what women believed was causing their symptoms), time from onset of symptoms until seeking and receiving treatment, AMI risk factors and comorbidities, and demographics. Interviews averaged 60 minutes. The team received 1,935 names; 554 were ineligible, could not be located, or had died. We consented 1,294. After excluding 24 women who failed the Blessed cognitive screen,22 the final sample included 1,270 women.
The authors limited the sample for these analyses to black and white women, excluding the small group of Hispanic women. After excluding women for whom delay time could not be meaningfully calculated (n = 23), either because they were already hospitalized at the time of symptom onset or because they did not recall the time their often subtle, symptoms began (n = 26), the final analytic sample included 509 black women and 500 white women.
For our primary end point, we dichotomized delay time (time in hours between symptom onset and hospital arrival) as ≤2 hours versus >2 hours, similar to Gibler et al.3 We used the log transformation of delay time in analyses using delay time as a continuous variable. We constructed a conservative proxy for eligibility for public insurance (Medicare or Medicaid) that was positive if a woman was 65 years or older or reported an annual family income <$10,000. We dichotomized attribution of symptoms as either to cardiac or to noncardiac causes.
The significance of bivariate differences was assessed using χ2 tests, t tests, analysis of variance, and Wilcoxon and Kruskal-Wallis tests, as appropriate. Multivariate logistic regression analysis was used to assess associations between delay time (dichotomized) and demographic, clinical, and attributional variables, and ordinary least-squares multiple linear regression was used to explore associations with delay time as a continuous variable. We constructed a series of cross-product interaction terms to allow us to check for interactions between race and attribution, eligibility for public insurance, education, first AMI, and/or the pre-AMI clinical variables. Variance inflation factors were examined to assess multicollinearity among independent variables. We used SPSS version 14 (SPSS Inc, Chicago, Ill) for study analyses.
The characteristics of the study sample, overall and by race, are summarized in Table 1. On average, black participants were significantly younger, less affluent (median income <$10,000 vs $20,000-$29,999 for white women), less well educated, and more likely to have been eligible for public insurance than their white counterparts. They reported significantly higher rates of pre-AMI diagnoses of diabetes, hypertension, and chronic heart failure (CHF). Black women were significantly less likely to report having attributed their symptoms to an AMI. There were no significant differences by race for other clinical variables or for delay time (continuous or categorical).
In bivariate analyses, attribution of symptoms to AMI (χ2 = 34.87, 1df; P < .01) and eligibility for public insurance (χ2 = 5.94, 1df; P < .05) were significantly associated with reports of receiving treatment within 2 hours of symptom onset. Bivariate associations with all other sociodemographic and clinical variables were nonsignificant, as were associations with year of AMI and state of residence. We observed the same pattern of associations for the log-transformed continuous delay-time variable: attribution of symptoms to AMI and eligibility for public insurance (t = 4.01, 556df; P < .01); pre-AMI CHF (Wilcoxon w = 60,603; P = .07); and state, year of AMI, and other clinical variables (nonsignificant).
We regressed delay time (medical treatment within ≤2 hours of symptom onset) on attribution, eligibility for public insurance, race, age, education, first AMI, pre-AMI clinical history variables (angina, CHD, CHF, diabetes, hypertension), and an interaction term for race and insurance eligibility. The model was statistically significant (P < .001) as were the attribution (P < .001) and insurance variables (P = .001). None of the interactions between race and demographic or clinical variables was statistically significant at the α level of .05. However, given the results for the interaction between race and eligibility for public insurance (P = .055), odds ratios and 95% confidence intervals are shown separately by race in Table 2. After adjustment for all other variables in the model, symptom attribution and eligibility for public insurance increased the likelihood that both black and white women would get to treatment within 2 hours of symptom onset. The magnitude of the effect differed substantially by race. Black women who attributed their symptoms to AMI were nearly 3 times as likely to get to treatment within 2 hours as those who did not, whereas white women who attributed their symptoms to AMI were nearly 4 times as likely to get treatment within 2 hours. Black women who were eligible for public insurance (Medicare or Medicaid) were 2.3 times as likely to get treatment within 2 hours as those who were not, whereas white women eligible for public insurance were only 1.35 times as likely to get to treatment within 2 hours. The association for white women was not statistically significant. Because the stratified analyses have half the sample size of the combined analyses, the associations of pre-AMI CHD and CHF with delay time are less significant, although they are of the same magnitude.
When we repeated the analyses using multiple linear regression with log-transformed continuous delay time as the outcome variable (data not shown), the results were essentially the same. Attribution of symptoms to a cardiac origin and eligibility for public insurance were statistically significant in the main effects models, overall (β = −.40, t = −6.18, 1df; P < .001, and β = − .233, t = −3.41, 1df; P = .001, respectively) and stratified by race. No other variable approached statistical significance.
This study contributes detailed information about a relatively large sample of black women and allows direct comparison to a similar group of white women. Although our subsamples of black and white women were comparable in size, there were significant differences. Black women were younger, less affluent and educated, and more likely to be eligible for public insurance. They were also significantly more likely to report a history of diabetes, hypertension, and CHF. Although numerous studies6-8 indicate that these characteristics should increase delay time, interestingly, black women in our study had a mean delay time of 12.9 hours and median of 1.0 hour compared with white women's 13.3 and 1.5 hours. These differences were nonsignificant, but the racial difference approached significance in the main effects model.
Other large studies3,7,8 have reported median delay times of 1.4 to 2.5 hours. Recent studies that analyzed delay by race reported median delays among blacks of 2.418 to 4.38 hours17 but did not report gender-specific data. Others investigating women's median delay times reported ranges from 2.0 to 4.25 hours8,11,23-25 but did not report findings by race or had predominantly white samples. Although these studies identified blacks and women as more likely to delay, they did not look at gender and race simultaneously. Banks and Dracup's26 recent study of 32 black women with AMI also reported a median delay of 4.42 hours.
Characteristics of the sample may have combined to reduce delay time overall and equalize those for black and white women in our sample. Although overall whites were more affluent and better educated than blacks, women in both groups were primarily of low socioeconomic status. Annual family income was under $30,000 for 57% of white women and under $10,000 for 21%. In addition, 67% of white women had a high school education or less, and 12% had less than a 9th grade education. The majority of both groups was eligible for public insurance.
National campaigns to increase awareness of heart disease in women have effectively increased awareness from 30% in 1997 to 55% in 2005.27 Because black women were recruited later than white women and the Heart Truth campaign was initiated in 2002, we explored whether delay time might have decreased as awareness of CHD as the number one cause of death in women increased. Despite improved awareness of CHD in the population, as in other studies,3,7,8 we did not observe a significant change in delay time over the 6 years of this study.
In our study, attribution of symptoms to cardiac causes was a significant predictor of decreased delay time. Black and white women who attributed their symptoms to AMI were nearly 3 and 4 times, respectively, as likely to get to treatment within 2 hours as those who did not. In our study, significantly more whites than blacks correctly attributed symptoms to AMI, a finding supported by others.11,12 Like the women in our study, women in a study by Moser et al.,11 who attributed symptoms correctly, sought treatment approximately 3 times as fast as those who did not.
Another significant finding in our study was that black women eligible for public insurance (70%) were 2.3 times as likely to seek treatment within 2 hours after symptom onset as those who were not public-insurance eligible. Although not significant, white women eligible for public insurance (65%) were 1.35 times as likely to seek treatment within 2 hours as those who were not eligible for public insurance. There is conflicting information in the literature about the effect of insurance type on delay time. Some investigators have found no effect.28,15,29 However, Gibler et al3 and O'Donnell et al30 compared those with public versus private insurance and reported that private insurance holders demonstrated shorter delays. Investigators who examined the effect of insurance on Emergency Medical System (EMS) use24 among patients with chest pain found that individuals in low-income groups (<$30,000 annually) who had public insurance that included EMS coverage were 3.8 times as likely to activate the EMS as those without EMS coverage. It is well documented that ambulance use significantly decreases delay time.18
Public insurance may be an increasingly important factor in reducing delay among women because out-of-pocket costs associated with private insurance, such as co-payments, have skyrocketed. Kaiser reported that a 1-year 45% increase in out-of-pocket costs contributed to 18% of insured individuals postponing medical care.31 Out-of-pocket costs have continued to escalate since the Kaiser Report and are most likely further influencing delay in those with private insurance. Because fewer white women in this study were eligible for public insurance, concerns about costs may have contributed to their longer delays in seeking treatment. However, being eligible for public insurance had a much stronger impact on delay times for black than for white women (odds ratio = 2.3 vs 1.35). Additional research using a more precise measure of insurance status is needed to better understand the apparent differential influence of eligibility for public insurance on delay in seeking treatment.
This secondary analysis is not without limitations. First, we used self-reported, retrospective data collected from a nonprobability sample. However, Zerwic et al25 documented an r = 0.86 correlation in delay intervals between self-report and medical record abstraction, indicating that self-report is a reliable source. Second, all data in this study are from survivors; delay times and attribution of symptoms for women who died immediately or within the first 3 months after their AMI may have differed. Third, our insurance indicator is an imprecise and very conservative proxy. We recognize that some of the women identified as not being eligible for public insurance may have had public insurance, thus potentially biasing results. However, such misclassification should bias findings toward a finding of no difference. Finally, because of differential recruitment rates, white women were recruited earlier than black women. Because media outlets have increasingly discussed women's symptoms of AMI, those recruited later in the study, predominantly black women, may have recognized AMI symptoms more quickly than those recruited earlier in the study, potentially influencing delay time. However, as noted earlier, we did not observe a significant change in delay during the course of this study.
The strongest and most striking finding from this secondary analysis was that those who attributed their symptoms to AMI were more likely to seek medical assistance within 2 hours after onset of AMI symptoms. This clearly indicates the need for more education for all women, especially those at high risk for AMI. A 2-pronged approach of increasing awareness of the consequences of CHD in women and of identifying the most common AMI symptoms for women of all races/cultures is essential.
Another important observation was that delay time did not decrease significantly over the study period, although public campaigns were instituted during this period. Public awareness of CHD as the major killer in US women is increasing.27 The most recent survey indicates that 55% of women now recognize that CHD is the leading cause of death among women, but awareness was greater in white and lower in minority women.27 Although encouraging, 45% of all women have not received this message. Women must understand their own risk factors and we must assist them in linking awareness of women's CHD with their own personal risk factors. Nurses are in a prime position to elevate awareness and assist women to develop an accurate personal risk analysis during routine examinations, follow-up care for CHD risk conditions, such as diabetes and hypertension, or any hospitalization. We must take advantage of these teachable moments to identify risk and increase awareness.
We also need additional research with racially diverse women to develop a comprehensive evidence base identifying the most common AMI symptoms. Although in this study, women who correctly attributed their symptoms to cardiac causes were 3 to 4 times as likely to seek care within 2 hours, it is striking that less than 20% of women reported attributing their symptoms to AMI. It is difficult for women to attribute symptoms correctly and to decrease delay times if they do not know women's most common AMI symptoms. Recognition of symptoms and correct attribution to cardiac causes coupled with an accurate personalized assessment for CHD risk should provide women with the impetus to seek timely treatment should AMI symptoms develop.
1. Lurie N, Fremont A, Jain AK, et al. Racial and ethnic disparities in care: the perspectives of cardiologists. Circulation
2. Wong MD, Shapiro MF, Boscardin WJ, Ettner SL. Contribution of major diseases to disparities in mortality. N Engl J Med
3. Gibler WB, Armstrong PW, Ohman EM, et al. Persistence of delays in presentation and treatment for patients with acute myocardial infarction
: the GUSTO-I and GUSTO-III experience. Ann Emerg Med
4. American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics-2006 update. Circulation
5. Gruppo Italiano per lo Studio della Streptochinasi nell' Infarcto Miocardico (GISSI). Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction
6. Moser DK, Kimble LP, Alberts MJ, et al. Reducing delay in seeking treatment by patients with acute coronary syndrome and stroke. A scientific statement from the AHA Council on Cardiovascular Nursing and Stroke Council. Circulation
7. McGinn AP, Rosamond WD, Goff DC Jr, Taylor HA, Miles JS, Chambless L. Trends in prehospital delay time and use of emergency medical services for acute myocardial infarction
: experience in 4 US communities from 1987-2000. Am Heart J
8. Goldberg RJ, Gurwitz JH, Gore JM. Duration of, and temporal trends (1994-1997) in, prehospital delay in patients with acute myocardial infarction
: the second National Registry of Myocardial Infarction
. Arch Intern Med
9. National Heart Lung & Blood Institute. Educational strategies to prevent prehospital delay in patients at high risk for acute myocardial infarction
. Bethesda, MD: National Institutes of Health; 1997. Report No: 97-3787.
10. Lefler LL, Bondy KN. Women
's delay in seeking treatment with myocardial infarction
: a meta-synthesis. J Cardiovasc Nurs
11. Moser DK, McKinley S, Dracup K, Chung ML. Gender differences in reasons patients delay in seeking treatment for acute myocardial infarction
symptoms. Patient Educ Couns
12. Meischke H, Yasui Y, Kuniyuki A, Bowne DJ, Anderson R, Urban N. How women
label and respond to symptoms of acute myocardial infarction
. Heart Lung
13. Finnegan JR, Meischke H, Zapka JG, et al. Patient delay in seeking care for heart attack symptoms: findings from focus groups conducted in five U.S. regions. Prev Med
14. Zerwic JJ. Symptoms of acute myocardial infarction
: expectations of a community sample. Heart Lung
15. Morgan DM. Effect of incongruence of acute myocardial infarction
symptoms on the decision to seek treatment in a rural population. J Cardiovasc Nurs
16. Lee H. Typical and atypical clinical signs and symptoms of myocardial infarction
and delayed seeking of professional care among blacks. Am J Crit Care
17. Klingler D, Green-Weir R, Nerenz D, et al. Perceptions of chest pain differ by race. Am Heart J
18. Goff DC Jr, Feldman HA, McGovern PG, et al. Prehospital delay in patients hospitalized with heart attack symptoms in the United States: the REACT trial. Am Heart J
19. Canto JG, Shlipak MG, Rogers WJ, et al. Prevalence, clinical characteristics, and mortality among patients with myocardial infarction
presenting without chest pain. JAMA
20. Sheifer SE, Rathore SS, Gersh BJ, et al. Time to presentation with acute myocardial infarction
in the elderly: associations with race, sex, and socioeconomic characteristics. Circulation
21. McSweeney JC, O'Sullivan P, Cody M, Crane PB. Development of the McSweeney Acute and Prodromal Myocardial Infarction
Symptom Survey. J Cardiovasc Nurs
22. Katzman R, Brown T, Fuld P, Peck A, Schechter R, Schimmel H. Validation of a short orientation-memory-concentration test of cognitive impairment. Am J Psychiatry
23. Rosenfeld AG. Treatment-seeking delay among women
with acute myocardial infarction
: decision trajectories and their predictors. Nurs Res
24. Siepmann DB, Mann NC, Hedges JR, Daya MR. Association between prepayment systems and emergency medical services use among patients with acute chest discomfort syndrome. For the REACT Study. Ann Emerg Med
25. Zerwic JJ, Ryan CJ, DeVon HA, Drell MJ. Treatment seeking for acute myocardial infarction
symptoms: differences in delay across sex and race. Nurs Res
26. Banks AD, Dracup K. Factors associated with prolonged prehospital delay of African Americans with acute myocardial infarction
. Am J Crit Care
27. Mosca L, Mochari H, Christian A, et al. National study of women
's awareness, preventive action, and barriers to cardiovascular health. Circulation
28. Magid DJ, Koepsell TD, Every NR, et al. Absence of association between insurance copayments and delays in seeking emergency care among patients with myocardial infarction
. N Engl J Med
29. Ho KK, Lee SW, Ooi SB, Lateef F, Lim SH, Anantharaman V. Acute coronary syndrome-factors causing delayed presentation at the emergency department. Ann Acad Med Singapore
30. O'Donnell S, Condell S, Begley C, Fitzgerald T. Prehospital care pathway delays: gender and myocardial infarction
. J Adv Nurs
31. Henry J. Kaiser Family Foundation. Underinsured in America: Is health coverage adequate?
Washington, DC; 2002. Available at: http://www.kff.org/uninsured
. Accessed July 24, 2006.