Correspondence

Rates of Death from Coronary Heart Disease

N Engl J Med 1999; 340:730-732March 4, 1999

Article

To the Editor:

Levy and Thom (Sept. 24 issue),1 in their editorial on rates of death from coronary heart disease (CHD), note an apparent paradox in the declining prevalence of causal risk factors for myocardial infarction and the lack of change or slight increase in the incidence of that disease process from 1987 to 1994, as reported by Rosamond et al. in the same issue.2 Levy and Thom seek to explain this paradox by looking at the conventional risk factors. I believe this view is too restricted and that, instead of a paradox, there is a failure to look for and thus to see what might otherwise be obvious.

Socioeconomic factors appear to have a substantial role in the incidence of many diseases, including atherosclerotic disease. A landmark study found up to a fourfold difference in mortality from coronary artery disease over a period of 7.5 years in civil servants in London, with the difference progressively increasing in categories of jobs considered to reflect declining socioeconomic status; the combined effects of smoking, cholesterol levels, and blood pressure accounted for distinctly less than half the differences noted.3 Moreover, the study population was one in which the deleterious effects of poor access to care were minimized by health insurance coverage for all, and all the subjects were employed — circumstances that are clearly more favorable than those that can be expected in broader population samples in the United States.

We know that in the United States the inequality in income between persons in the highest income bracket and those in the lowest is increasing 4 and that higher mortality rates, including the rate of death from CHD, are correlated with the degree of income inequality in each state.5 A similar association is seen in metropolitan areas.6 The applicability of these findings to the Atherosclerosis Risk in Communities (ARIC) study reported by Rosamond et al. is certainly arguable, since no data were presented to allow consideration of a possible association between socioeconomic status and the incidence or outcome of myocardial infarction. However, in seeking to account for the “paradox,” Levy and Thom fail even to acknowledge possible factors beyond the limited vision of biomedical science.

Daniel L. Gornel, M.D., M.P.H.
1480 Rexford Dr., Los Angeles, CA 90035

6 References

1
Levy D, Thom TJ. Death rates from coronary disease -- progress and a puzzling paradox. N Engl J Med 1998;339:915-917
Full Text | Web of Science | Medline

2
Rosamond WD, Chambless LE, Folsom AR, et al. Trends in the incidence of myocardial infarction and in mortality due to coronary heart disease, 1987 to 1994. N Engl J Med 1998;339:861-867
Full Text | Web of Science | Medline

3
Marmot MG, Rose G, Shipley M, Hamilton PJ. Employment grade and coronary heart disease in British civil servants. J Epidemiol Community Health 1978;32:244-249
CrossRef | Web of Science | Medline

4
Weinberg DH. A brief look at postwar U.S. income inequality. Washington, D.C.: Bureau of the Census, 1996. (Or see: http://www.census.gov/pub/hhes/income/incineq/p60tb4.html.)

5
Kennedy BP, Kawachi I, Prothrow-Stith D. Income distribution and mortality: cross sectional ecological study of the Robin Hood index in the United States. BMJ 1996;312:1004-1007[Erratum, BMJ 1996;312:1194.]
CrossRef | Web of Science | Medline

6
Lynch JW, Kaplan GA, Pamuk ER, et al. Income inequality and mortality in metropolitan areas of the United States. Am J Public Health 1998;88:1074-1080
CrossRef | Web of Science | Medline

To the Editor:

In their interesting editorial, Levy and Thom note the “puzzling paradox . . . that no decline in the incidence of myocardial infarction was observed in the ARIC sample during a period when the prevalence of causal risk factors was reduced.” When facts do not fit theories, theories have to change. Surely the evidence suggests the existence of other risk factors. As medical generalists,1 we expect that these factors are in the psychosocial dimensions of medicine. Everson et al. have reported “hopelessness as an independent predictor of cardiovascular disease morbidity and mortality in both American and Finnish populations.”2

Denis Pereira Gray, F.R.C.G.P.
Russell Steele, F.R.C.G.P.
Kieran Sweeney, M.R.C.G.P.
University of Exeter, Exeter EX2 5DW, United Kingdom

2 References

1
Gray DP, Steele R, Sweeney K, Evans P. Generalists in medicine. BMJ 1994;308:486-487
CrossRef | Web of Science

2
Everson SA, Kaplan GA, Goldberg DE, Salonen R, Salonen JT. Hopelessness and 4-year progression of carotid atherosclerosis. Arterioscler Thromb Vasc Biol 1997;17:1490-1495
CrossRef | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: The opinion of Levy and Thom that the flat trend in the incidence of myocardial infarction between 1987 and 1994 that we reported is a “puzzling paradox” deserves careful consideration. Dr. Gornel suggests that any discourse on stalled trends in incidence must include the consideration of socioeconomic factors. Both views underscore the complex nature of evolving patterns in the occurrence of disease and the methods used to measure them. Three points are relevant to the debate.

First, we submit that the flat trends in the incidence of myocardial infarction are real. We found no evidence of a statistically significant decline in any of the race–sex groups examined. As Levy and Thom theorized, it is possible that the stable incidence rates of hospitalization for myocardial infarction could be explained in part by a shift over time in the number of patients still alive when they reach the hospital and the number who die before arriving at the hospital. In fact, there was a statistically significant average annual decline in out-of-hospital deaths from CHD among men without a history of myocardial infarction (–3.9 percent; 95 percent confidence interval, –6.9 to –0.8). However, there was essentially no change in deaths among women (–0.3 percent; 95 percent confidence interval, –5.1 to 4.7). Community surveillance of incidence rates of myocardial infarction is an elaborate process subject to methodologic imperfections. Continued surveillance of the ARIC communities, as well as studies by others, is needed to clarify our findings.

Second, quantifying the component causes of contemporary trends in the incidence of myocardial infarction requires a wide perspective. Although the causal relation of elevated blood pressure, cigarette smoking, hyperlipidemia, and physical inactivity to heart disease is beyond doubt, the interactions of these factors with genetic, social, and other biologic factors are not fully understood. Individual as well as community-level socioeconomic factors play a part in the occurrence of disease and have their own evolving trends. A broader perspective, as Dr. Gornel suggests, can provide insights into what at first appears to be a paradox. Unfortunately, data on socioeconomic status were not available in our study, nor did we have enough communities to consider ecologic associations between trends in CHD and income inequality.

Third, the final measure of the effect of any set of risk factors, biologic or social, is the population-based trend in disease rates. Tremendous progress has been made in reducing mortality from CHD over the past 30 years.1 Our data suggest that headway in preventing the occurrence of a first myocardial infarction may have slowed in this decade. An active debate on the comparative merits of primary and secondary prevention, as well as biologic versus social risk factors, is essential to our understanding of this phenomenon.

Wayne D. Rosamond, Ph.D.
Lloyd E. Chambless, Ph.D.
University of North Carolina, Chapel Hill, NC 27514

Aaron R. Folsom, M.D.
University of Minnesota, Minneapolis, MN 55454

1 References

1
Gillum RF. Trends in acute myocardial infarction and coronary heart disease death in the United States. J Am Coll Cardiol 1994;23:1273-1277
CrossRef | Web of Science | Medline

Author/Editor Response

In our editorial, we linked nationally observed declines in mortality from CHD to declines in the prevalence of three key risk factors — cigarette smoking, high blood pressure, and high cholesterol levels — and we suggested that failure to observe a parallel fall in the incidence of myocardial infarction in the ARIC study is a paradox. We highlighted these three risk factors for CHD both because their causal roles have been proved unequivocally and because there is overwhelming evidence that their modification can reduce the risk of CHD.1

We agree with Dr. Gornel and with Gray et al. that other risk factors contribute to the incidence of CHD and to its outcome; the growing list of risk factors includes socioeconomic and psychosocial factors. At present, however, evidence establishing socioeconomic and psychosocial factors as causal contributors to the risk of CHD and data on the effect of their modification on the outcome of CHD are lacking.1

National data show that mortality from CHD has declined by more than 50 percent over the past 30 years, and this decline has been observed in all demographic groups.2 A potential role of socioeconomic factors in determining the risk of CHD is suggested by the fact that the declines in mortality from CHD have been slower in blacks than in whites and in persons with lower socioeconomic status than in those with higher socioeconomic status.3 Similarly, Rosamond et al. report that between 1987 and 1994 there were smaller declines in mortality and less favorable case-fatality trends in blacks than in whites in the ARIC study sample. Most of the blacks in the study were from Jackson, Mississippi, the study site with the lowest income level. Furthermore, these trends in CHD coincide with increases in the number of uninsured persons in this country and with the widening inequality of income that can result in disparities in access to medical care.

Although we do not reject the notion that changes in socioeconomic status or increased hopelessness in American society has an appreciable effect on trends in CHD, such relations are not well supported by the available literature on the subject. Furthermore, it is unclear how changes in socioeconomic or psychosocial factors could explain the divergent trends in the incidence of and mortality from CHD reported by Rosamond et al.

We must maintain a broad vision and continue our quest to identify additional risk factors for CHD. This is a matter of great importance for public health, especially if such information results in new approaches to the prevention of heart disease, the leading cause of death in the United States.

Daniel Levy, M.D.
National Heart, Lung, and Blood Institute, Framingham, MA 01702

Thomas J. Thom
National Heart, Lung, and Blood Institute, Bethesda, MD 20892

3 References

1
Pasternak RC, Grundy SM, Levy D, Thompson PD. Spectrum of risk factors for coronary heart disease. J Am Coll Cardiol 1996;27:978-990
CrossRef | Web of Science | Medline

2
National Heart, Lung, and Blood Institute. Morbidity and mortality: 1996 chartbook on cardiovascular, lung, and blood diseases. Bethesda, Md.: National Institutes of Health, 1996.

3
Tyroler HA, Wing S, Knowles MG. Increasing inequality in coronary heart disease mortality in relation to educational achievement: profile of places of residence, United States, 1962 to 1987. Ann Epidemiol 1993;3:Suppl:S51-S54