A. Gerald Shaper

Quotes

First, there were a number of things going on when I was a medical student that really did affect my way of thinking about medicine. The political situation was very obvious and I was interested in not just clinical medicine, but I was interested in medicine in the broader community. We had a very large non-white community called Colored People and there was the beginnings of the migration of African black people from other parts of South Africa into Cape Town. Interracial studies were starting and I was very conscious of studies going on in the White, Colored, and African communities in Cape Town at that time. They focused, through people like Bronte-Stewart and others, on cardiovascular disease. The key issues emerging were nutritional differences between racial groups. The possible effect this might have on disease outcomes was made paramount even to medical students.

We are talking about the late 1940s. It was about the time after World War II that malnutrition was becoming of international interest and WHO had begun to bring out reports in the late 1940s by Brock, Autret and others on nutrition. So nutrition, not only in terms of malnutrition, but in terms of cardiovascular disease was of interest. At the same time, there were coagulation and hematology studies going on by Clarence Mersky, who later went to the States. Again, the issues were always comparative, between the different racial communities which, of course, were different socio-economic and nutritional communities, not just racial. So nutrition, cardiovascular disease, hematology, coagulation and all of this was going on when I was a medical student.

But the nutritional issues affected me most because it had a political slant to it. And as I was very politically involved at the time, these were the aspects that mattered to me. (1)

Diabetics without Coronary Disease

I started with diabetes and that initiated my research in East Africa. It had considerable cardiovascular implications later. So I was running this diabetic clinic and over the years it became apparent as the clinic grew larger and larger, that here I had people with advanced neglected diabetes but no coronary heart disease. This was an African community where coronary heart disease was extremely rare. I mean so rare that a single case in an African judge had been written up in a medical journal.

I eventually did electrocardiograms on several hundred African diabetics, most of whom had had neglected diabetes for decades and yet there was no coronary heart disease. Some retinopathy, a lot of hypertension and obesity, but no cardiovascular disease of the atherosclerotic kind. And it fitted in with what I was beginning to feel about nutrition and cardiovascular disease, that unless you had a particular nutritional and lipid background, you were not subject to coronary heart disease even if you were a diabetic. (6)

So I have two issues [ed. statistical adjustment and attributable risk] that irritate me the whole time and I feel that it is very important for biological scientists not to become mechanistic scientists. I’m making a criticism of our non-biological, non-medical colleagues because they think that public health is too important to be run by doctors. However, there is an aspect of public health that requires understanding of the biological processes inherent in it.

African Cholesterol Levels

But very early on, when I arrived in Uganda, I was also very concerned to do work relating to the blood cholesterol story that had emerged from the Seven Countries Study … In 1958 or 1959 I did a study in African and Asian children aged 10, in African and Asian subjects in their middle 20s, and those aged 40 to 60. I was looking at their blood cholesterol in particular and I wanted to know what was happening to these two different communities as they aged. The Asian community in Kampala had a high rate of coronary heart disease and also had diabetes. In fact they had all the Western-pattern diseases while the Africans had no coronary heart disease at all. And what was interesting was that at the age of 10, remember we are now talking about the late 1950s when we were making this observation there were striking differences between the blood cholesterol levels in the African children who were averaging about 120 mg/dl and the Asian children who were averaging about 160mg/dl. . . And, of course, as Africans got older their blood cholesterol levels did not rise. But as Asians got older, their levels rose considerably.

What was most interesting was the response of my British mentor, Sir John McMichael, who was very opposed to the whole cholesterol hypothesis and who felt that this kind of work was a betrayal of our relationship… He was very opposed to the “American cholesterol hypothesis,” which was prominent at that time. And, of course, the Seven Countries Study was then starting. (9)

But I came back to Kampala even more enthused by the impact of the Seven Countries philosophy than before. So, when I went back to Uganda we began to have visits from WHO, Zdenek Fejfar in particular. And he encouraged us into the possibility that there was comparative work that could be done in Uganda that probably could not be done anywhere else. We had an African community with no coronary heart disease. We had an Asian community with a lot of coronary heart disease. And we had a selected European community with as much coronary heart disease as you get in other European communities. But young, of course, because they were selected by the work they were doing. So, I began to do two things. I continued work in the blood lipid area and I began to work in the coagulation and fibrinolysis area. . . before most other people were working in it. (10)

Coagulation

And in the early 1960s we published the finding that when you compared Africans and Asians, again at different age groups, one of the most striking findings was the difference in fibrolytic activity; the Africans could break down clots very actively but the Asians had striking inhibition of the fibronolytic mechanism. I also looked at platelet adhesiveness and many other coagulation phenomena and it was interesting and exciting, but nobody elsewhere was very much interested. This appeared to be a great problem that few workers were interested in.

As I went further and further into the fibrinolytic and platelet behavior and coagulation side, I think I made a major mistake. What I realized from the work we were doing was that a tremendous amount of what was going on in coagulation, lysis, and platelet behavior, was related to the blood lipid environment in which coagulation, lysis and so on was taking place. So I reasoned that fibrinolysis and the coagulation side of it was secondary to the nutritional aspects. So if one really wanted to do fundamental research, one continued on the nutritional and blood lipid side rather than on the coagulation side. I missed the possibility that you could interfere pharmacologically in these intermediate mechanisms. It did not cross my mind at that time. It was an opportunity missed. (11)

Shepherds’ Exotic Diets

The other thing I became aware of was that in parts of Uganda and Kenya there were people on odd diets. There were the Masai-speaking groups who had a milk, meat and blood diet. And I speculated that if these people had such a diet and if the diet-heart hypothesis is correct, then they ought to have high blood cholesterol level and they ought to be subject to atherosclerosis and coronary heart disease. So over a period of two years, we borrowed a Wellcome Trust mobile laboratory belonging to the physiology department. I brought a group of medical school colleagues together and we went out in the field and for two years did studies amongst the nomads in northern Kenya, on the border with the Sudan. So it was great stuff, very exciting and great fun…

First, that amongst these nomadic people on diets of milk, meat and blood, blood cholesterol and blood lipids in general, remained low and level throughout life. Blood cholesterol concentration at 20, 30, 40, 50, and 60 years of age, stayed the same. We had an anthropologist who worked with us, Paul Spencer, who could estimate their ages from recalled life events. As far as one could tell, there was no rise in cholesterol with age despite this diet. There was no increase in body weight with age either. There was no increasing adiposity or blood pressure with age. These findings hit me very hard in the sense of realizing that it was not biologically normal for blood cholesterol to rise with age. In fact, the biologically normal state was a steady lipid level throughout life, and the biological normal was a steady weight throughout life, and a steady blood pressure throughout life. Which is our [evolutionary] heritage. But that concept was not around at that time and it struck me very forcibly.

Cattle versus Camel Herding

But one odd thing turned up. Most of the nomads we worked with were cattle herders and I decided that camel herders were of interest because camels produced a more copious and richer milk and for longer periods of the year. So I thought that people who were camel herders might be different from people who were cattle herders. We carried out one study where we took blood samples from three different tribal groups, two that were cattle herding and one that was camel herding. They were examined in the same laboratory in random order, so nobody knew which samples they were examining. It was quite astonishing. The cattle herders, as expected, had absolutely flat cholesterol levels throughout life. But the camel herders had exactly the same pattern that one would have seen in a Western community. They rose with age to about the 40s and 60s, which was quite surprising. Again, this was a group with no hypertension. These were Rendille, Masai speaking and closely related to other Masai-speaking tribes. They were only slightly different in their geographic origins.

So again, it showed that you could have a community that had high blood lipids but still did not have coronary heart disease, because they had none of the other aggravating factors. The raised blood cholesterol was present, but it was not sufficient to produce atherosclerosis. They were lean, had no hypertension, were physically active, and did not smoke. So again, it was another interesting aspect of these groups. (13)

Rural-Urban Weight and Blood Pressure

When I came back to work in Kampala after these nomadic studies, what struck me was that we had a tremendous amount of hypertension in our local community in Kampala and in the areas around it. We also had a lot of overweight and obesity. It was a rural-urban difference. So we did studies in blood pressure and body weight in the peri-urban communities in the villages around Kampala. And it was quite clear that blood pressure rose with age exactly the same way as it did in Western communities, even though there was no coronary heart disease.

Again, you could have the same patterns of blood pressure as you had in Western communities but you did not have coronary heart disease, presumably because their blood cholesterols on a high carbohydrate diet were relatively low. It kept coming back to the issue that nutrition was absolutely fundamental to the atherosclerosis-coronary heart disease problem. To my mind, it was the essential prerequisite without which you could not have coronary heart disease no matter how many other aggravating factors you had.

Of course, at that time the Seven Countries Study was beginning to show us that the Japanese could smoke and be hypertensive, the Greeks could smoke heavily and be hypertensive, and both populations still had low rates of coronary heart disease. (15)

Statistics versus Biology

I think that two things still bug me about current attitudes to coronary heart disease and to vascular disease in general.

One is the power of the statisticians to overwhelm us with technique, and I talk now about two things in particular. One is adjustment and the other is attributable risk. Many of the people working in the area of attributable risk, and I know that WHO is deep into the issue at the moment because it has huge political implications, have no concept of the true nature of atherosclerosis. They think that all factors can be expressed simply in terms of a percentage of attributable risk. . . They do not think in terms of a biological model. For example, in coronary heart disease, blood cholesterol, smoking, blood pressure, all of these they consider to have equal weight and a percentage contribution. That is not how the biology of vascular disease works. . . . I’m worried about the whole process of attributable risk. The fact that we make major policy decisions on alcohol, on diet, on prevention of smoking, in terms of attributable risk worries me. I do not understand how biological scientists have managed to be conned by this or how WHO has managed to accept it without question.

The second thing is adjustment. We do adjustments in order to avoid the whole problem of confounders… It ignores pathways and ignores the fact that almost everything we adjust for is either poorly measured, is measured only once, or is measured inaccurately, or is only part of the process. For example, we put in obesity or body weight or BMI as an adjusting factor. Yet it is one of the most complex issues. Fat is one of the most complex organs in the body.

So I have [these] two issues that irritate me the whole time and I feel that it is very important for biological scientists not to become mechanistic scientists. I’m making a criticism of our non-biological, non-medical colleagues because they think that public health is too important to be run by doctors. However, there is an aspect of public health that requires understanding of the biological processes inherent in it. And although many statisticians and non-medical epidemiologists acquire a tremendous understanding of the processes, there is a tendency, particularly for new people, young people coming into the business to see it almost as a technical method. That all you do is adjust and you attribute things and that gives you the whole story. I have to say that we have not come to the point where we can really be at peace with this. (23)

The Correlation of Coronary Disease and Water Hardness

HB: I’ll tell you what Stoney Stallones said about that: ‘The good Lord doesn’t make us aware of the important issues for mankind with a correlation of 0.2!’ (24)

Then there’s genetics and health inequities

I’m concerned about molecular epidemiology because it’s ignoring what we already know in simple terms, and it is pursuing pathways which are intrinsically important in themselves, but will contribute little to the prevention of cardiovascular disease. . .or make healthy populations. We are beguiled by this. It’s the moons that dazzle us. We love them.

Molecular biology and genetics. Even in this country, you cannot do a study without focusing on inequalities. And the implicit promise is that we will be able to obliterate inequalities! There are people and groups who focus on inequalities with the presumption that this is the answer to our lifestyle problems.

I’m not saying that the issue of social inequalities isn’t important. But in terms of research into cardiovascular disease the inequalities issues is not fundamental. For example, we have socio-economic differences and social class differences in coronary heart disease. Yet when you begin to take into account their lifestyle differences, these socio-economic differences disappear. (26)

Public Health Conservative Policy in a Labor Government

I think that once you accept that nutrition is a fundamental issue, then you’re into major conflict with your national policy on nutrition, your dietary policy, your industrial policy, and marketing. Industry is what runs the political parties. So this is too big a concept to accept readily. We have a National Heart Forum which is an organization of more than 40 organizations, all having something to do with cardiovascular disease. The National Heart Forum recently issued a consensus statement, but did so only indirectly. You probably have not seen their recent report. It is titled: “Looking to the future: Making coronary heart disease an epidemic of the past.”

There is a chapter on causes of coronary heart disease. I was chairman of this chapter group and managed to get what I wanted to say into it. We looked at all the “new risk factors” and found that they made no significant additional contribution. The contributing authors included George Alberti who has been the President of the Royal College of Physicians, Michael Marmot and David Barker, and they were all prepared to put their names to it, They all saw the full document and they all contributed sections to it. But the problem is that this document like most documents, is now filed away. It circulated to the people who were interested and then it went on a shelf and that’s it.

The government committee on Medical Aspects of Food Policy (COMA) has been disbanded because it was too radical and we now have the Food Standards Agency, which is much more politically influenced. So the group of experts who used to meet and put forward documents which were really quite radical – Jerry Morris sat on a lot of them – are gone. (30)

The Ancel Keys video documentary

Looking at this video I’m on the edge of tears because of the deep sense of gratitude toward Ancel [Keys], an awareness of the integrity and vigor that went into the thinking behind all this and the impact that it had on generations of scientific and medical workers, and on the public health.

I regret that the lag time between all of this happening in the United States and in the United Kingdom is a lag period of something like 20 years. Even today, what you accept as the norm for the understanding of the dietary hypothesis of heart disease is still not firmly established in the hearts and minds of people and scientists in this country.

I think the public are ahead of the game in the sense that they know that there’s something centrally important about nutrition in this problem. But amongst the medical community, there is still the concept that this is what we call multifactorial. Now, of course, everything is multifactorial, but it’s multifactorial in the sense that any risk factor has equal value in producing coronary heart disease and that concept has still not been eradicated in this country… there is an essential (necessary) element which is nutritional and without it there is no coronary heart disease, even though the other risk factors will effect changes that are deleterious to health. But that message has really not gone through deeply to the United Kingdom, scientific or public. I can’t understand why. (34)