- Email: [email protected]
Differences in care-seeking behavior for acute chest pain in the United States and Japan1
Differences in care-seeking behavior for acute chest pain in the United States and Japan Lawrence Liao, MD,b David J. Whellan, MD, MHS,b Katsuhiko Tabuchi, MD,c and Kevin A. Schulman, MDa Durham, NC, and Tokyo, Japan
Background Delay from onset of acute myocardial infarction symptoms to the delivery of medical care is a major determinant of prognosis. Although studies have explored patient factors for delay in seeking care, there are limited data on international differences in care-seeking behavior. Methods We surveyed 1032 people in the United States and 1422 people in Japan in January 1997 on decisionmaking responses to a chest pain scenario representing acute MI. Participants were asked about how they would seek initial care and how promptly they would seek care. Results
The mean age was 43.6 years in the United States and 48.3 years in Japan. For the hypothetical scenario, US respondents were more likely to seek care at an emergency department (22.9% vs 16.2% in Japan) or through emergency medical services/911 (55.9% vs 32.9% in Japan, P ⫽ .001). American subjects were also more likely to seek care immediately (83.1% vs 56.4% in Japan, P ⫽ .001).
Conclusion
Respondents in the United States and Japan differed substantially in their responses to a hypothetical chest pain scenario. Whether these differences result from cultural or health care system factors and whether these apparent attitudes produce gaps in real responses to acute coronary syndromes must be explored in further studies. (Am Heart J 2004;147:630 –5.)
Delay from the onset of symptoms to the delivery of medical care directly impacts the prognosis of patients with acute myocardial infarction (MI).1 A delay of as little as 30 minutes in the administration of reperfusion therapy for ST segment elevation MI may reduce life expectancy by an average of 1 year.2 Examination of this process has shown that the single largest share of the pre-hospital time is taken by patient delay in seeking or summoning help.3,4 Reasons for these delays include patients thinking pain would go away, not realizing the pain was serious, and attempting self-treatment with nitrates or other medications.5 Other studies have found that race, sex, and socioeconomic status may also substantially impact patient alacrity in seeking care.6,7
From the aCenter for Clinical and Genetic Economics, bDuke Clinical Research Institute, Duke University Medical Center, Durham, NC, and cDepartment of Internal Medicine, National Tokyo Medical Center, Tokyo, Japan. Supported by an educational grant from the Pfizer Japan Foundation. Guest Editor for this manuscript was Harlan M. Krumholz, MD, Yale University School of Medicine, New Haven, Conn. Submitted June 7, 2003; accepted October 13, 2003. Reprint requests: Kevin A. Schulman, Center for Clinical and Genetic Economics, Duke Clinical Research Institute, Box 17969, Durham, NC, 27715. E-mail: [email protected] 0002-8703/$ - see front matter © 2004, Elsevier Inc. All rights reserved. doi:10.1016/j.ahj.2003.10.006
Although several studies have explored patient factors for delay in seeking care, there are limited data on international differences in care-seeking behavior. With its distinct culture, universal health insurance system, and a coronary artery disease mortality that is one fifth the rate in the United States (US), Japan possesses several attributes that might impact care-seeking behavior of its residents.8,9 The purpose of this study was to examine similarities and differences in patient responses to a hypothetical chest pain scenario in Japan and the US.
Methods We commissioned the Gallup Organization (Princeton, NJ) to assess responses by nationally representative heads of household in the US and Japan to hypothetical chest pain scenarios. In January 1997, US households were randomly selected from 150 sampling locations across the country on the basis on the 1990 US Census report. A maximum of 8 respondents were surveyed from each of the 150 sampling locations. During the same period, Gallup interviewed heads of household in Japan randomly selected from 2000 possible panel homes in 157 sample locations in Japan.
Survey instrument The survey instrument was designed in English and translated into Japan with back-translation to ensure the consis-
American Heart Journal Volume 147, Number 4
tency of survey content (Appendix). Both instruments were reviewed by a bilingual Japanese physician. The survey consisted of questions related to decision-making in response to a scenario of acute chest pain representing acute MI. Participants were asked how they would seek initial care for their symptoms (call 911/emergency medical services, travel to emergency department, see their personal physician, see a cardiologist, go to a local health clinic, not seek care). They were asked how long they would wait to seek care (immediately, within a few hours, within a day, schedule an appointment, or not seek care). They were asked about their perceptions of emergent care costs and insurance coverage, the geographical proximity of emergency medical services to them, and their relations with a primary care physician.
Data collection The data for this analysis were collected as part of the US and Japanese Omnibus surveys administered by the Gallup Organization in January 1997. In addition, demographic data were gathered for all respondents. Responses from heads of household were obtained via personal interviews. The interviewers involved in the project were experienced and professionally trained Gallup staff who had been specifically briefed on the implementation of the interview. Survey items were administered as part of the monthly Gallup Omnibus Survey in the US and Japan. The US Omnibus survey averages from 150 to 200 items, whereas the Japanese Omnibus survey averages approximately 60 items. Surveys were completed during a 3-week time period during evenings and weekends. For both the US and Japan Omnibus surveys, when a respondent was not available during the first contact, 1 attempt was made to return to the respondent’s location to complete the interview.
Statistical analysis All data on demographics and responses are reported for the US and Japanese cohorts as continuous or categorical variables. For the only continuous variable (age), data are reported as means. Categorical variables are reported as percentiles with cohort differences assessed by 2 tests.
Results Gallup surveyed 1032 people in the US and 1422 people in Japan. Demographic characteristics of both populations are presented in Table I. The mean age was 43.6 years in the US and 48.3 years in Japan. The US respondents were more likely to be professionals or own a business (25% vs 15%) and less likely to be homemakers (23% vs 8%, P ⫽ .0001 for the occupation comparison). The US sample had a racial composition comparable to the national population (81% white, 15% African American); as expected, there was no racial variation in the Japanese sample.
Access to care and perceptions of insurance coverage Data on perceptions of access to care and emergent cost coverage are shown in Table II. Respondents in
Liao et al 631
Table I. Population demographics
No. Females (%) Mean age (y) Age breakdown (%) 18–29 y 30–39 y 40–49 y 50–59 y ⱖ60 y Race (%) White Black Other Education (%) Less than high school High school graduate Some college/trade school College graduate or beyond Occupation (%) Farmer/fisherman Professional/own business Office worker/clerical/sales Laborer/skilled worker Freelance/manager/other Homemaker Full-time student Nonworker/retired
United States
Japan
1032 51 43.6
1433 54 48.3
21 24 20 13 23
13 16 22 20 29
81 15 4
N/A N/A N/A
19 29 31 22
22 52 N/A* 26
1 25 10 33 1 8 4 14
3 15 19 22 4 23 2 13
*This category was not collected in the Japanese survey.
the US claimed greater access to care than respondents in Japan. More subjects in the US reported living ⬍30 minutes from the closest emergency department, and fewer subjects reported living ⬎2 hours from the closest emergency department. Respondents in the US also reported greater cost coverage for emergent care. In the US, 33.9% of respondents, compared with 8% of respondents in Japan, reported insurance coverage exceeding 90% of costs (P ⫽ .0001). Although Japan has legally mandated universal health coverage, nationally regulated uniform fee schedules, and strictly prohibited balance billing (provider charges for more than the legally set fee schedule), 9.7% of Japanese respondents reported having no emergent care cost coverage.8,10 US respondents were also more likely to claim having a regular physician, although historically, Japanese consult doctors more often than Americans (11.9 visits and 1.7–7.1 visits per person per year, respectively).8
Response to potential myocardial infarction scenario Data for the potential myocardial infarction scenario are shown in Table III. Differences between the 2 countries in preferences for where to seek care were apparent (P ⫽ .001). Respondents in the US were more likely to seek care at an emergency department
American Heart Journal April 2004
632 Liao et al
Table II. Access to care and perceptions of insurance coverage
Travel time to closest ER No. ⬍30 Minutes (%) 30–60 Minutes (%) 1–2 Hours (%) ⬎2 Hours (%) Don’t know/other (%) Has a regular physician No. Yes (%) No (%) Don’t know (%) Reported emergent cost coverage No. None/no health insurance (%) ⬍50% (%) 50–75% (%) 76–90% (%) 91–99% (%) 100% (%) Flat co-pay (%) Coverage varies by cost of services (%) Don’t know/other (%)
United States
Japan
1031 85.8 11.3 1.5 0.1 1.4
1385 80.9 12.3 1.9 3.2 1.6
1031 72.9 26.5 0.6
1198 57.8 41.8 0.4
1031 12
1379 9.7
3.5 7.9 28.8 13.4 20.5 2.7 4.6
28 14.4 12 4.3 3.7 1.6 4.2
6.6
23.9
P
Discussion .0001
.0001
Care seeking site No. EMS/911 (%) Emergency room (%) Local health clinic (%) Personal physician (%) Cardiologist (%) Not seek care (%) Don’t know/other (%) Care seeking time No. Immediately (%) Within a few hours (%) Within a day (%) Schedule appointment for the next few days (%) Don’t know/other (%)
Japan
1031 55.9 22.9 1.9 9 3.3 3.7 3.3
1433 32.9 16.2 10.1 23.5 6.8 6.8 3.7
993 83.1 8.5 2.7 3.4
1335 56.4 15.2 16 8.8
2.3
3.6
In this first study to explore international care-seeking behavior for chest pain, respondents in the US and Japan differed substantially in their responses to hypothetical chest pain scenarios. US respondents more readily recognized the emergent nature of the potential MI scenario. Respondents in Japan were much less likely to respond immediately or seek care in an emergent setting for the potential MI scenario. These findings may reflect differences in culture, health education, or health care systems.
Previous work .0001
Table III. Care-seeking response to potential myocardial infarction scenario United States
to seek care (P ⫽ .001). Compared with their Japanese counterparts, US survey subjects were more likely to seek care immediately (83.1% vs 56.4% in Japan).
P
.001
.001
(22.9% vs 16.2% in Japan) or through emergency medical services/911 (55.9% vs 32.9% in Japan). US respondents were less likely to seek care with their personal physician (9% vs 23.5% in Japan) or not seek any care (3.7% vs 6.8% in Japan). Similarly, there were clear differences in preferences in the 2 countries for time
Several studies have examined reasons for pre-hospital delay in patients with acute coronary syndromes. These studies have found that a variety of factors including race, sex, socioeconomic status, and individual patient attitudes may all substantially impact alacrity in seeking care.5–7,11–16 In their survey of 313 consecutive patients with MI in Great Britain, Leslie and colleagues found that the 2 most frequently cited reasons for delay in seeking medical help were “thinking that symptoms would go away” and “not thinking it was serious.”5 Other researchers have found that individual demographic factors impact care-seeking delay for acute chest pain. Sheifer and colleagues examined 102,339 elderly patients from the Cooperative Cardiovascular Project and found that diabetes mellitus, history of angina, female sex, black race, and poverty were all associated with late arrival (ⱖ6 hours after symptom onset).6 Similar findings were seen when Ell and colleagues surveyed 448 African-American patients admitted with acute chest pain in Los Angeles.7 In that study, public hospital patients and patients without health insurance had significantly longer care-seeking delays.7 In their study of care-seeking behavior, Weissman and colleagues surveyed 12,068 patients in 5 Massachusetts hospitals and found that black race, poverty, lack of insurance, and lack of a regular physician were all associated with pre-hospital delays in care.17 In the more recent Global Registry of Acute Coronary Events (GRACE; 1999 –2001), the 3693 patients with ST-segment elevation MI had a median pre-hospital delay of 3.0 hours (mean, 6.1 hours).18 Factors associated in GRACE with longer delays included age, prior heart failure, early morning symptom onset, and presentation in Europe or South America.18
American Heart Journal Volume 147, Number 4
Differences in time to presentation in the US and Japan for acute coronary syndromes There are limited data comparing pre-hospital delays for acute coronary syndrome patients in the US and Japan. In the US, the largest contemporary cohort was recently reported from the National Registry of Myocardial Infarction 3 (NRMI-3).19 For the 114,405 patients with confirmed MI, the mean time to presentation was 5.96 hours. In both the Worcester Heart Attack Study (1997 study year) and in the baseline phase of the Rapid Early Action for Coronary Treatment (REACT) multicenter trial (1995–1996), the median pre-hospital delay was 2.0 hours (mean delay, 4.3 hours and 5.0 hours, respectively).20,21 Data from comparable Japanese cohorts are not currently available. There are data from smaller series, including Takahashi and colleagues’ 106 patients who underwent primary angioplasty for acute MI.22 In this selected group, the mean ischemic time was 4.7 ⫾ 2.6 hours.22 Such times are similar to reperfusion intervals seen in the recent Stent PAMI multicenter trial.23 In 2 older series from Japan, the mean pre-hospital patient delay times have ranged from 52 minutes in Asahikawa City to almost 6 hours in Tokyo.24 –26 Because of the comparative paucity of recent data from Japan regarding pre-hospital delays, it is difficult to assess whether the attitudes toward chest pain seen in this survey result in measurable differences in pre-hospital delay for acute coronary syndrome patients in Japan.
Policy implications Because of the tremendous morbidity and mortality associated with delays in presentation for acute coronary syndromes, interventions to minimize these delays would have the potential to significantly impact patient outcomes. In particular, because of the numerous previous studies of US care seeking showing lack of insurance may contribute to pre-hospital delays, interventions to reduce or eliminate the number of uninsured patients may be attractive to policy makers. However, our findings suggest that such strategies would not be a panacea; patient-reported care-seeking delays for the potential MI scenario were much higher in Japanese respondents even though Japan’s national health care system actually has mandatory universal coverage.10 Interventions specifically targeted to improve access through increasing insurance coverage have proven disappointing.27 Public education efforts have also shown mixed effects on patient behavior.28 –30 In the REACT trial, an aggressive 18-month public education campaign encompassing community and patient groups, small media (direct mail, billboards, public service announcements), and mass media (television, newspapers, radio) had no appreciable benefit on time from symptom on-
Liao et al 633
set to hospital arrival for patients with chest pain.31 In the intervention and reference communities, median time to hospital arrival was 130 minutes and 126 minutes, respectively.31 These results contrast with those of media and educational campaigns undertaken in Europe, which have produced measurable decreases in delay times.29 One hypothesis for the failure of efforts such as REACT to improve delay times in the US relates to the short median delay times at baseline compared with the longer delays (3– 4 hours) in the successful European studies.29,31 It may be difficult to reduce delay times to ⬍2 hours with an educational campaign. Historically, Japanese delay times appear closer to the lengths reported in the various European studies.26,29 Consequently, more vigorous public educational efforts, such as that reported by Shibata and colleagues in Asahikawa City, might produce significant improvements in patient delays and resulting survival.25
Strengths and limitations This study has several important strengths. First, this is the first study to examine differences in care-seeking behavior in the US and Japan. Second, samples were drawn from nationally representative population samples in the 2 countries on the basis of each country’s national population census. Third, respondents were randomly selected for in-person interviews. Fourth, interviews were conducted by professional, trained Gallup Organization personnel. This study has several limitations. First, these data reflect survey responses to hypothetical chest pain scenarios. Although care was made to match the US and Japan surveys, it is possible that word choices or other elements of the scenarios may have influenced 1 group differently from the other. Second, this study compared national samples from the US and Japan. Such groups have inherently different demographic and cultural characteristics. In assessing care-seeking responses, we cannot separate the effects of these disparate factors from the influence of the different national health systems. Japanese respondents reported substantially less emergent health care coverage than US respondents, although the Japanese, by law, have universal health coverage. These results may reflect a substantial misconception by Japanese respondents of either the survey question or their legally mandated health insurance benefits. Third, these data were collected in January 1997. Although it is unlikely that care-seeking behaviors in the 2 countries have been radically altered since the survey was completed, it is possible that details such as the growth of the uninsured population in the US may subtly change the results of more contemporary examinations.
634 Liao et al
Conclusion Respondents in the US and Japan differed substantially in their responses to a hypothetical chest pain scenario. Whether these differences result from cultural or health care system factors and whether these apparent attitudes produce delays in real responses to acute coronary syndromes must be explored in further studies. If our results reflect genuine delays in Japanese patient responses, improved public educational efforts may be helpful in addressing these disparities.
References 1. Penny WJ. Patient delay in calling for help: the weakest link in the chain of survival? Heart 2001;85:121–2. 2. Rawles JM. Quantification of the benefit of earlier thrombolytic therapy: five-year results of the Grampian Region Early Anistreplase Trial (GREAT). J Am Coll Cardiol 1997;30:1181– 86. 3. The pre-hospital management of acute heart attacks: recommendations of a Task Force of the European Society of Cardiology and the European Resuscitation Council. Eur Heart J 1998;19:1140 – 64. 4. Leitch JW, Birbara T, Freedman B, et al. Factors influencing the time from onset of chest pain to arrival at hospital. Med J Aust 1989;150:6 –10. 5. Leslie WS, Urie A, Hooper J, et al. Delay in calling for help during myocardial infarction: reasons for the delay and subsequent pattern of accessing care. Heart 2000;84:137– 41. 6. 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 2000;102: 1651– 6. 7. Ell K, Haywood LJ, Sobel E, et al. Acute chest pain in African Americans: factors in the delay in seeking emergency care. Am J Public Health 1994;84:965–70. 8. Otaki J. Considering primary care in Japan. Acad Med 1998;73: 662– 8. 9. Iso H, Shimamoto T, Kitamura A, et al. Trends of cardiovascular risk factors and diseases in Japan: implications for primordial prevention. Prev Med 1999;29:S102–5. 10. Ikegami N. The economics of health care in Japan. Science 1992; 258:614 – 8. 11. Yarzebski J, Goldberg RJ, Gore JM, et al. Temporal trends and factors associated with extent of delay to hospital arrival in patients with acute myocardial infarction: the Worcester Heart Attack Study. Am Heart J 1994;128:255– 63. 12. Goldberg RJ, Gurwitz J, Yarzebski J, et al. Patient delay and receipt of thrombolytic therapy among patients with acute myocardial infarction from a community-wide perspective. Am J Cardiol 1992;70:421–5. 13. Moss AJ, Wynar B, Goldstein S. Delay in hospitalization during the acute coronary period. Am J Cardiol 1969;24:659 – 65. 14. Bett N, Aroney G, Thompson P. Delays preceding admission to hospital and treatment with thrombolytic agents of patients with possible heart attack. Aust N Z J Med 1993;23:312–3. 15. Schmidt SB, Borsch MA. The prehospital phase of acute myocardial infarction in the era of thrombolysis. Am J Cardiol 1990;65: 1411–5.
American Heart Journal April 2004
16. Cooper RS, Simmons B, Castaner A, et al. Survival rates and prehospital delay during myocardial infarction among black persons. Am J Cardiol 1986;57:208 –11. 17. Weissman JS, Stern R, Fielding SL, et al. Delayed access to health care: risk factors, reasons, and consequences. Ann Intern Med 1991;114:325–31. 18. Goldberg RJ, Steg PG, Sadiq I, et al. Extent of, and factors associated with, delay to hospital presentation in patients with acute coronary disease (the GRACE registry). Am J Cardiol 2002;89: 791– 6. 19. Mathew V, Gersh B, Barron H, et al. Inhospital outcome of acute myocardial infarction in patients with prior coronary artery bypass surgery. Am Heart J 2002;144:463–9. 20. Goldberg RJ, Yarzebski J, Lessard D, et al. Decade-long trends and factors associated with time to hospital presentation in patients with acute myocardial infarction: the Worcester Heart Attack Study. Arch Intern Med 2000;160:3217–23. 21. 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 1999;138:1046 –57. 22. Takahashi T, Anzai T, Yoshikawa T, et al. Effect of preinfarction angina pectoris on ST-segment resolution after primary coronary angioplasty for acute myocardial infarction. Am J Cardiol 2002; 90:465–9. 23. Brodie BR, Stone GW, Morice MC, et al. Importance of time to reperfusion on outcomes with primary coronary angioplasty for acute myocardial infarction (results from the Stent Primary Angioplasty in Myocardial Infarction Trial). Am J Cardiol 2001;88: 1085–90. 24. Shibata J, Tanazawa S, Hirasawa K, et al. Deaths in early phase of acute myocardial infarction and approaches for reducing the in-hospital and out-hospital case-fatality rates of the disease. Jpn Circ J 1987;51:325–31. 25. Shibata J, Hirasawa K, Tateda K. The importance of early treatment of myocardial infarction and approaches for shortening delays in securing medical care. Jpn Circ J 1984;48: 721– 8. 26. Takano T, Endo T, Tanaka K, et al. Current status of prehospital care of patients with acute myocardial infarction in Tokyo: analysis of 3-year experience with coronary care unit network. Jpn Circ J 1987;51:338 – 43. 27. Siemiatycki J, Richardson L, Pless IB. Equality in medical care under national health insurance in Montreal. N Engl J Med 1980; 303:10 –5. 28. Ho MT, Eisenberg MS, Litwin PE, et al. Delay between onset of chest pain and seeking medical care: the effect of public education. Ann Emerg Med 1989;18:727–31. 29. Blohm MB, Hartford M, Karlson BW, et al. An evaluation of the results of media and educational campaigns designed to shorten the time taken by patients with acute myocardial infarction to decide to go to hospital. Heart 1996;76:430 – 4. 30. Rowley JM, Hill JD, Hampton JR, et al. Early reporting of myocardial infarction: impact of an experiment in patient education. Br Med J 1982;284:1741– 6. 31. Luepker RV, Raczynski JM, Osganian S, et al. Effect of a community intervention on patient delay and emergency medical service use in acute coronary heart disease: the Rapid Early Action for Coronary Treatment (REACT) Trial. JAMA 2000;284: 60 –7.
American Heart Journal Volume 147, Number 4
Appendix Survey scenario “For the next few questions, I would like you to tell me what you would do if you were in the situation I am about to describe. For the first example, please imagine that while carrying groceries you suddenly feel a crushing pressure in the middle of your chest. The pressure feels almost as if someone was squeezing your chest. The pain is very bad and you are forced to sit down. You notice that you are out of breath and that you have begun to sweat, even though it is not hot outside. You are surprised by the pain, because it came on rather suddenly. The pain does not go away for 10
Liao et al 635
minutes. Which one of the following best describes what you would do if you were in this situation?”
Other survey questions “How quickly would you seek care if you were in this situation?” “How long would you estimate that it would take you to get to the emergency room or hospital closest to your home?” “If you needed emergency medical care, how much of the cost would you estimate your health insurance plan would cover? If you do not have health insurance, please tell me so.” “Do you currently have a doctor who you see on a regular basis?”
Background Delay from onset of acute myocardial infarction symptoms to the delivery of medical care is a major determinant of prognosis. Although studies have explored patient factors for delay in seeking care, there are limited data on international differences in care-seeking behavior. Methods We surveyed 1032 people in the United States and 1422 people in Japan in January 1997 on decisionmaking responses to a chest pain scenario representing acute MI. Participants were asked about how they would seek initial care and how promptly they would seek care. Results
The mean age was 43.6 years in the United States and 48.3 years in Japan. For the hypothetical scenario, US respondents were more likely to seek care at an emergency department (22.9% vs 16.2% in Japan) or through emergency medical services/911 (55.9% vs 32.9% in Japan, P ⫽ .001). American subjects were also more likely to seek care immediately (83.1% vs 56.4% in Japan, P ⫽ .001).
Conclusion
Respondents in the United States and Japan differed substantially in their responses to a hypothetical chest pain scenario. Whether these differences result from cultural or health care system factors and whether these apparent attitudes produce gaps in real responses to acute coronary syndromes must be explored in further studies. (Am Heart J 2004;147:630 –5.)
Delay from the onset of symptoms to the delivery of medical care directly impacts the prognosis of patients with acute myocardial infarction (MI).1 A delay of as little as 30 minutes in the administration of reperfusion therapy for ST segment elevation MI may reduce life expectancy by an average of 1 year.2 Examination of this process has shown that the single largest share of the pre-hospital time is taken by patient delay in seeking or summoning help.3,4 Reasons for these delays include patients thinking pain would go away, not realizing the pain was serious, and attempting self-treatment with nitrates or other medications.5 Other studies have found that race, sex, and socioeconomic status may also substantially impact patient alacrity in seeking care.6,7
From the aCenter for Clinical and Genetic Economics, bDuke Clinical Research Institute, Duke University Medical Center, Durham, NC, and cDepartment of Internal Medicine, National Tokyo Medical Center, Tokyo, Japan. Supported by an educational grant from the Pfizer Japan Foundation. Guest Editor for this manuscript was Harlan M. Krumholz, MD, Yale University School of Medicine, New Haven, Conn. Submitted June 7, 2003; accepted October 13, 2003. Reprint requests: Kevin A. Schulman, Center for Clinical and Genetic Economics, Duke Clinical Research Institute, Box 17969, Durham, NC, 27715. E-mail: [email protected] 0002-8703/$ - see front matter © 2004, Elsevier Inc. All rights reserved. doi:10.1016/j.ahj.2003.10.006
Although several studies have explored patient factors for delay in seeking care, there are limited data on international differences in care-seeking behavior. With its distinct culture, universal health insurance system, and a coronary artery disease mortality that is one fifth the rate in the United States (US), Japan possesses several attributes that might impact care-seeking behavior of its residents.8,9 The purpose of this study was to examine similarities and differences in patient responses to a hypothetical chest pain scenario in Japan and the US.
Methods We commissioned the Gallup Organization (Princeton, NJ) to assess responses by nationally representative heads of household in the US and Japan to hypothetical chest pain scenarios. In January 1997, US households were randomly selected from 150 sampling locations across the country on the basis on the 1990 US Census report. A maximum of 8 respondents were surveyed from each of the 150 sampling locations. During the same period, Gallup interviewed heads of household in Japan randomly selected from 2000 possible panel homes in 157 sample locations in Japan.
Survey instrument The survey instrument was designed in English and translated into Japan with back-translation to ensure the consis-
American Heart Journal Volume 147, Number 4
tency of survey content (Appendix). Both instruments were reviewed by a bilingual Japanese physician. The survey consisted of questions related to decision-making in response to a scenario of acute chest pain representing acute MI. Participants were asked how they would seek initial care for their symptoms (call 911/emergency medical services, travel to emergency department, see their personal physician, see a cardiologist, go to a local health clinic, not seek care). They were asked how long they would wait to seek care (immediately, within a few hours, within a day, schedule an appointment, or not seek care). They were asked about their perceptions of emergent care costs and insurance coverage, the geographical proximity of emergency medical services to them, and their relations with a primary care physician.
Data collection The data for this analysis were collected as part of the US and Japanese Omnibus surveys administered by the Gallup Organization in January 1997. In addition, demographic data were gathered for all respondents. Responses from heads of household were obtained via personal interviews. The interviewers involved in the project were experienced and professionally trained Gallup staff who had been specifically briefed on the implementation of the interview. Survey items were administered as part of the monthly Gallup Omnibus Survey in the US and Japan. The US Omnibus survey averages from 150 to 200 items, whereas the Japanese Omnibus survey averages approximately 60 items. Surveys were completed during a 3-week time period during evenings and weekends. For both the US and Japan Omnibus surveys, when a respondent was not available during the first contact, 1 attempt was made to return to the respondent’s location to complete the interview.
Statistical analysis All data on demographics and responses are reported for the US and Japanese cohorts as continuous or categorical variables. For the only continuous variable (age), data are reported as means. Categorical variables are reported as percentiles with cohort differences assessed by 2 tests.
Results Gallup surveyed 1032 people in the US and 1422 people in Japan. Demographic characteristics of both populations are presented in Table I. The mean age was 43.6 years in the US and 48.3 years in Japan. The US respondents were more likely to be professionals or own a business (25% vs 15%) and less likely to be homemakers (23% vs 8%, P ⫽ .0001 for the occupation comparison). The US sample had a racial composition comparable to the national population (81% white, 15% African American); as expected, there was no racial variation in the Japanese sample.
Access to care and perceptions of insurance coverage Data on perceptions of access to care and emergent cost coverage are shown in Table II. Respondents in
Liao et al 631
Table I. Population demographics
No. Females (%) Mean age (y) Age breakdown (%) 18–29 y 30–39 y 40–49 y 50–59 y ⱖ60 y Race (%) White Black Other Education (%) Less than high school High school graduate Some college/trade school College graduate or beyond Occupation (%) Farmer/fisherman Professional/own business Office worker/clerical/sales Laborer/skilled worker Freelance/manager/other Homemaker Full-time student Nonworker/retired
United States
Japan
1032 51 43.6
1433 54 48.3
21 24 20 13 23
13 16 22 20 29
81 15 4
N/A N/A N/A
19 29 31 22
22 52 N/A* 26
1 25 10 33 1 8 4 14
3 15 19 22 4 23 2 13
*This category was not collected in the Japanese survey.
the US claimed greater access to care than respondents in Japan. More subjects in the US reported living ⬍30 minutes from the closest emergency department, and fewer subjects reported living ⬎2 hours from the closest emergency department. Respondents in the US also reported greater cost coverage for emergent care. In the US, 33.9% of respondents, compared with 8% of respondents in Japan, reported insurance coverage exceeding 90% of costs (P ⫽ .0001). Although Japan has legally mandated universal health coverage, nationally regulated uniform fee schedules, and strictly prohibited balance billing (provider charges for more than the legally set fee schedule), 9.7% of Japanese respondents reported having no emergent care cost coverage.8,10 US respondents were also more likely to claim having a regular physician, although historically, Japanese consult doctors more often than Americans (11.9 visits and 1.7–7.1 visits per person per year, respectively).8
Response to potential myocardial infarction scenario Data for the potential myocardial infarction scenario are shown in Table III. Differences between the 2 countries in preferences for where to seek care were apparent (P ⫽ .001). Respondents in the US were more likely to seek care at an emergency department
American Heart Journal April 2004
632 Liao et al
Table II. Access to care and perceptions of insurance coverage
Travel time to closest ER No. ⬍30 Minutes (%) 30–60 Minutes (%) 1–2 Hours (%) ⬎2 Hours (%) Don’t know/other (%) Has a regular physician No. Yes (%) No (%) Don’t know (%) Reported emergent cost coverage No. None/no health insurance (%) ⬍50% (%) 50–75% (%) 76–90% (%) 91–99% (%) 100% (%) Flat co-pay (%) Coverage varies by cost of services (%) Don’t know/other (%)
United States
Japan
1031 85.8 11.3 1.5 0.1 1.4
1385 80.9 12.3 1.9 3.2 1.6
1031 72.9 26.5 0.6
1198 57.8 41.8 0.4
1031 12
1379 9.7
3.5 7.9 28.8 13.4 20.5 2.7 4.6
28 14.4 12 4.3 3.7 1.6 4.2
6.6
23.9
P
Discussion .0001
.0001
Care seeking site No. EMS/911 (%) Emergency room (%) Local health clinic (%) Personal physician (%) Cardiologist (%) Not seek care (%) Don’t know/other (%) Care seeking time No. Immediately (%) Within a few hours (%) Within a day (%) Schedule appointment for the next few days (%) Don’t know/other (%)
Japan
1031 55.9 22.9 1.9 9 3.3 3.7 3.3
1433 32.9 16.2 10.1 23.5 6.8 6.8 3.7
993 83.1 8.5 2.7 3.4
1335 56.4 15.2 16 8.8
2.3
3.6
In this first study to explore international care-seeking behavior for chest pain, respondents in the US and Japan differed substantially in their responses to hypothetical chest pain scenarios. US respondents more readily recognized the emergent nature of the potential MI scenario. Respondents in Japan were much less likely to respond immediately or seek care in an emergent setting for the potential MI scenario. These findings may reflect differences in culture, health education, or health care systems.
Previous work .0001
Table III. Care-seeking response to potential myocardial infarction scenario United States
to seek care (P ⫽ .001). Compared with their Japanese counterparts, US survey subjects were more likely to seek care immediately (83.1% vs 56.4% in Japan).
P
.001
.001
(22.9% vs 16.2% in Japan) or through emergency medical services/911 (55.9% vs 32.9% in Japan). US respondents were less likely to seek care with their personal physician (9% vs 23.5% in Japan) or not seek any care (3.7% vs 6.8% in Japan). Similarly, there were clear differences in preferences in the 2 countries for time
Several studies have examined reasons for pre-hospital delay in patients with acute coronary syndromes. These studies have found that a variety of factors including race, sex, socioeconomic status, and individual patient attitudes may all substantially impact alacrity in seeking care.5–7,11–16 In their survey of 313 consecutive patients with MI in Great Britain, Leslie and colleagues found that the 2 most frequently cited reasons for delay in seeking medical help were “thinking that symptoms would go away” and “not thinking it was serious.”5 Other researchers have found that individual demographic factors impact care-seeking delay for acute chest pain. Sheifer and colleagues examined 102,339 elderly patients from the Cooperative Cardiovascular Project and found that diabetes mellitus, history of angina, female sex, black race, and poverty were all associated with late arrival (ⱖ6 hours after symptom onset).6 Similar findings were seen when Ell and colleagues surveyed 448 African-American patients admitted with acute chest pain in Los Angeles.7 In that study, public hospital patients and patients without health insurance had significantly longer care-seeking delays.7 In their study of care-seeking behavior, Weissman and colleagues surveyed 12,068 patients in 5 Massachusetts hospitals and found that black race, poverty, lack of insurance, and lack of a regular physician were all associated with pre-hospital delays in care.17 In the more recent Global Registry of Acute Coronary Events (GRACE; 1999 –2001), the 3693 patients with ST-segment elevation MI had a median pre-hospital delay of 3.0 hours (mean, 6.1 hours).18 Factors associated in GRACE with longer delays included age, prior heart failure, early morning symptom onset, and presentation in Europe or South America.18
American Heart Journal Volume 147, Number 4
Differences in time to presentation in the US and Japan for acute coronary syndromes There are limited data comparing pre-hospital delays for acute coronary syndrome patients in the US and Japan. In the US, the largest contemporary cohort was recently reported from the National Registry of Myocardial Infarction 3 (NRMI-3).19 For the 114,405 patients with confirmed MI, the mean time to presentation was 5.96 hours. In both the Worcester Heart Attack Study (1997 study year) and in the baseline phase of the Rapid Early Action for Coronary Treatment (REACT) multicenter trial (1995–1996), the median pre-hospital delay was 2.0 hours (mean delay, 4.3 hours and 5.0 hours, respectively).20,21 Data from comparable Japanese cohorts are not currently available. There are data from smaller series, including Takahashi and colleagues’ 106 patients who underwent primary angioplasty for acute MI.22 In this selected group, the mean ischemic time was 4.7 ⫾ 2.6 hours.22 Such times are similar to reperfusion intervals seen in the recent Stent PAMI multicenter trial.23 In 2 older series from Japan, the mean pre-hospital patient delay times have ranged from 52 minutes in Asahikawa City to almost 6 hours in Tokyo.24 –26 Because of the comparative paucity of recent data from Japan regarding pre-hospital delays, it is difficult to assess whether the attitudes toward chest pain seen in this survey result in measurable differences in pre-hospital delay for acute coronary syndrome patients in Japan.
Policy implications Because of the tremendous morbidity and mortality associated with delays in presentation for acute coronary syndromes, interventions to minimize these delays would have the potential to significantly impact patient outcomes. In particular, because of the numerous previous studies of US care seeking showing lack of insurance may contribute to pre-hospital delays, interventions to reduce or eliminate the number of uninsured patients may be attractive to policy makers. However, our findings suggest that such strategies would not be a panacea; patient-reported care-seeking delays for the potential MI scenario were much higher in Japanese respondents even though Japan’s national health care system actually has mandatory universal coverage.10 Interventions specifically targeted to improve access through increasing insurance coverage have proven disappointing.27 Public education efforts have also shown mixed effects on patient behavior.28 –30 In the REACT trial, an aggressive 18-month public education campaign encompassing community and patient groups, small media (direct mail, billboards, public service announcements), and mass media (television, newspapers, radio) had no appreciable benefit on time from symptom on-
Liao et al 633
set to hospital arrival for patients with chest pain.31 In the intervention and reference communities, median time to hospital arrival was 130 minutes and 126 minutes, respectively.31 These results contrast with those of media and educational campaigns undertaken in Europe, which have produced measurable decreases in delay times.29 One hypothesis for the failure of efforts such as REACT to improve delay times in the US relates to the short median delay times at baseline compared with the longer delays (3– 4 hours) in the successful European studies.29,31 It may be difficult to reduce delay times to ⬍2 hours with an educational campaign. Historically, Japanese delay times appear closer to the lengths reported in the various European studies.26,29 Consequently, more vigorous public educational efforts, such as that reported by Shibata and colleagues in Asahikawa City, might produce significant improvements in patient delays and resulting survival.25
Strengths and limitations This study has several important strengths. First, this is the first study to examine differences in care-seeking behavior in the US and Japan. Second, samples were drawn from nationally representative population samples in the 2 countries on the basis of each country’s national population census. Third, respondents were randomly selected for in-person interviews. Fourth, interviews were conducted by professional, trained Gallup Organization personnel. This study has several limitations. First, these data reflect survey responses to hypothetical chest pain scenarios. Although care was made to match the US and Japan surveys, it is possible that word choices or other elements of the scenarios may have influenced 1 group differently from the other. Second, this study compared national samples from the US and Japan. Such groups have inherently different demographic and cultural characteristics. In assessing care-seeking responses, we cannot separate the effects of these disparate factors from the influence of the different national health systems. Japanese respondents reported substantially less emergent health care coverage than US respondents, although the Japanese, by law, have universal health coverage. These results may reflect a substantial misconception by Japanese respondents of either the survey question or their legally mandated health insurance benefits. Third, these data were collected in January 1997. Although it is unlikely that care-seeking behaviors in the 2 countries have been radically altered since the survey was completed, it is possible that details such as the growth of the uninsured population in the US may subtly change the results of more contemporary examinations.
634 Liao et al
Conclusion Respondents in the US and Japan differed substantially in their responses to a hypothetical chest pain scenario. Whether these differences result from cultural or health care system factors and whether these apparent attitudes produce delays in real responses to acute coronary syndromes must be explored in further studies. If our results reflect genuine delays in Japanese patient responses, improved public educational efforts may be helpful in addressing these disparities.
References 1. Penny WJ. Patient delay in calling for help: the weakest link in the chain of survival? Heart 2001;85:121–2. 2. Rawles JM. Quantification of the benefit of earlier thrombolytic therapy: five-year results of the Grampian Region Early Anistreplase Trial (GREAT). J Am Coll Cardiol 1997;30:1181– 86. 3. The pre-hospital management of acute heart attacks: recommendations of a Task Force of the European Society of Cardiology and the European Resuscitation Council. Eur Heart J 1998;19:1140 – 64. 4. Leitch JW, Birbara T, Freedman B, et al. Factors influencing the time from onset of chest pain to arrival at hospital. Med J Aust 1989;150:6 –10. 5. Leslie WS, Urie A, Hooper J, et al. Delay in calling for help during myocardial infarction: reasons for the delay and subsequent pattern of accessing care. Heart 2000;84:137– 41. 6. 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 2000;102: 1651– 6. 7. Ell K, Haywood LJ, Sobel E, et al. Acute chest pain in African Americans: factors in the delay in seeking emergency care. Am J Public Health 1994;84:965–70. 8. Otaki J. Considering primary care in Japan. Acad Med 1998;73: 662– 8. 9. Iso H, Shimamoto T, Kitamura A, et al. Trends of cardiovascular risk factors and diseases in Japan: implications for primordial prevention. Prev Med 1999;29:S102–5. 10. Ikegami N. The economics of health care in Japan. Science 1992; 258:614 – 8. 11. Yarzebski J, Goldberg RJ, Gore JM, et al. Temporal trends and factors associated with extent of delay to hospital arrival in patients with acute myocardial infarction: the Worcester Heart Attack Study. Am Heart J 1994;128:255– 63. 12. Goldberg RJ, Gurwitz J, Yarzebski J, et al. Patient delay and receipt of thrombolytic therapy among patients with acute myocardial infarction from a community-wide perspective. Am J Cardiol 1992;70:421–5. 13. Moss AJ, Wynar B, Goldstein S. Delay in hospitalization during the acute coronary period. Am J Cardiol 1969;24:659 – 65. 14. Bett N, Aroney G, Thompson P. Delays preceding admission to hospital and treatment with thrombolytic agents of patients with possible heart attack. Aust N Z J Med 1993;23:312–3. 15. Schmidt SB, Borsch MA. The prehospital phase of acute myocardial infarction in the era of thrombolysis. Am J Cardiol 1990;65: 1411–5.
American Heart Journal April 2004
16. Cooper RS, Simmons B, Castaner A, et al. Survival rates and prehospital delay during myocardial infarction among black persons. Am J Cardiol 1986;57:208 –11. 17. Weissman JS, Stern R, Fielding SL, et al. Delayed access to health care: risk factors, reasons, and consequences. Ann Intern Med 1991;114:325–31. 18. Goldberg RJ, Steg PG, Sadiq I, et al. Extent of, and factors associated with, delay to hospital presentation in patients with acute coronary disease (the GRACE registry). Am J Cardiol 2002;89: 791– 6. 19. Mathew V, Gersh B, Barron H, et al. Inhospital outcome of acute myocardial infarction in patients with prior coronary artery bypass surgery. Am Heart J 2002;144:463–9. 20. Goldberg RJ, Yarzebski J, Lessard D, et al. Decade-long trends and factors associated with time to hospital presentation in patients with acute myocardial infarction: the Worcester Heart Attack Study. Arch Intern Med 2000;160:3217–23. 21. 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 1999;138:1046 –57. 22. Takahashi T, Anzai T, Yoshikawa T, et al. Effect of preinfarction angina pectoris on ST-segment resolution after primary coronary angioplasty for acute myocardial infarction. Am J Cardiol 2002; 90:465–9. 23. Brodie BR, Stone GW, Morice MC, et al. Importance of time to reperfusion on outcomes with primary coronary angioplasty for acute myocardial infarction (results from the Stent Primary Angioplasty in Myocardial Infarction Trial). Am J Cardiol 2001;88: 1085–90. 24. Shibata J, Tanazawa S, Hirasawa K, et al. Deaths in early phase of acute myocardial infarction and approaches for reducing the in-hospital and out-hospital case-fatality rates of the disease. Jpn Circ J 1987;51:325–31. 25. Shibata J, Hirasawa K, Tateda K. The importance of early treatment of myocardial infarction and approaches for shortening delays in securing medical care. Jpn Circ J 1984;48: 721– 8. 26. Takano T, Endo T, Tanaka K, et al. Current status of prehospital care of patients with acute myocardial infarction in Tokyo: analysis of 3-year experience with coronary care unit network. Jpn Circ J 1987;51:338 – 43. 27. Siemiatycki J, Richardson L, Pless IB. Equality in medical care under national health insurance in Montreal. N Engl J Med 1980; 303:10 –5. 28. Ho MT, Eisenberg MS, Litwin PE, et al. Delay between onset of chest pain and seeking medical care: the effect of public education. Ann Emerg Med 1989;18:727–31. 29. Blohm MB, Hartford M, Karlson BW, et al. An evaluation of the results of media and educational campaigns designed to shorten the time taken by patients with acute myocardial infarction to decide to go to hospital. Heart 1996;76:430 – 4. 30. Rowley JM, Hill JD, Hampton JR, et al. Early reporting of myocardial infarction: impact of an experiment in patient education. Br Med J 1982;284:1741– 6. 31. Luepker RV, Raczynski JM, Osganian S, et al. Effect of a community intervention on patient delay and emergency medical service use in acute coronary heart disease: the Rapid Early Action for Coronary Treatment (REACT) Trial. JAMA 2000;284: 60 –7.
American Heart Journal Volume 147, Number 4
Appendix Survey scenario “For the next few questions, I would like you to tell me what you would do if you were in the situation I am about to describe. For the first example, please imagine that while carrying groceries you suddenly feel a crushing pressure in the middle of your chest. The pressure feels almost as if someone was squeezing your chest. The pain is very bad and you are forced to sit down. You notice that you are out of breath and that you have begun to sweat, even though it is not hot outside. You are surprised by the pain, because it came on rather suddenly. The pain does not go away for 10
Liao et al 635
minutes. Which one of the following best describes what you would do if you were in this situation?”
Other survey questions “How quickly would you seek care if you were in this situation?” “How long would you estimate that it would take you to get to the emergency room or hospital closest to your home?” “If you needed emergency medical care, how much of the cost would you estimate your health insurance plan would cover? If you do not have health insurance, please tell me so.” “Do you currently have a doctor who you see on a regular basis?”