Why is Premature Coronary Heart Disease Mortality Falling?

Introduction

In the USA, after a relatively stable, low incidence, coronary heart disease (CHD) suddenly 'took off' in 1920 (Stallones, 1980). In Britain it happened in the late 1920s, although there was a noticeable increase in 1909/10 (Mackinnon, 1987). There were similar experiences, all at about the same time, throughout the developed world. Incidences of deaths from the disease in all affected countries climbed steadily for between three and four decades, during which time CHD became the greatest single cause of death. But then, in the 1950s and 1960s, the numbers of premature deaths peaked and began to fall. There have been several attempts to explain increases in the disease early in the century and attempts have also been made to explain the decline (Stern, 1979). But the suggestion that recent dietary changes, reductions in serum cholesterol and smoking habit, improved hypertension control and, perhaps, exercise have accounted for the decline are unconvincing, particularly since the decline began before there was any significant change in these lifestyle patterns.

Nevertheless, there has been a dramatic decline and if the reasons for the decline are understood, perhaps we may find clues to the causes of CHD.

Many 'risk factors' for CHD have been suggested (Hopkins & Williams, 1981). A review, of the more widely accepted ones (Stallones, 1980) demonstrated that none fitted the perceived wisdom. Trends in dietary fats, smoking habit, hypertension and an increasingly sedentary lifestyle all follow patterns which bear little relation to the rises and falls in deaths attributed to CHD. Several other reviews of the data have also found little convincing evidence to support current lifestyle recommendations (Yudkin,1957; Marmot, et al, 1981; McCormick & Skrabanek, 1988; Oliver, 1992; Dunnigan, 1993).

People's lifestyles are being manipulated and changed, and the incidences of premature deaths from CHD are declining. Some will say that the latter is the result of the former. But this cannot be true as the current dietary and lifestyle recommendations are much too recent: the COMA Report, for example, on which dietary recommendations for Britain are based was not published until l984 but premature CHD mortality had peaked and started to decline in 1965 (see Figure 1). At that time we ate fried breakfasts, bread and dripping and 'went to work on an egg'.

When did premature CHD mortality really start to decline?

Based on the graph at Figure l, one might argue that the then current high-meat, high-fat dietary recommendations contributed to the decline. Indeed, that was what I expected to demonstrate. However, if there were some factor around 1965 which resulted in a decline in men aged 40 to 44, it could reasonably be expected also to have affected men in other age groups at the same time. But this is not the case: Declines in older age groups are as dramatic as those in Figure 1, - but they occur later. Investigation into these age groups throws up a curious and significant pattern. Mortality (cause) statistics for England and Wales are published in 5-year age groups. The declines for successive 5-year age groups of men, 45-49, 50-54, 55-59, and 60-64, all begin 5 years apart: in 1969, 1974, 1979 and 1984 respectively. The pattern is much the same for women.

Whatever the reason for the dramatic declines, it must surely be something that is common to them all. And the only thing these men have in common is their birth dates: premature CHD mortality rises in those born before 1920, peaks in those born 1920-24, and declines in those born after 1925. It is so remarkably consistent that it leaves room for only one conclusion: whatever the reason for the present decline in CHD mortality, it must have started during the latter half of the 1920s. And it must have been something dramatic.

What caused premature CHD mortality to fall?

Having determined that the decline in premature CHD mortality stems from some event(s) around, or soon after, 1925, the question now to be resolved is: what was that event? And here we are in the realms of speculation.

CHD is not a contagious disease: it is caused by environmental factors. It was originally thought to be linked to cholesterol intake (Gofman, et al , 1950) or saturated fat intake (Keys, 1953), but many clinical trials have demonstrated convincingly that while diet may play a significant part in the aetiology of CHD by increasing our life-expectancy, it plays little or no part in premature CHD mortality. That hypothesis is sustained only by the selective citation of supportive trials (Ravnskov, 1992).

There is no doubt that CHD is a disease of affluent countries. But premature CHD mortality tends not to afflict affluent people; it was and is more common among the poor in those countries (Blane, et al , 1996).

Societal trends of CHD rates in later life parallel trends in both neonatal and postneonatal mortality (Barker & Osmond, 1986). Barker and Osmond suggest that poor nutrition in early life increases susceptibility to the effects of an affluent diet later. Data from several countries also demonstrate an inverse relation between children's mean serum cholesterol concentrations and child mortality (Deutsch, 1995).

There were many changes which may have affected health in the first quarter of this century: changes in methods and places of childbirth, analgesia during childbirth, formula baby foods, bottle sterilization, changes in patterns of breast vs bottle feeding, slum clearances, Clean Air Acts, refrigeration of food, free school milk and dinners, and nutritional supplements. These make pinning the decline on any one very difficult.

For any to have been responsible for the dramatic reversal of the CHD trends seen in people born in the late 1920s and subsequently, their changes must have been dramatic and sudden. They are likely to have been factors which affected the foetus and/or newborn child.

Chronic bacterial infection

Chlamydia pneumoniae and Helicobacter pylori, ubiquitous today, may have a causal relationship with CHD (Patel, et al , 1995). Although only recently discovered (Marshall, 1988), H pylori must have played a large role in the first half of the century. Penicillin was discovered in 1928 which is at the time we are considering. But penicillin cannot have had the necessary dramatic effect at that time as it was not widely used until after the 1939-45 War. And if the various slum clearance programmes early in the century helped to reduce childhood infections, conditions prevailing before such programmes do not explain the low incidence of CHD prior to 1920.

Air Pollution.

In the early part of this century most houses burned coal and were lit with coal gas. Lamps and cookers gave off carbon monoxide. This may have increased carboxyhaemoglobin levels in expectant mothers and children. In 1926 in Britain the Electricity Act marked the start of a rapid change to universal electric light which may have reduced CO levels in houses considerably. This fits the timescale well but, again, it does not explain the previous low incidence.

The depression.

There was a period of unrest throughout Britain early in the century (Scott-James, 1930). Times for many were hard. Food, particularly relatively expensive proteins and fats, would have been scarce. This period of unrest ended abruptly with the General Strike in 1926, which again fits well the timescale.

Nutrition

Background. Many childhood diseases have declined dramatically since the mid-nineteenth century. Most of this decline happened long before antibiotics and immunisation were widely used (Porter, 1971). Although some of the decline can be attributed to improved housing and sanitation, an important factor was better nutrition that conferred a higher host-resistance to disease.

Dietary animal fat, by raising serum cholesterol levels, is hypothesized now to be the major factor predisposing to CHD. Committees recommend a change towards polyunsaturated vegetable fats (Grundy, et al , 1982; NACNE, 1983; COMA, 1984). But epidemiological studies and dietary trials have failed consistently to demonstrate any benefit from such a change.

An Indian study (Malhotra, 1967) showed seven times the occurrence of myocardial infarction and fifteen times higher mortality from CHD in south Indians compared with north Indians, even though the north Indians consumed nine times more fat, most of which was saturated animal fat. Similarly CHD was unknown in the milk- and meat-eating Maasai and Sambura tribes of Africa (Shaffer, 1963). Comparing the Eskimos of Greenland and Baffin Island with those of Labrador shows the same pattern (McClellan & Du Bois, 1930). The former, eating no vegetable foods, had no CHD, while the latter, who ate 'civilized' food - dried potatoes, flour, canned foods and cereals - did. After almost half a century of research into diet and cardiovascular diseases, the Framingham researchers say: "Intakes of fat and type of fat were not related to the incidence of the combined outcome of all cardiovascular diseases or to total or cardiovascular mortality" (Gillman, et al , 1997).

At the beginning of this century, when mortality and morbidity from CHD was low, fat intakes were predominantly of butter and animal fats. The rises in CHD mortality in Britain followed rises in consumption of margarine and vegetable shortenings (Yudkin,1957).

In the 1920s, Sir John Boyd Orr compared growth rates of children in public schools with those in state schools (Orr, 1936). He demonstrated convincingly that children of the socially deprived, who lived on a largely carbohydrate diet of bread and potatoes, benefited from a diet supplemented with full-cream milk.

The Change. In1920s Britain there was a change in farming policy. Acreage in arable production fell by 2 million. At the same time pasture and grazing increased by 2.5 million acres (Whitaker's Almanac. 1922, 1932). People were urged to eat more milk, meat, eggs and butter.

Conclusion.

Premature CHD started to decline at a time when a diet high in animal proteins and fat was advocated. The decline is now showing signs of faltering with the current low-fat regime as CHD mortality in younger women has risen in six of the last seven years (HMSO, 1988-95).

The decline in premature CHD deaths began in those born in the latter half of the 1920s. Several things changed almost simultaneously just at that time. It is possible that they all played some part. However, there is strong evidence that experience in infancy has a profound bearing on risk of cardiovascular diseases in later life. Thus the most important change in the 1920s and 1930s was probably the recommended dietary change in favour of foods of animal origin which gave children a better start in life.

That this may be so is supported by much research showing that present 'prudent' dietary guidelines have heralded a return of many childhood diseases which had disappeared under the previous dietary advice.

References

Barker DJP, Osmond C. (1986). Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet i; 1077-81.
Blane D, Hart CL, Davey Smith G, Gillis CR, Hole DJ, Hawthorne VM. (1996). Association of cardiovascular disease risk factors with socioeconomic position during childhood and during adulthood. British Medical Journal 313; 1434-8.
Committee on the Medical Aspects of Food Policy (COMA). (1984) Diet and Cardiovascular Disease . Department of Health and Social Security.
Deutsch ME. (1995) Restriction of fat in the diet of infants. Lancet 346; 1117-8.
Dunnigan MG. (1993). The problem with cholesterol: no light at the end of the tunnel. British Medical Journal 306; 1355-6.
Gillman MW, Cupples LA, Millen BE, Ellison RC, Wolf PA. (1997). Inverse association of dietary fat with development of ischemic stroke in men. Journal of the American Medical Association 278; 2145-50.
Gofman JW, Lindgren F, Elliott H, Mantz W, Hewitt J, Strisower B, Herring V. (1950) The role of lipids and lipoproteins in atherosclerosis. Science 111; 166-71, 186.
Grundy SM, Bilheimer D, Blackburn H, et al . (1982). Rationale of the diet heart statement of the American heart Association, report of the nutrition Committee. Circulation 65; 839A-854A.
Hopkins PN, Williams RR. (1981) A survey of 246 suggested coronary risk factors. Atherosclerosis 40; 1-
Keys A. (1953). Atherosclerosis: a problem in newer public health. Journal of the Mount Sinai Hospital 20; 118-139.
Mackinnon AU. (1987). The origin of the modern epidemic of coronary heart disease in England. Journal of the Royal College of General Practitioners 37; 174-6.
Malhotra SL. (1967). Serum lipids, dietary factors and ischemic heart disease. American Journal of Clinical Nutrition 20; 462-74.
Marmot MG, Booth M, Beral V. (1981). Changes in heart disease mortality in England and Wales and other countries. Health Trends 13; 33-42.
Marshall BJ. (1988). The Campylobacter pylori story. Scandinavian Journal of Gastroenterology. 146; Supplement, 58-66.
McClellan WS, Du Bois EF. (1930). Prolonged meat diets with a study of kidney function and ketosis. Journal of Biological Chemistry 87; 651-668.
McCormick J, Skrabanek P. (1988). Coronary heart disease is not preventable by population interventions. Lancet ii; 839-41.
National Advisory Committee on Health Education (NACNE). (1983). A Discussion Paper on Proposals for Nutritional Guidelines for Health Education in Britain . Health Education Council.
Oliver MF. (1992). Doubts about preventing coronary heart disease. British Medical Journal 304; 393-4.
Orr JB. (1936). Food, Health and Income . London
Patel P, Mendall MA, Carrington D, Strachan DP, Leatham E, Molineaux N, et al . (1995). Association of Helicobacter pylori and Chlamydia pneumoniae infections with coronary heart disease and cardiovascular risk factors. British Medical Journal 311; 711-14.
Porter RR. (1971). The contribution of the biological and medical sciences to human welfare. Presidential Address to the British association for the Advancement of Science, Swansea Meeting, 1971. Published by the association, 1972.
Ravnskov U. (1992). Cholesterol lowering trials in coronary heart disease: frequency of citation and outcome. British Medical Journal 305; 15-19.
Registrar General's Mortality (Cause) Statistics (1988-1995). ICD 410-414, Women aged 30-34. Her Majesty's Stationary Office, London.
Scott-James RA. (1930). The war period: when it began and when it ended. Saturday Review November; 661-2, 697-8.
Shaffer RD, Mann GV, Anderson RS, Sandstead HH. (1963). Heart disease among the milk and meat-eating Masai of Tanganyika. Proceedings of the International Congress on Nutrition. 6th, Edinburgh 1; 616.
Stallones RA. (1980). The Rise and Fall of Ischaemic Heart Disease. Scientific American 243; 43-9.
Stern MP. (1979). The recent decline in ischemic heart disease mortality. Annals of Internal Medicine 91; 630-40.
Whitaker's Almanac. (1922, 1932).
Yudkin J. (1957). Diet and coronary thrombosis. Hypothesis and fact. Lancet ii; 155-61.

Last updated 30 January 2003