NewsMaker Interview- Epidemiologist Dr. Charles Hennekens

Our first interview (November 1994) is with noted epidemiologist Charles Hennekens, M.D., Professor of Medicine, Harvard Medical School, and Chief, Division of Preventive Medicine, Brigham and Women's Hospital, Boston, MA.

We are bombarded with endless, sometimes conflicting, media reports announcing that this food is good for you or that activity is bad for you. Where do these claims originate and how much scientific merit do they have? For example, most of us have heard that low doses of aspirin taken over the long term significantly reduce the risk for heart attack. Dr. Hennekens is the director of the Physicians' Health Study, the large long-term study which demonstrated conclusively that aspirin did reduce the chance of a first heart attack in middle-age men. Another part of that trial is evaluating the potential role for the antioxidant vitamin beta-carotene in protecting against heart disease and cancer, and will be completed next year. I caught up with Dr. Hennekens at the annual meeting of the American Heart Association in Dallas and asked him to elaborate on the science of public health.

Q: Epidemiology involves studying large groups of people for clues into patterns of disease and potential causal factors. This seems at first glance to be far removed from basic science in the laboratory. How are basic science, clinical research and epidemiology connected?

A: Advances in scientific knowledge proceed on several fronts, optimally simultaneously. The basic researcher working in the lab doing in vitro and in vivo work provides us the crucial answers to the unique question of why an intervention reduces premature death and disability. Clinicians and nurses provide enormous benefits to patients by applying advances in treatment and diagnosis and by formulating hypotheses from their clinical experiences. The epidemiologists and statisticians formulate hypotheses from descriptive studies, correlational studies, and cross-sectional surveys and test them in observational cohort or case-controlled studies, as well as randomized trials when needed. These studies answer the crucial and complementary question of whether an intervention reduces premature death and disability.

All of these scientific disciplines provide important, relevant and complementary information contributing to a totality of evidence. At any point in time it is this totality of evidence that forms the basis for rational clinical decisions for patients, as well as policy decisions for the health of the general public.

Q: Could you elaborate on that point using the Physicians' Health Study as an example? A: The Physicians Health Study is a randomized trial designed to determine if aspirin was effective in reducing the risk of cardiovascular disease and if beta-carotene was effective in reducing the risk of cancer and cardiovascular disease. The study enrolled 22,071 male physicians, half of whom received aspirin and the other half placebo. In addition, each participant was randomly assigned to receive beta-carotene or placebo.

The aspirin component of the study was terminated early because of an observation that those receiving aspirin had an extreme, 44% lower risk of first heart attack than placebo recipients. The beta-carotene study continues until next year.

Q: Tell us how you formulated the hypotheses for this study. What made you think aspirin would reduce heart attacks in the first place?

A: There was a body of basic research indicating that low doses of aspirin inhibited the tendency of blood platelets to stick together, and by doing so might decrease coronary thrombosis, the proximate cause of virtually all heart attacks. There were also clinical observations of patients who had already survived a heart attack showing that aspirin lowered the risk for subsequent heart attack. This raised the question of whether healthy people might also derive a benefit from aspirin.

Q: What was the underlying hypotheses for evaluating beta-carotene?

A: Beta carotene is the vegetable form of vitamin A found in carrots, fruits and green leafy vegetables. Beta-carotene has the remarkable property of both trapping free radicals and quenching singlet oxygen, a byproduct of normal metabolism. Singlet oxygen is an excited molecule that may enhance carcinogenesis and possibly atherogenesis. So, by acting as an antioxidant, beta-carotene might break a chain of events leading to the development of cancer. Furthermore, there was supporting data from observational epidemiological studies suggesting that people whose diets were higher in beta-carotene-containing fruits and vegetables may have lower risks of cancer and heart disease.

What we don't yet know is whether it is the beta-carotene itself or other nutritional components of the diet or other elements of the lifestyle which lower the risk. The randomized evidence we will get from the Physicians Health Study in 22,000 men as well as the Women's Study involving 40,000 female health professionals will provide us with some firm answers to these questions.

Q: A lot happens in the science world during the five to ten year period from when a large study begins to its completion. How do you integrate new findings that come along during the course of a study?

A: We shouldn't change the design of studies already underway. But new discoveries can change our thinking. For example, our early observations in 333 men with chronic stable angina in the Physicians Health Study suggested that those randomized to beta-carotene had a lower risk of subsequent heart disease. This raised the question of whether we should investigate the possibility that beta-carotene could reduce the development of a first heart attack, and we will be looking for that. New scientific findings can help us understand the mechanisms of the effects we are observing in the epidemiological studies. In addition, new scientific discoveries can provide research methodologies to test hypotheses in future studies.

Q: The public can get very confused with some studies and clinical trials saying this is good for you and others saying that is not good for you. What can be done to reduce this confusion?

A: The public is understandably confused. Academic researchers tend to over report the results of any one finding, and the media does the same. What we have to do is to educate the public to qualitatively understand that no one study will change the totality of evidence. It might move us a little bit in one direction or another, but we need to understand that a large number of different studies from different places done by different researchers is needed to understand fully what's going on. There is no quick fix on these questions, we need to consider the totality of evidence from all sources in order to make the most rational decisions.

Q: Epidemiologists and geneticists often seem to be approaching the same problems, but from different sides. Do epidemiological studies help identify genetic predispositions and does identification of genes lead to new epidemiological hypotheses?

A: I think so. There are clear and important interactions between genes and environment. In genetic studies we try to control the environment and look for the genetic contribution. In epidemiology we try to control genetic variability by using large sample sizes and randomized allocation and then look for environmental determinants. There is a close relationship and each discipline contributes important, complementary information.

We want to identify genetically susceptible subgroups of people. Since we are, as yet, unable to change genes, we need to identify the environmental determinants of the disease which are modifiable. Ultimately, new improvements in molecular genetics suggest there may even be ways to alter the genetic make-up directly.

Q: Do you have any advice for future scientists?

A: Young future scientists ought to get a broad base of academic knowledge of biology, chemistry and physics and not specialize too quickly. When they do decide which track to pursue they will have a broad scientific base that will allow them to understand the contributions from different disciplines and how important they are. We need contributions from all of the scientific disciplines in order to get the kinds of evidence we need to make more rational clinical decisions for individual patients as well as policy decisions for the health of the general public.

Newsmakers Index

What's News Index

Feedback