History of Recommendations to the Public about Dietary Fat

Abstract

Recommendations to Americans concerning dietary fat and heart disease began to appear in the late 1950s. This followed the publications of Gofman et al. (1950) and Keys (1953) relating to techniques for separating plasma lipoprotein fractions and the epidemiologic correlations between dietary fat, serum cholesterol and heart disease, respectively. Advice to the public after 40 years is similar to that given originally, namely, to reduce total fat, saturated fat and cholesterol intake, although cholesterol intake per se is not correlated strongly with cholesterolemia. Newer players on the heart disease stage are homocysteinemia, chlamydia infection and cytomegalovirus. These findings, when amplified, may alter the thrust of medical and dietary advice. In the meantime, since 1950, deaths from all causes in the U.S. (per 100,000, age-adjusted) have fallen by 40% and deaths from heart disease and stroke have fallen by 53 and 70%, respectively.

cholesterol, dietary fat, dietary guidelines, lipids, lipoproteins

Recommendations to the American public regarding dietary fat have a relatively recent history. In a book entitled The American and His Food published in 1940 (Cummings 1940), the author traces the history of the American diet and the events that shaped it without mentioning dietary fat. Until the fourth decade of this century, heart disease and cancer made relatively minor contributions to total mortality. When the major cause of death was infectious disease, a calorie-rich diet was deemed helpful in recovering from illness.

Real interest in dietary fat and its effects—particularly with regard to its role in cardiovascular disease—was stimulated by several papers published in the early 1950s. Gofman and his colleagues at the University of California, Berkeley published a paper in Science (Gofman et al. 1950) that detailed findings related to their new technique of separating plasma lipoproteins by ultracentrifugation. They showed that levels of certain of these lipoprotein classes were related to atherosclerotic heart disease and implicated dietary fat as a factor in this relationship. At about the same time Ancel Keys embarked on his worldwide epidemiologic investigations of dietary fat and heart disease prevalence, which showed that the level of dietary fat was related to mortality from heart disease (Keys 1953). In his “Seven Countries” study, Keys (1970) found a significant association between fat and saturated fat intake and heart disease mortality. Yerushalmey and Hilleboe (1957) pointed out that if 21 other countries were included, the association observed by Keys was weak and that a similar association could be advanced between animal protein intake and heart disease.

The role of dietary cholesterol in the etiology of heart disease had been a subject of much earlier research and speculation. The early history is detailed in a book published in 1958 (Kritchevsky 1958). The observation that cholesterol was a constituent of the atherosclerotic plaque was noted in a pathology text published 150 years ago (Vogel 1847). Any number of investigators showed that atherosclerotic aortas contained significantly more cholesterol than normal ones. Others showed that cholesterol feeding alone was sufficient to establish cholesterol-rich lesions in the arteries of rabbits and chickens. Atherosclerotic lesions could be established in rats, dogs and monkeys by cholesterol feeding plus other dietary and hormonal manipulations. Although these findings led to innumerable experimental studies of cholesterol/fat feeding and atherosclerosis, Stehbens (1989) has argued persistently that the human and experimental lesions in animals are different enough to cast doubt on the validity of the experimental lesion as an example of human disease. Although the role of cholesterolemia and hyperlipidemia in the etiology of human atherosclerosis was not accepted unanimously in the 1950s, it was considered sufficient by some authorities to establish a case against high intakes of dietary fat and cholesterol. This was enough to open the door to dietary guidelines offered to the public for possible prevention or amelioration of heart disease.

THE EVOLUTION OF DIETARY GUIDELINES

The first guidelines were published in 1957 (Page et al. 1957) (Table 1). What stands out are the qualifiers (may, probably) that appear in every sentence. Four years later (Page et al. 1961), the guidelines (Table 2) had assumed the general shape that we recognize today. In 1968, the Committee on Nutrition of the American Heart Association published eight dietary guidelines as follows: 1) reduce animal fat, 2) decrease saturated fats and increase polyunsaturated fats, 3) reduce cholesterol, 4) maintain ideal body weight, 5) apply dietary recommendations early in life, 6) maintain the principles of good nutrition with the change in diet, 7) adhere to dietary recommendations, and 8) make sound food habits a family affair. In 1978, the Committee on Nutrition added a recommendation to reduce sodium intake. Earlier, in 1977, the Senate Select Committee on Nutrition and Human Needs published their dietary guidelines for Americans (U.S. Senate Select Committee on Nutrition and Human Needs 1977) as shown in Table 3.

Table 1

American Heart Association Report (1957)^1-1

1.Diet may play an important role in the pathogenesis of atherosclerosis.

2.The fat content and total calories in the diet are probably important factors.

3.The ratio between saturated and unsaturated fat may be the basic determinant.

4.A wide variety of other factors beside fat, both dietary and non-dietary, may be important

F1-1

Page et al. (1957).

Table 2

American Heart Association Report (1961)^2-1

1.Maintain a correct body weight.

2.Engage in moderate exercise, e.g., walking to aid in weight reduction.

3.Reduce intake of total fat, saturated fat, and cholesterol. Increase intake of polyunsaturated fat.

4.Men with a strong family history of atherosclerosis should pay particular attention to diet modification.

5.Dietary changes should be carried out under medical supervision.

F2-1

Page et al. (1961).

Table 3

Dietary Goals for the United States (1977)^3-1

1.Increase carbohydrate consumption to account for approximately 55 to 60% of energy intake.

2.Reduce overall fat consumption from 40 to 30% of energy intake.

3.Reduce saturated fat consumption to account for about 10% of total energy intake; and balance that with polyunsaturated and monounsaturated fat, which should account for 10% of energy intake each.

4.Reduce cholesterol consumption to about 300 mg/day.

5.Reduce sugar consumption by about 40% to account for about 15% of total energy intake.

6.Reduce salt consumption by about 50 to 85% to about 3 g/day.

F3-1

U.S. Senate Select Committee (1977).

McNutt discussed and summarized the dietary advice given to the public between 1957 and 1980 (McNutt 1980). She summarized eleven sets of guidelines presented to the American public between 1957 and 1980 and noted a general coherence among them.

In 1980, the Food and Nutrition Board of the National Academy of Sciences published a 24-page booklet entitled “Toward Healthful Diets” (Food and Nutrition Board 1980). Their five recommendations were as follows:

The FNB/NAS report was attacked with unbelievable vehemence by the professional and lay press. It appeared that few of the attackers had read the original publication but were spurred on by newspaper reports alleging that the FNB had ignored the role(s) of cholesterol and fats in the etiology of coronary disease and had indeed given dietary cholesterol a green light. In fact the report said the following: “For persons with a positive family history of heart disease and other risk factors, such as obesity, hypertension, and diabetes, concentrations of blood lipids and lipoprotein fractions should be determined and, if any are abnormal, therapy should be undertaken under a physician's guidance.” The 1990 and 1995 dietary guidelines for Americans are listed in Table 4. The similarities of these guidelines to those in “Toward Healthful Diets” are striking.

Table 4

Dietary Guidelines for Americans^4-1

1990	1995
Eat a variety of foods	Balance the food you eat with physical activity-maintain or improve your weight
Maintain health weight	Choose a diet with plenty of grain products, vegetables, and fruits
Choose a diet low in fat, saturated fat, and cholesterol	Choose a diet low in fat, saturated fat, and cholesterol
Choose a diet with plenty of vegetables, fruits, and grain products	Eat a variety of foods
Use sugars only in moderation	Choose a diet moderate in salt and sodium
Use salt and sodium only in moderation	Choose a diet moderate in sugars
If you drink alcoholic beverages, do so in moderation	If you drink alcoholic beverages, do so in moderation

F4-1

U.S. Department of Agriculture and U.S. Department of Health & Human Services.

THE ROLE OF CHOLESTEROL IN ATHEROSCLEROSIS

The role of dietary cholesterol in cholesterolemia and atherosclerosis is unclear. Gertler et al. (1950) compared cholesterol levels and cholesterol intake in patients with and without coronary disease. They looked at men with the lowest and highest levels of serum cholesterol or cholesterol intake. In every case, the serum cholesterol levels of men with coronary disease were significantly higher than those of the controls but they bore no relation to their habitual intake of cholesterol. Keys et al. (1950) had observed “... the serum cholesterol of ‘normal’ men is not significantly related to differences in the habitual cholesterol intake over a range of 250–800 milligrams per day.” Landé and Sperry (1936) examined blood cholesterol and aortic atherosclerosis in 83 cases of violent death and found severity of atherosclerosis to be a function of age rather than cholesterolemia: A factor that tends to neutralize the effects of dietary cholesterol on plasma cholesterol is the biosynthetic capacity of the liver, which is regulated in a manner reciprocal to the dietary intake.

Table 5

Age Adjusted Death Rates per 100,000 (U.S.)^5-1

	Year
	1950	1960	1970	1980	1992	Δ (%) 1950
All causes	840.5	760.9	714.3	585.8	504.5	−40
Heart disease	307.2	286.2	253.6	202.0	144.3	−53
Stroke	88.6	79.7	66.3	40.8	26.2	−70
Cancer	125.3	125.8	129.8	132.8	133.1	+6
Pneumonia	26.2	28.0	22.1	12.9	8.0	−69
Suicide	11.0	10.6	11.8	11.4	11.1	+1
Homocide	5.4	5.2	9.1	10.8	10.5	+94

F5-1

U.S. Department of Health and Human Services (1996).

INTERVENTION TRIALS AND A CAUTIONARY NOTE

The Nutrition Committee of the American Heart Association issued a detailed discussion of the AHA diet-heart statement (Grundy et al. 1982), reviewing earlier suggestions and presenting a rationale for the then current statement. They cautioned against inclusion of very large amounts (over 15% of calories) of polyunsaturated fats and recommended an upper limit of 10% of calories. They also cited American epidemiologic studies, which had failed to find significant correlations among dietary fat, serum cholesterol and risk of coronary heart disease (CHD). Among them was a report from Framingham (Kannel and Gordon 1970), in which no relation was found between diet and cholesterol levels in several hundred men and women, and the Tecumseh Study (Nichols et al. 1976), which failed to find correlations between any dietary component and coronary disease rates. A study that was not cited (Gordon et al. 1981), compared diets of men who had coronary events with those who did not in the Framingham, Puerto Rico and Hawaiian Heart Studies. They found significant differences only in calorie, carbohydrate and alcohol intake but none in cholesterol intake or ratio of polyunsaturated to saturated fat.

The report also suggested an “ideal” range of serum cholesterol to be between 130 and 190 mg/dL (3.36–4.91 mmol/L). But is it? There are many data pointing to some relationship between the risk of cancer and low cholesterol levels (Kritchevsky and Kritchevsky 1992). It is probably not a case of low cholesterol per se being a factor, but a connection to it appears to exist. Nevertheless, the occurrence of a U-shaped curve for all-cause mortality as a function of serum cholesterol level prompted the NHLBI to convene a conference to examine this relationship and explain why the all-cause death rate was similar at the extreme ends of the serum cholesterol range. Nineteen studies were considered and no definitive conclusions were reached. Further studies to investigate the relationship of noncardiovascular deaths to serum cholesterol levels were recommended (Jacobs et al. 1992). These findings stimulated an editorial by Hulley et al. (1992) suggesting changes in the direction of our public lipid-lowering policies particularly because primary prevention trials revealed as many deaths due to noncardiovascular disease as lives saved in persons with cardiovascular disease.

Research continues apace, and as new findings appear, it is necessary to reevaluate our conclusions and preventive medicine policies. A recent report from the Russian Lipid Research Clinics (Shestov et al. 1993) concluded: “The results disclose a sizable subset of hypocholesterolemics in this population at increased risk of cardiac death associated with lifestyle characteristics.” The characteristics included low LDL-cholesterol, high HDL-cholesterol, higher alcohol consumption, lean body mass and less education. A role for dietary fiber as protective against myocardial infarction and independent of fat intake has been suggested (Rimm et al. 1996) and has confirmed other findings (Khaw and Barrett-Connor 1987, Humble et al. 1993).

NEW FACTORS ASSOCIATED WITH ATHEROGENESIS

Several other non-lipid factors have been introduced into the heart disease arena. Homocysteinemia, which has been discussed as a factor in cardiovascular disease for several decades (McCully 1969), has now been shown to be significantly (r = 0.71) related to cardiovascular disease mortality in a substudy of the WHO/MONICA Project (Alfthan et al. 1997). The possible involvement of Chlamydia pneumoniae in atherogenesis has been raised as a real possibility, on the basis of several recent publications (Saikku et al. 1988, Thom et al. 1992). Cytomegalovirus infection has also been suggested to be a factor in heart disease (Melnick et al. 1993, Zhou et al. 1996). The cause of heart disease, particularly CHD, is not as cut and dried as it may have appeared to be only a few short years ago.

In the meantime the general health picture vis-a-vis coronary disease keeps improving. The mostly good news is summarized in Table 5.

FOOTNOTES

1

Presented as part of the symposium “Evolution of Ideas about the Nutritional Value of Dietary Fat” given at the Experimental Biology 97 meeting, April 9, 1997, New Orleans, LA. This symposium was sponsored by the American Society for Nutritional Science. Guest editor for the symposium publication was Robert E. Olson, University of South Florida, Tampa, FL.

2

Supported in part by a Research Career Award (HL00734) from the National Institutes of Health.