Bodybuilding.net - Bodybuilding Forum - View Single Post

EricT · 05-31-2007, 02:58 PM

Research and the Mass Media--An Introduction
Department of Nutrition -Harvard School of Public Health

Over the past 100 years, great strides have been made in improving the health of people the world over. In the United States, for example, life expectancy has increased by about 25 years for men and 30 years for women, largely due to scientific advances that have helped prevent and treat many serious diseases.

The end results of science undeniably benefit us, but to people who don't work in the field, the scientific process can also be frustrating. This is particularly true for anyone trying to make important decisions about his or her health. What should I eat? How much exercise should I get? How often should I see the doctor for health checks? We all expect that science--specifically, health research--will provide answers to these very important and personal questions.

Unfortunately, it's often hard to get a straight answer. One day the "experts" say one thing. The next, they seem to say another. Then it all appears to switch back again. Such flip-flops can be maddening, especially when you're making your best effort to live a healthy lifestyle. Why go to the trouble of making big changes when today's highly recommended choice may be tomorrow's bad example?

The classic case of such flip-flopping in nutrition is the butter-versus-margarine question. More than 30 years ago, as research began to suggest that saturated fat might be bad for the heart, recommendations were made that encouraged people to switch from butter, which is high in saturated fat, to low-saturated-fat margarine. However, further research showed that, in many cases, margarine contained a type of fat (known as trans fat) that's even worse for the heart than saturated fat. This seeming about-face led many people to throw up their hands in frustration.

But the butter-versus-margarine question is also a well-known, if somewhat vexing, example of how research often works. Scientific research is a dynamic process that moves forward slowly. Recommendations are made based on the best science available at the time. However, with new research and new results, these recommendations may be revised.

On the other hand, such radical shifts in advice are rare today. There's far more research on diet and health available now than there was 30 years ago, when the original recommendation to switch to margarine occurred. This means that contemporary diet recommendations tend to be better-grounded--based on the results of many studies--and involve much less guesswork than in the past.

But contradictions in research results still occur. They are an inevitable part of the scientific process. Researchers are constantly performing studies and reporting their results. And when so many different people study each topic in so many different ways, it's natural that the results won't always be the same. What is the key, though, and what drives health recommendations is the weight of evidence on a particular topic--what all the results as a whole point to.

The research process is like placing stones on an old-fashioned balance scale. When enough weight accumulates on one side, the scale tips in favor of a particular recommendation. And the more weight there is on one side, the stronger the recommendation is and the more evidence it would take to change it.

If, on one side of the scale, you have over 40 studies showing that moderate alcohol intake can lower the risk of heart disease and, on the other, one or two studies that contradict those results, the scale would hardly budge. The weight of evidence would still be greatly in favor of moderate alcohol intake protecting against heart disease. Indeed, the link between alcohol and heart disease is so strong that it's known as an established relationship.

But not all topics are as clear-cut as this. Often, the weight of evidence is not as great. In some cases, only a handful of studies have addressed a particular question. In other cases, a large number of studies may lie on one side, but there may also be some particularly significant studies on the other side as well--just enough to cast some doubt. In these instances, we'd say that there is a probable link between a behavior and a disease. The value of eating moderate amounts of nuts to protect against heart disease is an example of a probable link.

There are also possible links, where the weight of evidence is still less and, in effect, the scale only tips slightly to one side. Possible links often develop in new, emerging areas of study, where a few studies have found a relationship, but more studies need to be done to confirm the results. A high intake of trans fat and an increased risk of diabetes is an example of a possible relationship that needs to be confirmed.

To carry this analogy further, the scale's likelihood of tipping reflects not only the number of stones placed on one scale, but also the size of those stones. Bigger stones will make the scale tip faster than smaller ones. Likewise, big, well-designed studies tend to play a more important role in establishing a relationship--and in shaping health recommendations--than smaller, less-well-designed studies.

Although the details can get complicated, large studies that follow human participants over time (randomized trials and cohort studies) tend to provide more reliable results than smaller studies that ask people about their past activities (case-control studies).

Different Types of Research Studies

There are many different types of research studies, and each has distinct strengths and weaknesses. In general, randomized trials and cohort studies provide the best information when looking at the link between a certain factor (like diet) and a health outcome (like heart disease).

Laboratory and Animal Studies

These are studies done in laboratories on cells, tissue, or animals. Laboratories provide strictly controlled conditions and are often the genesis of scientific ideas that go on to have a broad impact on human health. However, laboratory studies are only a starting point. Animals or cells are no substitute for humans.

Case-control studies

These studies look at the characteristics of one group of people who already have a certain health outcome (the cases) and compare them to a similar group of people who do not have the outcome (the controls). While case-control studies can be done quickly and relatively cheaply, they aren't ideal for studying diet because they gather information from the past. People with illnesses often recall past behaviors differently from those without illness. This opens such studies to potential inaccuracy and bias in the information they gather.

Cohort studies

These studies follow large groups of people over a long period of time. Researchers regularly gather information from the people in the study on a wide variety of variables (like meat intake, physical activity level, and weight). Once a specified amount of time has elapsed, the characteristics of people in the group are compared to test specific hypotheses (like the link between carotenoids and glaucoma, or meat intake and prostate cancer).

Though time-consuming and expensive, cohort studies generally provide more reliable information than case-control studies because they don't rely on information from the past. Cohort studies gather the information all along and before anyone develops the disease being studied. As a group, these types of studies have provided valuable information about the link between lifestyle factors and disease. Two of the largest and longest-running cohort studies of diet are the Harvard-based Nurses' Health Study and Health Professionals Follow-up Study.

Randomized trials

Like cohort studies, these studies follow a group of people over time. However, with randomized trials, the researchers actually intervene to see how a specific behavior change or treatment, for example, affects a health outcome. They are called "randomized trials" because people in the study are randomly assigned either to receive or not receive the intervention. This randomization helps researchers hone in on the true effect the intervention has on the health outcome.

However, randomized trials also have drawbacks, especially when it comes to diet. While they are good at looking at topics like vitamin supplements and cancer, when the change in diet is more involved than say taking a vitamin pill, participants begin to have trouble keeping to their prescribed diets. Such involved interventions can also become very expensive.

For example: Fiber and Colon Cancer--Following the Scientific Trail

Because science is such a dynamic process, you can never exactly tell where it is going to lead you. Conclusions that once seemed logical and fairly solid may be revised--or completely overturned--as more and better research is done on a particular topic. One example of this is the relationship between fiber and colon cancer.

Starting about 30 years ago, a high fiber intake was regularly recommended as one way to lower the risk for colon cancer. This recommendation was largely based on observations that countries with a high fiber intake tended to have rates of colon cancer lower than the rates found in countries with a low fiber intake.

But such descriptive studies don't provide the most definitive information. While they are often good points to start a scientific journey, they only take a broad look at large groups of people. Descriptive studies generally can't address all of the factors that might account for differences in rates of disease. Fiber intake could indeed have something to do with the differences in colon cancer rates, but those differences could also involve many other things that differ between countries, including other diet or lifestyle factors.

When studies that can take such things into account on an individual level began to look at the issue of fiber and colon cancer, the picture became much less clear. A number of case-control studies found that a high fiber intake was linked to a lower risk of colon cancer, but many did not. Given these wavering results--and because case-control studies are not an optimal way to assess food intake, relying as they do on participants' recollections of what they ate in the past--more research using better methods was needed. In the meantime, many health professionals still regularly recommended a high fiber intake for people trying to lower their risk of colon cancer.

Not until the results of cohort studies came out did this recommendation begin to lose its backing. Because cohort studies observe a group of people over time, their findings are generally stronger than those of case-control studies, especially when it comes to something like diet and colon cancer. What most of these cohort studies found was that fiber intake had very little, if any, link with colon cancer.

Such findings were further bolstered by the results of randomized trials--types of studies that many consider the gold-standard of research. These studies took a group of people and randomly assigned individuals to one of two groups. One group was put on a high fiber diet, while the other group followed a lower fiber diet. After 3-4 years, the two groups were compared and no difference was found in rates of colon polyps--noncancerous growths that can turn into cancer. Of course, colon polyps are not cancer, but since it's thought that all colon cancers start as polyps, it is strong evidence that fiber intake has no direct link with colon cancer.

In this case, the path of discovery led from widespread belief in a clear link between fiber and colon cancer to acceptance of the likelihood that there was no strong link between the two. As such, it's an excellent example of how research can often develop. What may start as a clear connection based on findings from broad, descriptive studies can slowly unravel as more and better-quality research unveils the true nature of a relationship. However, keep in mind that a weak relationship is difficult to exclude altogether. Further studies might yet demonstrate a weak effect of fiber on colon cancer, although such a finding wouldn't alter the conclusion that other means must be sought to prevent colon cancer.

Deciphering Media Stories on Diet

Even at its best, science is a painstaking, deliberate process, which doesn't fit very well into the cut-and-dry, newer-is-always better world of the mass media. And it's the media reports on health that are responsible for much of the frustration the public feels toward the public health community. With their emphasis on short, "newsworthy" pieces, the media often only report the results of single studies, and many stories are chosen simply because the results run contrary to current health recommendations. Because such reports provide little information about how the new results fit in with other evidence on the topic, the public is left to assume that, once again, the scientists screwed up and are now backtracking.

Fortunately, in many cases it only takes a few incisive questions to get at the heart of a research-related news story and see how important the results are for you personally. One of the most crucial things to keep in mind is the issue we've already discussed above: how a given study fits into the entire body of evidence on a topic. Whenever reading or watching a news story on health, keep these questions in mind:

*Are they simply reporting the results of a single study? If so, where does it fit in with other studies on the topic?
*Only very rarely would a single study be influential enough for people to change their behaviors based on the results.
*How large is the study?
*Large studies often provide more reliable results than small studies.
*Was the study done in animals or humans?
*Mice, rats, and monkeys are not people. To best understand how food (or some other factor) affects human health, it must almost always be studied in humans.
*Did the study look at real disease endpoints, like heart disease or osteoporosis?
*Chronic diseases, like heart disease and osteoporosis, often take many decades to develop. To get around waiting that long, researchers will sometimes look at markers for these diseases, like narrowing of the arteries or bone density. These markers, though, don't always develop into the disease.
*How was diet assessed?
*Some methods of dietary assessment are better than others. Good studies will have evidence that the methods have validity.
*With these tips and a better understanding of the world of health research, you can look at health information with a more discerning eye. While this won't be a guarantee against frustration, it should help you embrace health recommendations--and the healthy lifestyle they promote--with more confidence.