Dr. CLEAVE. Quite so.

Senator McGOVERN. Could you explain that, in layman's terms, how nutrition affects the quality of blood?

Dr. CLEAVE. Well, the bloodstream-apart from the question of gases in the air, which is of no importance in this connection whatever the bloodstream reflects what you are eating. It all goes into the bloodstream, all the food absorbed from the gut. The bloodstream absolutely reflects your nutrition.

I therefore believe that if you want a natural bloodstream-which is the natural environment of the heart, this miraculous organ-you will have to eat natural materials. It is as easy as that.

I strive at all times to keep it very, very simple, especially in the document in your possession.

Senator McGoVERN. There is another phrase you used that I was intrigued with. You refer to "energy imbalance" resulting from overconsumption. Can you further define for us, what you mean by "energy imbalance"?

OVERCONSUMPTION RESULTS IN THE IMBALANCE

Dr. CLEAVE. Yes, I will. We will assume that a man is quite tired after a bit of work and he wants to renew his energy. One of the easiest ways of doing that is to eat something sweet.

Now, if he were to eat half a melon, shall we say, that will be done safely and correctly as nature ordained it. But now let us suppose, for example, that he ate a bar of chocolate instead. It is absolutely certain owing to the extreme concentration in the chocolate, that the tongue will not be able to tell him exactly where to stop. The way it would if he was eating the melon.

I said half a melon. He might eat three-quarters. He might eat a quarter. But he would know when to stop because it is bulky and dilute.

Now, if he eats a bit of sugary stuff which is possibly 10 times more concentrated, the tongue is no longer calibrated to tell him where to stop. He will almost certainly finish the bar, and he will probably take into his body anything up to 50 percent more sugar than he would have done in the case of a melon.

Now we have the energy imbalance before you. He wanted half a melon. He has ended up with three-quarters of a melon in the form of chocolate, and he has the imbalance of a quarter of a melon's load of energy which he would never have taken in under natural conditions. Senator McGOVERN. Dr. Cleave, just one more question I was concerned about. In your prepared testimony, you refer to the arguments by Ansel Keys that you say contradict your theory. Doesn't Keys point to some countries that have a high sugar consumption, but a low incidence of diabetes and heart disease?

Dr. CLEAVE. He does indeed. I would first of all point out that he lumps in these high-sugar consumption countries where the sugarcane is grown. Where admittedly a very large quantity of sugar is eaten, but mostly in its natural form. It makes a tremendous difference.

DISTINCTION IN FORMS OF CARBOHYDRATES

Second, he doesn't distinguish between all forms of carbohydrate consumption. You see, it would be quite easy in this hall to assemble about 20 of the fattest people in Washington, and they wouldn't have a sweet tooth among them, and they wouldn't like sugar.

What were they or what are they, these 20 people? Beer drinkers? You must consider all forms of sugar consumption, and all other forms of refined carbohydrate consumption. You can't possibly take only one, you have to consider the lot. When you consider the lot, you get a fundamentally different answer. Beer is full of malt sugar and enormously fattening.

We showed that in Trinidad. I went out to Trinidad because I know the place, and it was reported there that sugar in the Negroes was less than in the Indians, and yet the Indians consumed more sugar than the others. It was exactly the opposite of what you would expect, in regards to diabetes.

But, when we looked into the matter and found out what they were taking in the form of white flour, the difference was immediately obvious. Taking sugar and white flour, the relationship to diabetes of the two together, came out correct.

You have to consider all forms of refined carbohydrate consumption. You can't take a mountain of a man weighing 20 stone sitting therewithout having any interest in sweets, but drinking anything up to 10 pints of beer a day-you can't take that man and say, "There you are, there is a fat man, an enormously fat man who isn't touching sugar." You end up with a "reductio ad absurdum." You have to consider all refined carbohydrates another crucial point. Dr. Keys doesn't allow for the incubation period; some of these high-sugar consumers haven't coronary disease. No, not yet. We, in Britain, had very little in 1890-when the sugar consumption first reached to the 100-pound-a-year level. But, 30 years later, the coronary disease explosion took place in 1920. Dr. Keys made no allowance whatever for the incubation period-the timelag. Twenty years ago the Eskimos had a high sugar consumption-but no diabetes. Now they have plenty of it. Dr. Keys makes a huge error by not taking this incubation period into account.

Senator McGOVERN. Do you have any other points you would like to make in conclusion, Dr. Cleave?

Dr. CLEAVE. No, I was wound up for 20 minutes, Mr. Chairman. Senator MCGOVERN. Thank you so much. This was very fascinating testimony.

Dr. CLEAVE. Thank you very much.

Senator McGOVERN. The hearing tomorrow will be in room 1202, in this building, and it will begin at 10:30 instead of the usual time of 10 a.m.

The committee is in recess, to reconvene on Wednesday at 10:30 a.m. [Whereupon, at 12:25 p.m., the Select Committee was recessed, to reconvene at 10:30 a.m. on May 2, 1973, in room 1202 of the Dirksen Building.]

NUTRITION AND DISEASES

Sugar in the Diet, Diabetes, and Heart Diseases

WEDNESDAY, MAY 2, 1973

U.S. SENATE

SELECT COMMITTEE ON

NUTRITION AND HUMAN NEEDS

Washington, D.C.

The select committee met at 10:30 a.m., pursuant to call, in room 1202 of the Dirksen Building; the Honorable Richard S. Schweiker presiding.

Present: Senator Schweiker.

Staff members: Kenneth Schlossberg, staff director; and Vernon M. Goetcheus, senior minority professional staff.

Senator SCHWEIKER. The Senate Select Committee on Nutrition and Human Needs will please come to order.

OPENING STATEMENT BY SENATOR SCHWEIKER, PRESIDING

I would like to welcome you to this third and final day in a series of hearings on the relationship between diet and health. I personally would like to take this opportunity to thank all the witnesses who have given their valuable time and traveled such long distances to make these hearings possible.

Our first witness today is Dr. Peter Bennett. Dr. Bennett is the chief of epidemiology and field studies at the National Institute of Arthritis, Metabolism, and Digestive Diseases in Phoenix, Ariz. He is a leading expert on the occurrence of diabetes and has been engaged in extensive research on the Pima Indians of the Southwest to try to determine factors relating to their high incidence of diabetes.

Our second witness is Dr. Arthur Steinberg. Dr. Steinberg is the Herrick professor of biology at Case Western Reserve University. He is one of the Nation's leading geneticists in the area of diabetes. Dr. Steinberg has developed estimates generally agreed upon that 20 percent of the population may be genetically vulnerable to diabetes.

Our third witness is William E. Dulin. Dr. Dulin is the manager of diabetes and atherosclerosis research at the Upjohn Co. in Kalamazoo, Mich. He is one of the world's leading animal researchers in the area of diabetes with a special concern for identification of vulnerable individuals in infancy and for dietary treatment to prevent the disease from developing or to modify its severity.

So we have a very distinguished panel of experts in this area. I am very pleased to call then on our first witness scheduled today, Dr. Peter Bennett. Dr. Bennett, would you please come forward.

STATEMENT OF DR. PETER BENNETT, CHIEF, EPIDEMIOLOGY AND FIELD STUDIES BRANCH, NATIONAL INSTITUTE OF ARTHRITIS, METABOLISM, AND DIGESTIVE DISEASES, PHOENIX, ARIZ.

Dr. BENNETT. Mr. Chairman, members of the committee, it is a pleasure to be here in response to your request to testify on the relationship between diabetes and obesity.

My interest in the relationship between obesity and diabetes arose from the discovery that the Pima Indians, a tribe of American Indians residing on the Gila River Indian Reservation in central Arizona, have the highest prevalence of diabetes mellitus yet recorded in the world. Among the Pima Indians the prevalence of diabetes is 50 percent in those aged 35 years and over. Other desert dwelling American Indians including the Papago, Maricopa, and Cocopah Tribes also have diabetes frequencies close to that of the Pima, which is 10-15 times greater than that reported for the United States as a whole.1

OBESITY SIGNAL OF DIABETES-SUSCEPTIBILITY

The Pima Indians are mostly obese, and, on average, they weigh 50 percent more than their desirable body weight. The prevalence of diabetes is greater in the more obese Indians, especially in the younger adults, suggesting that the role of obesity may be to precipitate diabetes at a relatively early age among those who are genetically predisposed to the disease.

Detailed dietary studies conducted in the 25- to 44-year-old Pima females have shown that their calorie intake is excessive, averaging close to 3,200 calories per day compared to a recommended daily average of about 2,000 calories. No other peculiarities in dietary intake responsible for the excessive frequency of diabetes have been found, although at the time of the study the diabetics had a slightly lower carbohydrate and sugar intake than the nondiabetics-presumably representing dietary changes following the onset of diabetes, rather than being the cause of the diabetes.

We have postulated that the high frequency of diabetes among the Southwestern Indians is probably a relatively recent development, as has been suggested in other populations with presently high or presently increasing frequencies of diabetes, such as in the urbanized Indian in South Africa or his aboriginal counterpart in Australia.

3

It has been suggested that in populations where food supplies may be intermittent, with relatively frequent periods of starvation, that the prediabetic state, presumed at this time to be determined by hereditary or genetic factors, may be advantageous. Such persons genetically predisposed to diabetes may be better able to store food as adipose or fat tissue and conserve glucose than others. This capability is associated with an exaggerated insulin secretion from the pancreas. Under continued stress from an excessive amount of food over a prolonged time period, the pancreas may be unable to keep up with the production of

2 Bennett, P. H., Burch, T. A., and Miller, M.; Diabetes Mellitus in American (Pima) Indians. Lancet 2: 125-128, 1971.

2 Reid, J. M., Fulmer, S. D., Pettigrew, M. A., et al: Nutrient Intake of Pima Indian Women Relationships to Diabetes Mellitus and Gallbladder Disease. Amer. J. Clin. Nutrition 24: 1281-1289, 1971.

3 Neel, J. V.: Diabetes Mellitus: A "thrifty" genetotype rendered detrimental by "progress". Amer. J. Hum. Genetics 14: 353-362, 1962.

the excessive amounts of insulin required to metabolize the foodstuffs. Then glucose intolerance and diabetes, which are the result of a relative or absolute deficiency of insulin, may supervene.

If the predisposition to diabetes were advantageous in times of starvation, over many centuries the frequency of the diabetic gene would increase within the population. If that population were then exposed to constant or excessive food supply, diabetes mellitus could appear within the lifespan of a single generation and produce its severe manifestations and complications, which would be then deleterious to the population group.

This hypothesis could explain the high frequency of diabetes among the Southwestern American Indians and other peoples, and also explain the high frequency of spontaneous diabetes which occurs among a number of mammalian species when brought from their natural environment and placed in one of constant or excessive food supply in the laboratory.

The high prevalence of obesity at the onset of diabetes in human beings is universally recognized, and the incidence of overt diabetes among the obese with mild carbohydrate intolerance is extremely high. It is yet unknown, however, whether diabetes can be produced by overeating or obesity alone, or whether a genetic predisposition to diabetes is a sine qua non for its development. Since there is at present no certain way to identify the subject with a genetic predisposition to diabetes before he develops the disease, this question remains unanswered, but clearly those with a family history of diabetes and who are also obese are at greater risk of the development of overt diabetes than those who are not obese. This is particularly true of the offspring of two diabetic parents.

CALORIE RESTRICTION MEDICALLY DESIRABLE

In most obese adult patients, with diabetes of recent onset, calorie restriction can result in improvement or even normalization of carbohydrate tolerance. This suggests that if these same subjects had earlier restricted their calorie intake, and hence avoided becoming obese, the onset of diabetes would at least have been delayed, and quite possibly the disease would have been prevented.

The possible effects of regulation of food supply on diabetes on a national basis are well illustrated in data from England and Wales. During the food rationing periods in the 1914-18 war and in the Second World War diabetes mortality fell by 25 percent and 40 percent respectively. After each period, the relaxation of the restrictions produced by food rationing were accompanied by a slow, but steady return to the preexisting mortality levels over the subsequent 15 years. It is difficult to attribute these changes to anything other than the reduction of calorie intake to a medically more desirable level.

In the United States, mortality rates attributable to diabetes have increased steadily over the past 20 years. In part, this increase is attributable to the aging of the population, but even after adjustment for this, the death rates directly attributed to diabetes increased by approximately 10 percent in white males and in nonwhite persons by approximately 50 percent between 1954 and 1967.

1 Himsworth, H. P.: Diet in the aetiology of human diabetes. Proc. Roy. Soc. Med. 43: