The decline in coronary heart disease: Determining the paternity of success
Article Outline
An old aphorism, commonly remembered because of its use by President John F. Kennedy after the Bay of Pigs disaster, is that success has many fathers, but failure is an orphan. The implications are twofold. First, everyone wants to take at least their share of the credit when things go well, but human nature is to try to diminish or avoid responsibility for disappointing results. Second, obfuscation has historically been aided by the fact that, at least until recently, paternity, unlike maternity, has been hard to prove.
In the case of coronary heart disease, the success is undeniable. Age-adjusted coronary heart disease mortality has declined progressively since at least the mid-1960s, and these declines have been noted in men and women and in blacks and whites, although declines among African Americans have lagged behind the declines in whites (1). Similar declines have been found in Western Europe, Australia, and New Zealand, whereas rates have risen in the former Soviet Union (2).
What has been more difficult is the assignment of the proportional credit, or paternity, of success in the absence of DNA testing. Furthermore, a nearly 40-year phenomenon may be the net result of different explanations at different times. For example, progressive reductions in case fatality rates from acute myocardial infarction may have been explained first by the ability of coronary care units to resuscitate patients from sudden death in the 1970s, then by thrombolysis, and subsequently by primary coronary angioplasty.
The three major potential explanations for the reduction in coronary heart disease mortality are the primary prevention of coronary heart disease in the general population, interventions that reduce the case fatality rate from acute events such as myocardial infarction or cardiac arrest, and secondary prevention to prolong life in patients with diagnosed coronary heart disease. The assignment of credit can be based on local or national data on incidence and prevalence, or it can be deduced by analyzing the utilization of interventions with known benefits.
In this issue of the The Green Journal, Arciero and colleagues use data from Olmsted County to estimate changes in the incidence of coronary heart disease over time (3). The advantage of their approach is the availability of a reasonably closed population that receives most of its medical care locally. A major relative disadvantage is the problem of generalizing more broadly from a small, homogenous local area. Nevertheless, Arciero et al provide ample data that the incidence of coronary heart disease is declining in Olmsted County.
However, their study also emphasizes a confounding issue, namely the changing definition of coronary heart disease. Good diagnostic criteria exist for acute myocardial infarction, but they have evolved over time as the advent of new technologies, initially creatine-kinase isoenzymes and more recently troponin levels, has markedly increased the sensitivity for diagnosing acute myocardial infarction, especially small myocardial infarctions. The recent decision to redefine myocardial infarction based on an elevated troponin level alone (4) will undoubtedly lead to an increase in the number of incident and recurrent myocardial infarctions. Furthermore, ample evidence documents that a substantial proportion of acute myocardial infarctions are not diagnosed at the time that they occur but are confirmed subsequently. Although these myocardial infarctions do not carry the same acute mortality and costs as diagnosed myocardial infarctions, some result in short-term sudden death, and others have long-term prognostic implications. The diagnosis of angina and even asymptomatic coronary heart disease will occur earlier as screening exercise tests, tomographic scanning for coronary calcium, nuclear cardiologic procedures, and coronary angiography are more commonly used. Although some of these patients would have been diagnosed later in the course of their disease 30 years ago, others would have died from a variety of other causes without ever being considered as having coronary disease. In Olmsted County, the incidence of coronary heart disease declined despite this presumed improvement in diagnostic sensitivity.
A major challenge is that national data do not correspond ideally to these epidemiologic categories. Substantial data are available on the number of patients hospitalized with acute myocardial infarction in the United States (5), but it is not simple to determine how many of these patients had prior evidence of coronary heart disease or even a prior myocardial infarction. Furthermore, many national estimates arbitrarily categorize prehospital deaths as acute myocardial infarctions—an approach that may or may not be technically accurate.
Using the deductive approach to estimating the likely effect of known changes in risk factors and case fatality rates, my colleagues and I have previously estimated that about 25% of the decline in coronary mortality is due to a decrease in incidence and nearly 75% is due to a reduction in deaths among patients with known coronary heart disease (6). Of the latter, about 15% may be ascribed to reductions in mortality from acute myocardial infarction, about 30% to specific medical and surgical treatments of myocardial ischemia, and about 50% to a reduction in risk factors (e.g., hyperlipidemia, smoking, and hypertension) in patients with known coronary disease. Estimates from England and Wales are reasonably similar (7).
To understand how reductions in the incidence of new-onset coronary heart disease and improvements in the outcome of patients with coronary disease effect mortality, it is critical to understand the interplay between incidence and prevalence. From an arithmetic perspective, prevalence is the product of incidence multiplied by average life expectancy. The direct relationship between life expectance and prevalence means that most interventions to reduce mortality from coronary heart disease also increase prevalence, including reduced case fatality rates for first and recurrent acute myocardial infarctions, risk factor reduction in patients with known coronary heart disease, and all medical and interventional treatments of ischemic heart disease and its complications, such as heart failure and arrhythmias. Prevalence will also be increased by any increases in the incidence of diagnosed coronary heart disease, regardless of whether incidence is actually increasing or whether more sensitive tests diagnose it earlier. The only way truly to reduce incidence is to change the risk factors, known and unknown, that cause the disease. Even then, prevalence could increase despite decreasing absolute incidence if the life expectancy of prevalent coronary heart disease patients is increasing sufficiently. In 2000, the estimated prevalence of coronary heart disease in the U.S. in persons 25 to 94 year of age was 11.2 million (8), with an estimated annual incidence of new coronary heart disease cases in the U.S. of about 1.3 million (9,10), so average life expectancy after diagnosis is about 8.6 years. Just as 6-month increase in average life expectancy among prevalent cases could offset a 9.5% absolute decrease in incidence and maintain a constant absolute prevalence. If prevalence increases, even a decrease in the rate of events and deaths among prevalent cases could be offset by the increase in the absolute number of prevalent cases, so that the number of deaths among prevalent cases of coronary heart disease would remain stable despite declining death rates.
In the article by Ergin et al in this issue of The Green Journal (11), follow-up data from people who were included in the National Health and Nutrition Examination Survey epidemiologic follow-up study were assessed at baseline and over time. When the initial 1971–1982 cohort was compared with a later (1982–1992) cohort, the latter had a 31% lower cardiovascular disease mortality rate, a 21% lower cardiovascular disease incidence rate, and a 28% lower 1-month case fatality rate. Of note was that the age-, race-, and gender-adjusted incidence of coronary heart disease declined by only 13%, and the adjusted rate of acute myocardial infarction actually increased insignificantly, whereas the rate of recurrent myocardial infarctions declined significantly in all but white women. These data in aggregate indicate that reductions in the incidence of coronary heart disease, presumably explained by reductions in risk factors, have contributed to lower coronary heart disease death rates, but that the lower death rates are also substantially related to improvements in outcomes of acute myocardial infarction and to the effects of secondary prevention.
With all these declines, why are our hospitals so full?
The success in reducing age-adjusted rates of coronary heart disease should not be equated to a reduction in the absolute number of myocardial infarctions or deaths. The population of the United States is not only aging, it is also increasing, especially as the baby boom generation enters the coronary heart disease age range. The result is a marked increase in the number of persons at risk for coronary disease. Reductions in competing causes of death, especially stroke and to a much lesser extent (on a relative but not absolute basis) noncardiovascular death, keep more people alive to develop coronary heart disease. The net result of all of these changes, as well as changes in diagnostic criteria, is that the number of hospitalized acute myocardial infarctions increased from about 430,000 to about 930,000 between 1980 and 2000 (5).
Unless improvements continue, the absolute number of Americans with coronary heart disease will continue to rise. We have come a long way, but much work remains—both the extension of current, known, beneficial interventions to the full range of eligible patients, as well as continuing advances in prevention and treatment.
References
- . Morbidity & Mortality (2002 chart book on cardiovascular, lung, and blood diseases) . Rockville, Maryland: U.S. Department of Health and Human Services National Institutes of Health; 2002;
- Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality (10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease) . Lancet . 1999;353:1547–1557
- Arciero TJ, Jacobsen SJ, Reeder GS, et al. Temporal trends in the incidence of coronary disease. Am J Med. 2004:117:228–233
- . Myocardial infarction redefined—a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition for Myocardial Infarction . J Am Coll Cardiol . 2000;36:959–969
- . http://www.cdc.gov/nchs/about/major/hdasd/nhds.htm Accessed June 5, 2004
- The recent decline in mortality from coronary heart disease, 1980-1990 (the effect of secular trends in risk factors and treatment) . JAMA . 1997;277:535–542
- . Explaining the decline in coronary heart disease mortality in England and Wales between 1981 and 2000 . Circulation . 2004;109:1101–1107
- . ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NHIS/2000 Accessed June 4, 2004.
- . Some risk factors related to the annual incidence of cardiovascular disease and death using pooled repeated biennial measurements (Framingham Study, 30-year follow-up) . In: Kannel WB , Wolf PA , Garrison RJ editor. The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease . Washington, D.C.: Government Printing Office: NIH Publication 87-2703, section 34.; 1987;
- U.S. Census Bureau. Census 2000. Available at: http://factfinder.census.gov/servlet/DTTable?ds_name=DEC_2000_SF1_U&geo_id=01000US&mt_name=DEC_2000_SF1_U_PCT012. U.S. Census Bureau. Accessed June 28, 2003.
- Ergin A, Muntner P, Sherwin R. Secular trends in cardiovascular disease mortality, incidence, and case fatality rates in U.S. adults. Am J Med. 2004:117:219–227
PII: S0002-9343(04)00368-7
doi:10.1016/j.amjmed.2004.06.003
© 2004 Elsevier Inc. All rights reserved.
