Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 1996;93:1755-1764

This Article
Right arrow Extract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Epstein, F. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Epstein, F. H.

(Circulation. 1996;93:1755-1764.)
© 1996 American Heart Association, Inc.

Articles

Cardiovascular Disease Epidemiology

A Journey From the Past Into the Future

Frederick H. Epstein, MD, FRCP

From the Institute of Social and Preventive Medicine, University of Zurich, Switzerland.

Correspondence to Dr Henry Blackburn, Division of Epidemiology, School of Public Health, University of Minnesota, 1300 S 2nd St, Suite 300, Minneapolis, MN 55454.


*    Setting the Stage
up arrowTop
 *Setting the Stage
down arrowBrief Historical Overview
 down arrowGeographical Pathology and...
 down arrowFrom Individual to Community...
 down arrowIntervention Studies
 down arrowToward Coronary Heart Disease...
 down arrowToward Chronic Disease...
 down arrowThe Journey Ahead
 down arrowReferences
 
When Ancel Keys established the Laboratory of Physiological Hygiene at the University of Minnesota in 1946, cardiovascular disease (CVD) epidemiology did not exist. It came into being through the vision and efforts of a small group. Collaboration among the early workers developed, and by the end of the 1940s, this new branch of science was fully under way. The gate was opened by the launching of the first prospective studies, led by the Minnesota Business Men Study,1 and was soon followed by the Framingham Study.2 Keys's contribution was decisive from the start.3

CVD epidemiology, seen as one of the major deliberate scientific approaches to atherosclerosis research, almost seemed to come from nowhere, like a nova or the "big bang." In the late 1940s, there were forerunners, but, as always with the birth of creative ideas, there will remain some mystery around the reasons for this explosive development; it led within 30 years to coronary heart disease (CHD) becoming a largely preventable condition. The extent of this dramatic change is mirrored by the fact that the word "prevention" was not even mentioned in the report of the first symposium on CVD epidemiology, organized by Ancel Keys and Paul D. White on the occasion of the World Congress of Cardiology in Washington in 1954.4 It would seem that they saw no point in paying lip service to the self-evident target of prevention before the scientific foundations for preventive action had been laid.

In those years immediately following the end of the Second World War, there was a pervasive spirit of discovery, as if to regain the time lost for scientific work during the war years. An example of how fast things moved in those days is provided by the Cooperative Study on Lipoproteins,5 a prospective study of almost 5000 men to test whether specific lipoproteins predict CHD better than total cholesterol. This study was based on Gofman's investigations of low-density lipoprotein in rabbits, published in 19506 ; already in the same year the National Advisory Heart Council took the initiative to launch the cooperative study in humans. This example is typical of the excitement on many fronts of atherosclerosis research in those early days.

Many of the founding fathers of research in CVD epidemiology and public health were not trained in those fields but came from clinical investigation and medicine. The distinguished cardiovascular physiologist Alfred E. Cohn, of the Rockefeller Institute, with Claire Ling of the New York Heart Association, wrote a visionary book, The Burden of Diseases in the United States, which recognized the coming threat of chronic diseases.7 Ernst Boas, a leading clinical cardiologist, with David Adlersberg, published a series of papers on hypercholesterolemia in families and its relation to CHD,8 which was the only reference cited for justifying the measurement of cholesterol in Framingham.2 Ancel Keys himself is a physiologist by training and his contributions are by no means confined to epidemiology.9 Epidemiology being an interdisciplinary science, it is appropriate and desirable that it should include workers from other fields. Indeed, for the results of epidemiological research to be translated into preventive action, the involvement of many other disciplines is an essential prerequisite.


*    Brief Historical Overview
up arrowTop
 up arrowSetting the Stage
 *Brief Historical Overview
down arrowGeographical Pathology and...
 down arrowFrom Individual to Community...
 down arrowIntervention Studies
 down arrowToward Coronary Heart Disease...
 down arrowToward Chronic Disease...
 down arrowThe Journey Ahead
 down arrowReferences
 
The conscious and deliberate effort to bring epidemiological approaches and methods to bear on the causes of atherosclerosis and CHD dates to the late 1940s. The early beginnings, though not called epidemiology, reach into the last century. A treatise on geographical pathology by Professor August Hirsch of Berlin reviews all that was known at the time about the distribution of diseases around the world, stating that reliable data on heart disease were practically nonexistent.10 In the 1930s, Wilhelm Raab, then in Vienna, related geographical differences in CHD frequency to dietary differences, in particular dietary fat.11 The early data available on the relationship between atherosclerosis and diet were reviewed by Rosenthal.12 When the Second World War ended in the mid-1940s, the new era of research started almost explosively.

CVD epidemiology covers a wide range of scientific activities. This account is limited to mainstream research, with recognition that full understanding of the progress depends on a broad knowledge of all the contributions. There are two main parallel approaches: (1) comparison of the frequency of CVD in different populations and relating those differences to predisposing factors and (2) prospective cohort studies in individuals, aiming at identifying predisposing factors along a range from high to low disease risk.

The first approach will be covered under the heading of geographical epidemiology. The second, the cohort approach, leads from individual risk to the distribution of risk on the community level. These carry over to intervention studies because a point is reached in the course of observational studies when it becomes imperative to test whether predisposing factors not only predict clinical disease but whether their reduction causes a decline in the risk of disease. The time sequence can be depicted in a diagram (see Fig 1Down).



View larger version (10K):
[in this window]
[in a new window]
  		Figure 1. Cardiovascular disease epidemiology historical development.

Following the period of dawn and the founder years, the observational studies gained momentum until a stage was reached in the late 1960s when the evidence for a causal relationship between risk factors and disease risk became so strong that preventive trials had to be initiated. A second wave of observational studies is now under way in search of new risk factors to enhance the preventive potential inherent in the established risk factors.


*    Geographical Pathology and Epidemiology
up arrowTop
 up arrowSetting the Stage
 up arrowBrief Historical Overview
 *Geographical Pathology and...
down arrowFrom Individual to Community...
 down arrowIntervention Studies
 down arrowToward Coronary Heart Disease...
 down arrowToward Chronic Disease...
 down arrowThe Journey Ahead
 down arrowReferences
 
The investigation of geographical differences has always been a mainstay of epidemiological research.10 A conference of the International Society of Geographical Pathology on the subject of cardiovascular diseases was held in the Netherlands in 1934. The proceedings contain information on the frequency of atherosclerotic lesions in a number of countries. Included is a remarkable report by Anitschkow, famous for his groundbreaking work on experimental atherosclerosis in cholesterol-fed rabbits, about geographical differences in the clinical frequency of atherosclerotic diseases by social class and occupation. This provides a striking early illustration that epidemiological interests need not be confined to epidemiologists.13 It was Keys, however, who opened the new era by conducting a series of pilot field studies in the early 1950s ranging from Italy, Spain, England, and Scandinavia to South Africa and Japan.3 This culminated in the systematic Seven Countries Study, which started in 1957.14

The first written record of Keys's views on the relationship between serum cholesterol levels and geographical differences in the frequency of CHD is based on a lecture presented to a very small audience at Mount Sinai Hospital in New York in 1953.15 This was a momentous occasion because this cholesterol-disease relationship has become one of the cornerstones of CHD epidemiology and prevention. It is the mark of genius to make complicated things look simple and not to let complexity obscure the truth. Complexity was introduced in 1957 by Yerushalmi and Hilleboe in a much-discussed paper,16 indicating that the addition of more countries resulted in a good deal more scatter than suggested by Keys's chart, which showed a slightly curvilinear relationship between CHD mortality and the proportion of dietary calories provided by total fat. The emphasis at that time was on total dietary fat. In the ensuing years, it became increasingly apparent that different fats varied in their effect on serum cholesterol levels. These developments led Keys to carry out a series of metabolic ward experiments resulting in a formula named after him17 ; it indicates that the serum cholesterol–raising effect of saturated fatty acids is twice as potent as the lowering effect of polyunsaturated fatty acids.

In 1970, a report from the Seven Countries Study had a decisive impact on CHD prevention.18 It showed that CHD risk is strongly related to both serum cholesterol levels and the proportion of dietary calories derived from saturated fatty acids, measured previously in health, that diet saturated fat content leads to CHD, and that the relationship is mediated by serum cholesterol.18 The Seven Countries Study is an epidemiological investigation of cohorts in different geographical areas, and being prospective, guards against the "ecological fallacy," which is a potential disadvantage of cross-sectional, cross-cultural studies.19

There is much collateral evidence that CHD mortality and dietary fat are correlated, including data based on the CHD mortality statistics of the WHO and dietary data from the Food and Agricultural Organization of the United Nations. Correlations between these two sources of data in 20 countries are positive and significant, with coefficients of the order of .7 for all different types of fat except for a negative correlation with polyunsaturated fat.20 The same applies to the lipid score derived from Keys's formula, independent of fiber intake.21 The analyses assume a time lag of about 15 years between the dietary measurements and the mortality data, taking account of the time it presumably takes for dietary habits to affect CHD mortality. A recent report, based on similar data for 40 countries, shows a correlation of .78 between CHD mortality and an index for the intake of saturated fatty acids and dietary cholesterol.22

The cross-cultural data on the association between CHD risk and serum cholesterol levels are more equivocal than those for dietary fat consumption. Apart from the evidence of the Seven Countries Study, there are two other reports of a high correlation between CHD risk and the level of serum cholesterol in different countries.23 24 On the other hand, particularly within Europe, and looking at countries individually rather than "across the board," there are a number of countries with similar serum cholesterol levels that differ markedly in CHD mortality.25 Recent data from the WHO MONICA project (Fig 2Down) illustrate this point. A bird's-eye view of the charts, drawn from published data,26 27 hints of a curvilinear relationship, but the scatter is large. There is no real contradiction, however, between this picture and the message conveyed by the Seven Countries Study that serum cholesterol elevation represents the mechanism by which excess population intake of saturated fatty acids and cholesterol leads to atherosclerosis and CHD. The overall prevalence of coronary atherosclerosis in a country is largely determined by its ambient average serum cholesterol level. This sets the stage, while the observed frequency of atherosclerotic lesions and attendant clinical disease is the consequence of the countrywide constellation of other enhancing as well as protective risk factors.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 2. Total age-standardized cholesterol and ischemic heart disease mortality (per 100 000) in 30 MONICA centers: Men aged 35 to 64 years.

Looking at an individual country is like looking at an individual patient whose total risk depends on the balance of multiple influences. Clinical observations have been the fountainhead of many epidemiological discoveries. Thus, "clinical looks" at individual countries could yield etiological clues that might escape on regarding only the larger picture. In this light, the "French paradox" or the "Finnish paradox"22 assumes a new meaning. Any departure from the regression line of the relationship is a paradox to some extent. Interest and concern should include all countries along the line, covering both the outliers and those conforming to the rule. Such a research strategy is likely to be optimally effective in finding additional causes of geographical differences and, in this way, improving the potential for prevention at the level of public health. Meanwhile, the views of Keys have stood the test of time and they provide a safe harbor for journeys toward new territories. In this respect, geographical differences within the same country, with a few notable exceptions, have been neglected, and their investigation is likely to be rewarding.

Thus, the explanation of geographical differences requires that the entire constellation of risk factors be taken into account. Adding blood pressure and smoking to serum cholesterol, on the basis of data from the WHO MONICA project covering 35 populations, leaves more than three quarters of the variance in CHD mortality unexplained.28 In the Seven Countries Study, adding body mass index to the other three risk factors made a contribution toward explaining geographical variation among the 16 cohorts within the seven countries.29 Thus, investigations into the causes of geographical differences must look beyond the major risk factors that have been so successful in predicting individual risk. This is a priority for new research.

While the three major risk factors account only in part for geographic differences on cross-sectional study, it is conceivable that their changes over time could explain secular trends in CHD mortality. This is apparently the case for the United States and probably other countries in which there have been public campaigns to foster prevention and where the decline of CHD mortality has been paralleled by favorable changes in lifestyles and risk factors.30 The most recent and striking example comes from Finland, indicating that the observed decline in CHD mortality during the past years is almost entirely predicted by the changes in serum cholesterol, blood pressure, and smoking both in women and men.31 The central importance of nutrition is highlighted by the observation that there is a high correlation (r=.80, P=.0001) between the slope of the Keys score (which measures the intake of dietary fatty acids and cholesterol and reflects changes in serum cholesterol) between 1961 and 1988 and the slope of CHD mortality 10 years later in 26 countries (Stamler J, Ruth KJ, and Epstein FH, unpublished data). Furthermore, improvements in medical care have also had an impact on secular trends.32 All these findings suggest that the changes in mortality from CHD are largely due to factors that are known and can be controlled, notwithstanding gaps in understanding the reasons for variations in CHD mortality in different countries.

The quintessence of all geographical epidemiology is that dietary fat and cholesterol intake, the main determinants of population levels for serum cholesterol and low-density lipoprotein, play a fundamental role in explaining cross-cultural differences in the frequency of CHD. Serum lipids, in accordance with Keys's original views, set the stage for geographical variations, while the actual levels of mortality are determined by the interplay with additional factors. Among the latter, hemostatic function, also influenced by dietary fatty acids, is likely involved. The antioxidant content of foods must be considered, too.33 These are important new areas for research, aimed at understanding as completely as possible the causes of geographical differences. This knowledge will further enhance the potential of prevention. Geographical epidemiology has already provided one of the cornerstones of the evidence for the preventability of CHD. In addition, it has shown in the most direct way possible how high-risk countries can become low-risk countries and how low-risk countries can protect themselves against adopting lifestyles that turn them into high-risk countries.


*    From Individual to Community Risk
up arrowTop
 up arrowSetting the Stage
 up arrowBrief Historical Overview
 up arrowGeographical Pathology and...
 *From Individual to Community...
down arrowIntervention Studies
 down arrowToward Coronary Heart Disease...
 down arrowToward Chronic Disease...
 down arrowThe Journey Ahead
 down arrowReferences
 
Progress in understanding geographical differences has hinged on the knowledge of risk factors discovered in the course of prospective studies of CHD. These have been parallel developments. The Seven Countries Study, being both cross-cultural and prospective,14 provides a transition. It is not generally recognized how many prospective studies have been carried out all over the world (see TableDown). They would be a source for geographical comparisons of CHD incidence, but differences in methods make this impossible. The MONICA project allows such comparisons34 but not in terms of incidence, since it does not follow individuals and is not a cohort study.


View this table:
[in this window]
[in a new window]
  		Table 1. Prospective Studies of Coronary Heart Disease Epidemiology: Past and Current

It is now considered obvious that a prerequisite of prevention is the identification of predisposed persons by their risk factors before they develop manifest disease. That knowledge required prospective cohort studies. Surprisingly, the need was unrecognized until Keys made this revolutionary discovery in the late 1940s.35 From then on, descriptive cohort studies, measuring CHD incidence, were on their way (Fig 2Up). Surprisingly as well, the three major risk factors serum cholesterol, blood pressure, and smoking had been identified as early as 1956,36 and all further work only made this knowledge broader, deeper, and more secure. With time, more risk factors were found.

During the first two decades of the new era of CVD epidemiology and prevention, the main concern was detection of high-risk persons in the population and designing effective screening programs. The term "risk factor" appeared for the first time in 1961, in a publication from the Framingham Study.37 This "high-risk" strategy was the logical consequence of the success, based on the results of prospective studies, in detecting "carriers" of high risk. Tests of high sensitivity, specificity, and predictive power were sought, attempting to concentrate within the smallest possible proportion of the population the maximal number of individuals destined to develop CHD.38 39 Toward the end of the 1960s, realization grew that prevention and control programs could not rely solely on the detection and protection of the most susceptible persons in the community but must be rooted in the whole community and involve the entire population. This development carried over into intervention studies (Fig 1Up). A new focus on the community approach was also suggested by the realization that most of the CHD events in the population do not occur among those at high risk but rather among those at moderate risk, indicating that the high-risk strategy alone would miss half or more of the future victims. Thus, the entire population must adopt preventive measures.40 This concept was proposed in 1972 but was largely overlooked, having been part of a broad review of current issues in CHD epidemiology, until Rose,41 in 1981, published his paper on "The Strategy of Prevention." In it, he defined the "high-risk strategy" and the "mass strategy," later renamed "population strategy." The idea had an immediate impact and, along with a subsequent formal elaboration of the theme,42 has remained valid ever since.

One of the purposes of cohort studies is to test new risk factors and to assess their relative importance. The identification of new risk factors may improve predictive power in the individual.39 It should also improve what has been called "preventive potential."39 Implicit in the risk factor concept is that knowledge of more factors causally related to the disease will increase the ability to prevent the disease by their modification. Lately, much attention is given to "new risk factors."43 44 However, some questions on screening have not been adequately addressed. For example, it would be impossible to screen everybody for every well-established risk factor. Some system of staging needs to be devised, but little thought has been given to the problem. For the population strategy of prevention, it would be desirable in theory to change environment and lifestyles in such a way that the prevalence of all risk factors is reduced. However, what would be the threshold of prevalence of a risk factor or its attributable risk for the whole population below which intervention would not be warranted? Who should be screened for which genetic risk factors to take care of coexistent risk factors that can be modified? These questions should be addressed, though some of the information required to answer them is not yet available. It should be forthcoming from the new wave of prospective studies (Fig 1Up).


*    Intervention Studies
up arrowTop
 up arrowSetting the Stage
 up arrowBrief Historical Overview
 up arrowGeographical Pathology and...
 up arrowFrom Individual to Community...
 *Intervention Studies
down arrowToward Coronary Heart Disease...
 down arrowToward Chronic Disease...
 down arrowThe Journey Ahead
 down arrowReferences
 
The need for systematic intervention studies arose in the late 1960s when the evidence for sequential relationships between risk factor elevations and subsequent CHD risk became so strong that it had to be tested whether risk factor reduction would, in fact, reduce disease risk. Seen as a whole, the many trials conducted yielded positive results.45 46 Where the results fell short of expectation, the negative outcome could be ascribed to insufficient statistical power to demonstrate a difference between the experimental and control groups. In fact, the trial results have been of the order anticipated from the risk factor reductions actually achieved.47 Trials in which CHD mortality was reduced frequently do not show a reduction in total mortality, suggesting a compensatory increase in non–cardiovascular disease mortality. This conclusion and concern are not warranted because to date, the trials have not had the power to assess the effect of intervention on total mortality.48 In the most recently published trial, however, both coronary and total mortality were reduced.49

Participants in intervention studies have been high-risk persons because the conduct and cost of mass trials among persons covering the whole range from high to low risk would be prohibitive.50 In view of the graded relationship between risk factors and disease risk, however, it is not unreasonable to extrapolate trial findings on high-risk persons to those at intermediate or even low risk. The intermediate range of risk is of vital importance, since half or more of the CHD events in the population occur in this range. In the absence of randomized, controlled trials in that range, community studies that include the whole population in intervention programs should fill part of this gap. Community studies have shown so far that favorable changes in lifestyles and risk factors can be achieved, but they fall short of the desirable targets. Nevertheless, taking serum cholesterol as an example, striking reductions have taken place in countries as a whole, such as the United States51 or Finland.52 Indeed, the main problem in demonstrating success of intervention programs in community studies is that changes in the same direction take place in the comparison communities, probably because the many health messages on a national level are effective. It is too early to tell whether systematic and organized prevention programs in intervention communities lead to a greater disease risk reduction than that observed in control communities. Meanwhile, it is gratifying that this process also takes place in the latter communities and in countries as a whole. The decline in mortality from CHD that was first noted in the United States in the 1960s has now extended, with some exceptions, to most countries reporting mortality statistics.53


*    Toward Coronary Heart Disease Prevention
up arrowTop
 up arrowSetting the Stage
 up arrowBrief Historical Overview
 up arrowGeographical Pathology and...
 up arrowFrom Individual to Community...
 up arrowIntervention Studies
 *Toward Coronary Heart Disease...
down arrowToward Chronic Disease...
 down arrowThe Journey Ahead
 down arrowReferences
 
The worldwide tendency for CHD mortality to decline might suggest that "tomorrow is already here." While this trend is encouraging and, in countries where it could be analyzed, seems to be the result of favorable changes in lifestyles, it is less certain whether there has been a concomitant decline in CHD incidence. In any case, there are no grounds for complacency because even in countries such as the United States, prevention has a long way to go. Intervention studies and collateral evidence have left no reasonable doubt that CHD is preventable. The problem lies in translating this knowledge into action. On the whole, the forces supporting action have been gaining over those slowing or obstructing it, but there still exists strong resistance, ranging from frank opposition to disinterest, especially regarding cholesterol. The most serious among an array of claims is that low or lowering serum cholesterol leads to an increase in noncardiovascular disease, in particular cancer risk. According to the best evidence currently available, the excess cancer risk, if it exists, is small and confined to persons with very low cholesterol levels that are rare or will rarely be reached with preventive treatment; the excessive risk said to be related to noncancer and noncardiovascular diseases, including traumatic deaths, is probably entirely due to confounding by preexisting disease, detrimental habits such as smoking, or excessive drinking and influences related to social class.54 55

Another argument comes from those who favor reduction of serum cholesterol for high-risk persons or survivors of myocardial infarction but state that there is no evidence from controlled trials that those at moderate or low risk would benefit from lowering their levels. They do, in fact, oppose the population strategy of prevention, disregarding the fact that at least half or more of the heart attacks in the population, including sudden deaths, occur below the threshold of "high risk." These persons would, therefore, suffer a heart attack or die as a result of it without ever having had the chance to protect themselves against exposure to suboptimal levels of cholesterol. There is every reason to think that lowering serum cholesterol is protective for those with only moderately elevated or even low levels. Serum cholesterol, like blood pressure, is a graded characteristic: there is no point along the curve relating disease risk to risk factor level where risk stops and immunity begins. The risk declines but never disappears, just as the atherosclerotic lesions themselves cover a continuous spectrum from persons with severe to those with mild involvement. Populations with high average serum cholesterol levels are "sick populations"42 in terms of serum cholesterol. Their cure consists not of treating only those with elevated levels but shifting the entire distribution curve to the left, which will at the same time reduce the prevalence of high-risk people. Most important, the population strategy ensures that children adopt healthy eating habits early in life, slowing the rise of serum cholesterol with age, creating a new generation with a lower average and less-frequent high levels. Thus, the high-risk and population strategies are complementary and reenforce each other. Those supporting the high-risk strategy alone are seemingly willing to look on while serum cholesterol reaches high-risk levels, alongside the development of excessive atherosclerotic lesions, and then to institute preventive treatment. This is surely a caricature of preventive medicine, quite apart from the fact that a large part of the population at risk remains unprotected.

The population strategy is essential not only in regard to the dynamics of pathogenetic mechanisms of atherosclerosis and CHD but also because of its key role in health promotion. It is very difficult for an individual to garner strength for changing lifestyle toward the maintenance of health, for example, in a climate where smoking is an accepted social behavior. It is not easy to swim against the stream and stop the habit. A climate favorable to the concept of prevention is a prerequisite for the social support that people need to develop living habits to promote health. A community is more than the sum of its inhabitants. It represents a complex social system of its own that includes a community spirit. The creation of a spirit friendly toward prevention is important not only for the success of the population strategy. A high-risk individual will be helped in accepting and adhering to medical advice if there is social support from the immediate surroundings and from the climate pervasive in the community and country. The family is the smallest epidemiological unit in which transactions take place between human beings with their genetic predispositions and the environment. The community is an extension of this unit into a larger universe.

The effective pursuit of preventive cardiology hinges on close coordination of the high-risk and the population strategies. Great strides have been made in establishing the high-risk strategy as an integral part of medical practice as evidenced by commitments and recommendations made by responsible national and international bodies.56 57 This does not mean that the recommendations are being turned into action as forcefully as they should be, but a solid basis has been established. The broader population strategy, on the other hand, is less strongly promoted and supported. Unlike the high-risk strategy, it does not have full backing by the medical profession, the first and most immediate concern of practicing physicians being treatment and cure rather than prevention. The population strategy also depends almost entirely on the financial support of health agencies, while the high-risk strategy receives substantial help from the pharmaceutical industry. For the population strategy to prevail, it must become an integral part of official public health policy. Only the population strategy will ensure that prevention will cover the whole life span for all people and achieve the ultimate aim of true prevention: creation of a new generation in which low risk is the rule and high risk the exception.

Prevention must start early in life, though it is probably never too late. In an aging population, as age advances, latent disease becomes more frequent and the dividing line between primary and secondary prevention becomes more blurred. Concerning secondary prevention, half of the events of myocardial infarction in the community occur in persons with manifest CHD. This again puts the community view into focus. Preventive care on the community level, covering the population, recognizes not only individuals but families; it should know no ethnic boundaries and should make it easier to integrate curative and preventive medicine. The population strategy will also provide a base for the much-needed chronic disease surveillance systems that should become as much taken for granted as those for acute disease surveillance.

Disease prevention programs on a national scale need to be monitored to measure their progress and success. This is part of good public health housekeeping. Such measures are presently available, even in the United States, only in terms of mortality. There exists no reliable, direct evidence that incidence and prevalence have changed, and, if so, to what degree and in which direction compared with the observed decline in CHD and in total mortality. While it is likely that this decline can be attributed in large part to primary prevention, this is not known with any degree of accuracy. The necessity of acute disease surveillance is accepted to prevent rapidly spreading epidemics. Chronic diseases spread more insidiously but, in their final impact, are no less threatening. It would be inefficient to set up surveillance systems for cardiovascular diseases only because other major diseases require monitoring just as much, especially because some are strongly linked with CHD. The experience gained along the road to success of CHD prevention provides a model for chronic disease prevention as a whole. Measuring progress by surveillance is a prerequisite for optimally effective prevention and control programs.


*    Toward Chronic Disease Prevention
up arrowTop
 up arrowSetting the Stage
 up arrowBrief Historical Overview
 up arrowGeographical Pathology and...
 up arrowFrom Individual to Community...
 up arrowIntervention Studies
 up arrowToward Coronary Heart Disease...
 *Toward Chronic Disease...
down arrowThe Journey Ahead
 down arrowReferences
 
The progress along the road traveled so far has been encouraging. CHD has become by far the most preventable of all chronic diseases, with the possible exception of lung cancer prevention through smoking cessation. One of the reasons for hope is the thought that the changes in lifestyles that protect against CHD may help to maintain health on a broader front. An early vision of the concept that diseases should be viewed in their entirety was represented by the Tecumseh Study, a study of health and disease in a total community. It was conceived by one of the outstanding epidemiologists of the last generation, Thomas Francis, Jr, whose distinguished career in infectious disease epidemiology eventually led him to include chronic disorders in his encompassing picture of human ecology.58 59 The first publication from this study related risk factors to several chronic diseases and the diseases to each other.60 In the Seven Countries Study, Ancel Keys also went beyond the primary concern with CHD; the subtitle of his book is "A Multivariate Analysis of Death and Coronary Heart Disease," and the book itself includes data on total mortality, cancer, deaths caused by violence and accidents, and other causes.14

The thesis that a number of chronic diseases share common causes and therefore may be prevented by similar measures has been documented in detail elsewhere.61 62 The evidence, admittedly preliminary, falls under the headings "mortality interrelationships," "sharing of risk factors," "sharing of lifestyles," and "secular trends," summarized as follows: (1) there are strong correlations between cardiovascular mortality and noncardiovascular mortality in the states of the United States63 and between mortality from CHD and colorectal cancer in different countries64 ; (2) CHD risk factors are also predictive for other diseases,65 66 and men in the MRFIT study with low risk factors have a low mortality not only from CHD but from other major causes of death67 ; (3) mortality from several chronic diseases is inversely related to a health practices index, a high index indicating a healthy lifestyle68 ; in North Karelia, both CHD mortality and, after a time lag, cancer mortality fell in the course of a community prevention program69 ; social class differences in mortality, related to psychological factors and working conditions, are similar for all major causes of death70 ; nutritional habits, especially fat consumption and alcohol intake, are related to CHD and several types of cancer in similar ways71 72 73 74 ; physical inactivity is associated with a higher risk of death for a number of diseases75 ; (4) the secular decline of CHD mortality in the United States was paralleled by declines in all major causes of death except lung cancer76 ; and internationally, there is concordance of the rises and falls in mortality from CHD and a number of other conditions.77

The parallelism of international secular mortality trends for major chronic diseases is presented in some detail from a recently published analysis of international mortality trends between 1950 and 1983.77 78 If we look first at Fig 3ADown at the secular trends for heart disease versus cancer other than lung cancer, in two time periods and two age groups, contemporary analysis indicates that there are more countries showing concordance than discordance, but the difference for men is not marked. However, in Fig 3ADown there is much more concordance than discordance on a time-lag analysis, if we assume that cancer mortality has a longer incubation period and takes 15 years longer to react to changes in lifestyles and the environment than heart disease. In this analysis, heart disease trends in the first period are compared with cancer trends during the second period. A striking degree of concordance emerges. For women, the trends are mostly concordant, even on contemporary analysis, reflecting the universal tendency toward a secular decline of mortality from most causes in women. For heart disease versus lung cancer trends (Fig 3BDown), there is a distinct tendency toward concordance for men, somewhat more marked on time-lag analysis. This tendency is unexpected because the overall trends for the two conditions generally have been in opposite directions; for this reason, it is not surprising that the degree of concordance differs between the two periods. The lack of concordance for women is expected in view of the almost universal decline of their heart disease trends in contrast to those for lung cancer. Heart disease and cerebrovascular disease mortality trends in Fig 3CDown show much concordance, especially during the second period, even on contemporary analysis. These results are encouraging and lend support to the hypothesis proposed. The main purpose, at this point, is to encourage thinking and working along these lines to provide stronger evidence whether prevention of heart disease goes a long way toward preventing chronic disease more generally.





View larger version (108K):
[in this window]
[in a new window]
  		Figure 3. A, Concordance of secular mortality trends: Heart disease (HD) versus cancer other than lung cancer (COL). B, Concordance of secular mortality trends: Heart disease versus lung cancer (LC). C, Concordance of secular mortality trends: Heart disease versus cerebrovascular disease (CD).

If it should prove to be true that there are strong links between major chronic diseases in terms of common causes and particular lifestyles, it may be asked whether it is sensible to direct health messages toward the prevention of heart disease only, knowing that the recommended measures will be more broadly protective. The public is used to thinking in terms of specific diseases such as heart disease or cancer, and campaign appeals to preserve health more widely might, at first, fail to reach their target on account of being perceived as vague. New health messages would have to be developed, addressing heart disease as well as other chronic disorders, especially those with common links to heart disease. In the same way, preventive programs in the community and for individuals ought to be comprehensive, including multiple chronic diseases. In such programs, the identity of the separate diseases and of the private or public agencies that sponsor them should not be lost. The need is for coordination between these agencies. In addition, this would lead to more efficient and better use of resources.79 The WHO has made moves in this direction.80

Comprehensive chronic disease prevention programs may be expected to lead to further increases in longevity. The increases that have occurred during the last decades have been largely due to improvements in medical care. With more elderly and old people in the population, disability and invalidism have also been rising. It is not possible at this time to guess how many of these impairments could have been avoided if, along with better medical therapy, there had been populationwide preventive programs to encourage better lifestyles. There is no greater research priority today than to find evidence that the preventive measures that help to postpone death will, in addition, help to maintain physical, mental, and social function with advancing age. Preventive cardiology has established itself, but in an aging population, there is an equal need for preventive gerontology to ensure that people not only live longer but stay healthy and able-bodied. The comprehensive prevention of chronic disease will only be a gain to society as a whole if it can be backed by progress in preventive gerontology. It is not easy to develop research methods to show how disability can be avoided or postponed. One beginning has been made by an international working group using an index of "disability-free life expectancy"—healthy life expectancy—to be compared with the actual life expectancy in different countries.81 82 The ratio of disability-free to actual life expectancy is an index of health, and the task is to find the causes of higher and lower ratios. This would seem to be one promising approach toward identifying the factors associated with infirmity and health, respectively, in the later years of life.


*    The Journey Ahead
up arrowTop
 up arrowSetting the Stage
 up arrowBrief Historical Overview
 up arrowGeographical Pathology and...
 up arrowFrom Individual to Community...
 up arrowIntervention Studies
 up arrowToward Coronary Heart Disease...
 up arrowToward Chronic Disease...
 *The Journey Ahead
down arrowReferences
 
A review at the time of Keys's definitive book on the Seven Countries Study14 ended with the words: "Keys's contributions to the epidemiology of coronary heart disease are enormous, but this report suggests that what he has begun may be far greater than what he has done."83 In a similar way, CVD epidemiology as a whole has not only suggested a pattern of approach toward the prevention of other chronic diseases but would seem to have opened new horizons. The chances are that the future will bring advances in scientific knowledge of disease mechanisms, putting preventive measures on even stronger foundations. It is also likely that there will be a deepened understanding of the forces that determine individual and group attitudes and behaviors relating to the preservation of health. On the other hand, it is uncertain whether or to what extent future socioeconomic and political developments, in different parts of the world and cultures, will favor a climate in which healthy lifestyles can establish themselves. Competing demands coming from priority needs to control infectious diseases in some countries also must be taken into account. At this time, at the crossroads of change, it is hazardous to predict the direction in which the world is moving. However, as far as preventive medicine is concerned, there is no alternative to holding a steady course in the hope that the gains already made can be consolidated and extended into the future.

In this connection, it would be appropriate to ask why preventive cardiology has come to be such a success story. The reasons are, in large part, scientific. At the beginning, there were hypotheses that turned out to be correct and yielded powerful predictors, serving at the same time to institute preventive measures. The risk factor base was extended and the epidemiological study methods improved and strengthened. In addition, however, a part of the secret consisted in the remarkable degree of goodwill and understanding among the scientific workers sharing a common cause. The productive spirit of collaboration was maintained throughout the crucial years, an amazingly short period, considering that the study of chronic diseases, by their very nature, needs time. The magic of these days is no longer what it has been, and this is more in some countries than others. Arguments and controversies are essential for scientific progress, but, depending on the way they are conducted, they can be destructive on a number of levels. Along the road ahead, it will sometimes be well to look back.

Traditionally, the point of departure in the quest for the causes of diseases and their prevention have been the diseases themselves. It would, however, also be possible to start out with patterns of lifestyles and the search for clusters of diseases that share similar patterns. If the focus is on the preservation and promotion of health, it is logical to begin with the habits of daily living compatible with optimal well-being. This approach is particularly appropriate when the same harmful lifestyles cause a number of different disorders. This new point of departure would possibly provide stronger motivations toward changing detrimental health behavior than an initial focus on disease. There are still formidable barriers to such changes, based on evidence from controlled trials and community intervention studies. Novel approaches to foster reorientation of attitudes concerning health are needed. Historically, medical thinking, even in preventive medicine, centers on disease. To promote health among the people, the first thought should be given to what people do. In practice, the two approaches are not alternatives but ought to be combined to best advantage.

Advances in the epidemiology of cardiovascular diseases, cancer, diabetes, and other chronic conditions have, so far, been mostly along parallel and rarely intersecting lines. There would have been little point in thinking of interrelationships between these disorders before knowledge of their epidemiology individually had reached an adequate level for these purposes. This stage having now been reached, future work would benefit from focusing on the intersecting lines. The hope envisaged here is to find common denominators in lifestyles for several chronic disorders so that they might become amenable to the same or similar preventive measures. Epidemiological research in disease-specific fields, such as the cardiovascular disorders, must, of course, go on. Increasing attention, however, should be given to develop the interdisciplinary field of "chronic disease epidemiology," a field devoted to searching for links between the separate disorders.

In this Ancel Keys Lecture, Keys's own work has been emphasized repeatedly, though many of his major accomplishments have not been touched. The account started with the establishment of the Laboratory of Physiological Hygiene, which Keys led to fame. Health is the maintenance of physiology, warding off pathology, as Hygeia tries in Aesculapian medicine to postpone the time of calling. Cardiovascular disease epidemiology has come a long way and the future is promising. It is hoped that it has set an example and provided a model for the prevention of other chronic diseases.


*    Footnotes
 
Reprint requests to Dr Felix Gutzwiller, Institut fur Sozial und Präventivmedizin, Der Universität Zürich, Sumatrastrasse 30, CH-8006 Zurich, Switzerland.


*    References
up arrowTop
 up arrowSetting the Stage
 up arrowBrief Historical Overview
 up arrowGeographical Pathology and...
 up arrowFrom Individual to Community...
 up arrowIntervention Studies
 up arrowToward Coronary Heart Disease...
 up arrowToward Chronic Disease...
 up arrowThe Journey Ahead
 *References
 

  1. Keys A, Taylor HL, Blackburn H, Brozek J, Anderson JT, Simonson E. Coronary heart disease among Minnesota business and professional men followed 15 years. Circulation. 1963;28:381-395. [Abstract/Free Full Text]
  2. Dawber TR. The Framingham Study: The Epidemiology of Atherosclerotic Disease. Cambridge/London: Harvard University Press; 1980.
  3. Keys A. The inception and pilot surveys. In: Kromhout D, Menotti A, Blackburn H, eds. The Seven Countries Study: A Scientific Adventure in Cardiovascular Epidemiology. Utrecht, Netherlands: Brouwer Offset bv; 1994:15-27.
  4. Keys A, White PD. Cardiovascular Epidemiology. New York, NY: Paul B. Hoeber; 1956.
  5. Cooperative Study of Lipoproteins and Atherosclerosis. Evaluation of serum lipoprotein and cholesterol measurements as predictors of clinical complications of atherosclerosis. Circulation. 1956;14(part II):691-741.
  6. Gofman J, Lindgren F, Elliott H, Mantz W, Hewitt J, Strisower B, Herring V. The role of lipids and lipoproteins in atherosclerosis. Science. 1950;111:166-171. [Free Full Text]
  7. Cohn AE, Ling C. The Burden of Diseases in the United States. New York, NY: Oxford University Press; 1950.
  8. Adlersberg D, Parets AD, Boas EP. Studies of families with xanthoma and unselected patients with coronary artery disease under the age of 50 years. JAMA. 1949;141:246-254. [Medline] [Order article via Infotrieve]
  9. Blackburn H. Introduction to Ancel Keys Lecture: Ancel Keys, pioneer. Circulation. 1991;84:1402-1404. [Free Full Text]
  10. Hirsch A. Handbook of Geographical and Historical Pathology. Translated by Charles Creighton. London, England: The New Sydenham Society; 1883.
  11. Raab W. Alimentäre Faktoren in der Entstehung von Arteriosklerose und Hypertonie. Med Klin. 1932;28:487-521.
  12. Rosenthal SR. Studies in atherosclerosis: chemical, experimental and morphological: roles of cholesterol metabolism, blood pressure and structure of aorta, fat angle of aorta (FAA) and infiltration-expression theory of lipid deposit. Arch Pathol. 1934;18:473 and 660.
  13. Anitschkow N. Pathologische Anatomie und allgemeine Pathologie der Arteriosklerose. In: Askanazy M, ed. Deuxième Conférence Internationale de Pathologie Géographique. Utrecht, Netherlands: A. Oosthoek; 1934:44-101.
  14. Keys A. Seven Countries. A Multivariate Analysis of Death and Coronary Heart Disease. Cambridge/London: Harvard University Press; 1980.
  15. Keys A. Atherosclerosis: a problem in newer public health. J Mt Sinai Hosp. 1953;20:118-139.
  16. Yerushalmi J, Hilleboe HE. Fat in the diet and mortality from heart disease. NY State J Med. 1957;57:2343-2354. [Medline] [Order article via Infotrieve]
  17. Keys A, Anderson JT, Grande F. Serum cholesterol response to changes in the diet, IV: particular fatty acids in the diet. Metabolism. 1965;14:776-787.
  18. Keys A. Coronary heart disease in seven countries. Circulation. 1970;41(suppl I):I-1-I-211.
  19. Epstein FH. Public health implications of the Seven Countries Study. In: Kromhout D, Menotti A, Blackburn H, eds. The Seven Countries Study: A Scientific Adventure in Cardiovascular Disease Epidemiology. Utrecht, Netherlands: Brouwer Offset bv; 1994:169-174.
  20. Stamler J. Population Studies. In: Levy R, Rifkind B, Dennis B, Ernst N, eds. Nutrition, Lipids, and Coronary Heart Disease. New York, NY: Raven Press; 1979:25-88.
  21. Liu K, Stamler J, Trevisan M, Moss D. Dietary lipids, sugar, fiber and mortality from coronary heart disease: bivariate analysis of international data. Arteriosclerosis. 1982;2:221-227. [Abstract/Free Full Text]
  22. Artaud-Wild SM, Connor SL, Sexton G, Connor WE. Differences in coronary mortality can be explained by differences in cholesterol and saturated fat intakes in 40 countries but not in France and Finland. Circulation. 1993;88:2771-2779. [Abstract/Free Full Text]
  23. Simons LA. Interrelations of lipids and lipoproteins with coronary artery disease mortality in 19 countries. Am J Cardiol. 1985;57:5G-10G.
  24. Law MR, Wald NJ. An ecological study of serum cholesterol and ischaemic heart disease between 1950 and 1990. Eur J Clin Nutr. 1994;48:305-325. [Medline] [Order article via Infotrieve]
  25. Epstein FH. Myocardial infarction in Europe: geographical differences and time trends. In: Mason DT, Neri-Serneri GG, Oliver MF, eds. Florence International Meeting on Myocardial Infarction. Amsterdam/Oxford/Princeton: Excerpta Medica; 1979:11-21.
  26. World Health Statistics Quarterly. Vol 41. Geneva, Switzerland: World Health Organization; 1988.
  27. World Health Statistics Annual. Geneva, Switzerland: World Health Organization; 1989.
  28. Stewart AW, Kuulasma K, Beaglehole R. Ecological analysis of the association between mortality and major risk factors of cardiovascular disease: the World Health Organization MONICA Project. Int J Epidemiol. 1994;23:505-516. [Abstract/Free Full Text]
  29. Menotti A, Seccareccia F. Risk factors and mortality patterns in the Seven Countries Study. In: Toshima H, Koga Y, Blackburn H, Keys A, eds. Lessons for Science From the Seven Countries Study. Tokyo/Berlin/Heidelberg/New York: Springer; 1994:17-34.
  30. Papers from a workshop on Trends and Determinants of Coronary Heart Disease Mortality: International Comparisons. Int J Epidemiol. 1989;18(suppl 1):S1-S230.
  31. Vartiainen E, Puska P, Pekkanen J, Tuomilehto J, Jousilathi P. Changes in risk factors explain changes in mortality from ischaemic heart disease in Finland. Br Med J. 1994;309:23-27. [Abstract/Free Full Text]
  32. Higgins MW, Luepker RV. Trends in Coronary Heart Disease Mortality: The Influence of Medical Care. New York, NY: Oxford University Press; 1988.
  33. Gey KF, Puska P, Jordan P, Moser UK. Inverse correlation between plasma vitamin E and mortality from ischemic heart disease in cross-cultural epidemiology. Am J Clin Nutr. 1991;53(1 Suppl):326S-334S.
  34. Tunstall-Pedoe H, Kuulasma K, Amouyel P, Arveiler D, Rajakangas A-M, Pajak A. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project: registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation. 1994;90:583-612. [Abstract/Free Full Text]
  35. Keys A. The physiology of the individual as an approach to a more quantitative biology of man. Fed Proc. 1949;8:523-529. [Medline] [Order article via Infotrieve]
  36. Measuring the risk of coronary heart disease in adult population groups: a symposium. Am J Public Health. 1957;47:1-63.
  37. Kannel WB, Dawber TR, Kagan A, Revotskie L, Stokes J III. Factors of risk in the development of coronary heart disease: six-year follow-up experience. Ann Intern Med. 1961;55:33-50.
  38. Epstein FH. Predicting coronary heart disease. JAMA. 1967;201:795-800. [Medline] [Order article via Infotrieve]
  39. Epstein FH. Predicting, explaining, and preventing coronary heart disease. Mod Concepts Cardiovasc Dis. 1979;48:7-12. [Medline] [Order article via Infotrieve]
  40. Epstein FH. Coronary heart disease epidemiology revisited: clinical and community aspects. Circulation. 1973;48:185-194. [Abstract/Free Full Text]
  41. Rose G. Strategy of prevention: lessons from cardiovascular disease. Br Med J. 1981;1:1847-1851.
  42. Rose G. Sick individuals and sick populations. Int J Epidemiol. 1985;14:32-38. [Abstract/Free Full Text]
  43. WHO Scientific Group. Cardiovascular Disease Risk Factors: New Areas for Research. Geneva, Switzerland: World Health Organization; 1994.
  44. Postiglione A, Panico S, Lewis B, Eisenberg S, Fruchart JC, Hanefeld M, Heistad DD, Kesaniemi YA, Mann JI, Olsson AG, Paulson OB, Rossouw JE, Seidel D, Singh RB. New and candidate risk factors for coronary heart disease. Nutr Metab Cardiovasc Dis. 1994;4:233-255.
  45. Law MR, Wald NJ, Thompson SG. By how much and how quickly does reduction in serum cholesterol concentration lower risk of ischaemic heart disease? BMJ. 1994;308:367-372. [Abstract/Free Full Text]
  46. Holme I. An analysis of randomized trials evaluating the effect of cholesterol reduction on total mortality and coronary heart disease incidence. Circulation. 1990;82:1916-1924. [Abstract/Free Full Text]
  47. Tyroler HA. Review of lipid-lowering clinical trials in relation to observational studies. Circulation. 1987;76:515-522. [Abstract/Free Full Text]
  48. Collins R, Keech A, Peto R, Sleight P, Kjekshus J, Wilhelmsen L, MacMahon S, Shaw J, Simes J, Braunwald E, Buring J, Hennekens C, Pfeffer M, Sacks F, Probstfield J, Yusuf S, Downs JR, Gotto A, Cobbe S, Ford I, Shepherd J. Cholesterol and total mortality: need for larger trials. BMJ. 1992;304:1689. Letter.
  49. Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383-1389. [Medline] [Order article via Infotrieve]
  50. Mass Field Trials of the Diet-Heart Question: Their Significance, Feasibility and Applicability. Report of the Diet-Heart Review Panel of the National Heart Institute. American Heart Association Monograph No. 28, June 1969.
  51. Johnson CL, Rifkind BM, Sempos CT, Carrol MD, Bachoric PS, Briefel RR, Gordon DJ, Burt VL, Brown CD, Lippel K, Cleeman JI. Declining serum total cholesterol levels among US adults: the National Health and Nutrition Surveys. JAMA. 1993;269:3002-3008. [Abstract]
  52. Vartiainen E, Puska P, Jousilahti P, Korhonen HJ, Tuomilehto J, Nissinen A. Twenty-year trends in coronary risk factors in North Karelia and other areas of Finland. Int J Epidemiol. 1994;23:495-504. [Abstract/Free Full Text]
  53. Thom TJ. International mortality from heart disease: rates and trends. Int J Epidemiol. 1989;18(suppl 1):S20-S28.
  54. Epstein FH. Low cholesterol, cancer, and other non-cardiovascular disorders. Atherosclerosis. 1992;94:1-12. [Medline] [Order article via Infotrieve]
  55. Epstein FH. Interpreting the association between low serum cholesterol and disease: confounding and causality. In: Lewis B, Paoletti R, Tikkanen MJ, eds. Low Blood Cholesterol: Health Implications. London, England: Current Medical Literature Ltd; 1993:10-17.
  56. National Cholesterol Education Program. Second Report of the Expert Panel on detection, evaluation and treatment of high blood cholesterol in adults (Adult Treatment Panel II). Circulation. 1994;89:1329-1445.
  57. Pyörälä K, De Backer G, Poole-Wilson P, Wood D, on behalf of the Task Force. Prevention of coronary heart disease in clinical practice: recommendations of the Task Force of the European Society of Cardiology, European Atherosclerosis Society, and European Society of Hypertension. Eur Heart J. 1994;15:1300-1331. [Free Full Text]
  58. Francis T Jr. Aspects of the Tecumseh Study. Public Health Reports. 1961;76:963-966. [Medline] [Order article via Infotrieve]
  59. Francis T Jr. The epidemiological approach to human ecology. Am J Med Sci. 1959;237:677-684. [Medline] [Order article via Infotrieve]
  60. Epstein FH, Francis T Jr, Hayner NS, Johnson BC, Kjelsberg MO, Napier JA, Ostrander LD Jr, Payne MW, Dodge HJ. Prevalence of chronic diseases and distribution of selected physiologic variables in a total community: Tecumseh, Michigan. Am J Epidemiol. 1965;81:307-322. [Free Full Text]
  61. Epstein FH. Prevention of coronary heart disease: links with other chronic disorders. In: Lenzi S, Descovitch GC, eds. Atherosclerosis and Cardiovascular Disease. Lancaster/Boston/The Hague/Dordrecht: MTP Press Ltd; 1987:157-162.
  62. Epstein FH. Will measures to prevent coronary heart disease protect against other chronic disorders? In: Toshima H, Koga Y, Blackburn H, Keys A, eds. Lessons for Science from the Seven Countries Study. Tokyo/Berlin/Heidelberg/New York: Springer; 1994:179-193.
  63. Sauer HI, Enterline PE. Are geographic variations in death rates from cardiovascular disease real? J Chron Dis. 1959;10:513-524. [Medline] [Order article via Infotrieve]
  64. Sidney S, Farquhar JW. Cholesterol, cancer and public health policy. Am J Med. 1983;75:494-508. [Medline] [Order article via Infotrieve]
  65. Menotti A, Conti S, Giampaoli S, Mariotti S, Signoretti P. Coronary risk factors predicting coronary and other causes of death in fifteen years. Acta Cardiol. 1980;35:107-120. [Medline] [Order article via Infotrieve]
  66. Menotti A, Mariotti S, Seccareccia F, Diampaoli S. The 25-year estimated probability of death from some specific causes as a function of twelve risk factors in middle-aged men. Eur J Epidemiol. 1987;4:60-67.
  67. Stamler J, Dyer AR, Shekelle RB, Neaton J, Stamler R. Relationship of baseline major risk factors to coronary and all-cause mortality and to longevity: findings from long-term follow-up of Chicago cohorts. Cardiology. 1993;82:191-222. [Medline] [Order article via Infotrieve]
  68. Wingard DL, Berkman LF, Brand RJ. A multivariate analysis of health-related practices: a nine-year mortality follow-up of the Alameda County Study. Am J Epidemiol. 1982;116:765-775. [Abstract/Free Full Text]
  69. Puska P, Korhonen HJ, Torppa J, Tuomilehto J, Vartiainen E, Pietinen P, Nissinen A. Does community-wide prevention of cardiovascular disease influence cancer mortality? Eur J Cancer Prev. 1993;2:457-460. [Medline] [Order article via Infotrieve]
  70. Marmot MG, Shipley MJ, Rose G. Inequalities in death-specific explanations of a general pattern. Lancet. 1984;1:1003-1006. [Medline] [Order article via Infotrieve]
  71. Committee on Diet and Health, Food and Nutrition Board, Commission on Life Sciences, National Research Council. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington, DC: National Academy Press; 1989.
  72. WHO Study Group. Diet, Nutrition, and the Prevention of Chronic Diseases. World Health Organization Technical Report Series 797. Geneva, Switzerland: World Health Organization; 1990.
  73. Waldron KW, Johnson IT, Fenwick GR. Food and Cancer Prevention: Chemical and Biological Aspects. Cambridge, England: Royal Society of Chemistry; 1993.
  74. Kesteloot H, Lesaffre E, Joosens JV. Dairy fat, saturated animal fat, and cancer risk. Prev Med. 1991;20:226-236. [Medline] [Order article via Infotrieve]
  75. Paffenbarger RS, Hyde RT, Wing AL, Lee I-M, Jung DL, Kampert JB. The association of changes in physical activity level and other lifestyle characteristics with mortality among men. N Engl J Med. 1993;328:538-545. [Abstract/Free Full Text]
  76. Thom TJ, Kannel WB, Feinleib M. Factors in the decline of coronary heart disease mortality. In: Connor WE, Bristow JD, eds. Coronary Heart Disease: Prevention, Complications, Treatment. Philadelphia, Pa: Lippincott; 1988:5-20.
  77. Thom TJ, Epstein FH, Feldman JJ, Leaverton PE, Wolz M. Total Mortality and Mortality From Heart Disease, Cancer and Stroke From 1950 to 1987 in 27 Countries: Highlights of Trends and Their Interrelationships Among Causes of Death. NIH publication No. 92-3088. Bethesda, Md: National Institutes of Health, National Heart, Lung and Blood Institute; 1992.
  78. Thom TJ, Epstein FH. Heart disease, cancer, and stroke mortality trends and their interrelationships: an international perspective. Circulation. 1994;90:574-582. [Abstract/Free Full Text]
  79. Epstein FH, Holland WW. Prevention of chronic diseases in the community: one disease versus multiple disease strategies. Int J Epidemiol. 1983;12:135-137. [Free Full Text]
  80. Glasunov IS, Grabauskas V, Holland WW, Epstein FH. An integrated programme for the prevention and control of non-communicable diseases. J Chron Dis. 1983;36:419-426. [Medline] [Order article via Infotrieve]
  81. Myers GC. International research on healthy life expectance. In: Feinleib M, ed. Comparative International Vital and Health Statistics Report. Series 7, No. 7. Hyattsville, Md: Vital and Health Statistics, DHHS Publication No. (PHS) 93-1483. 1993:33-42.
  82. Robine J-M, Blanchet M, Dowd JE. Health Expectancy: First Workshop of the Healthy Life Expectancy Network. London, England: HMSO; 1992.
  83. Stallones RA. Patterns of health risk: Seven Countries. Science. 1980;208:1138-1139.[Free Full Text]