Most Americans care about the food they eat. They want to eat a healthy diet. But what is a healthy diet?
Sadly, the answer to that question remains a mystery. Federal dietary guidelines were first issued in 1980 to curb the rising incidence of heart disease, obesity and diabetes. But current measurements of those diseases show little or no progress. A revised set of recommendations was released this month that includes a new cap on added sugar but this is unlikely to have much additional impact.
Steven E. Nissen, chairman of the cardiology department at The Cleveland Clinic, and Nina Teicholz, author of “The Big Fat Surprise”, tell us there may be hope for improvement. Writing in The Wall Street Journal, these authors tell us the problem is the government relying on weak science. But Congress recently mandated the first-ever outside review of the evidence underlying the dietary guidelines and the process that produces them.
The National Academy of Medicine will conduct the review this year. But the quality of their review and the credibility of their recommendations depends upon the quality of the science that backs their conclusions. The best scientific evidence comes from clinical trials where conditions are controlled to exclude bias and the variables being measured. But this is difficult in dietary studies because subjects must be closely monitored, even provided all of their diet, to avoid introducing contamination into the study.
As a result, most dietary studies instead rely on observations using a scientific method known as prospective epidemiology. Researchers send out questionnaires to large numbers of people, asking about diet and lifestyle. Then they follow these people for years and record their health outcomes.
This is poor science, at best. These kinds of studies can only show associations, not causations. For instance, obesity might be associated with sitting in front of a television set all day. But these same people might also be eating junk food. Does their obesity come from eating junk food or the lack of exercise? Or is there some other unknown variable that is the real cause of their obesity?
Nissen and Teicholz note that of the enormous number of associations generated by observational studies, only a small number are ultimately confirmed. In 2005, John Ioannidis of Stanford University analyzed several dozen highly cited studies and concluded that subsequent clinical trials could reproduce only 20% of observational findings.
A 2011 paper published by the statistics journal Significance analyzed 52 claims made in nutritional studies, and none – 0% – withstood the scrutiny of subsequent clinical trials! Yet policy makers often rely on such studies to make recommendations for our health.
Not surprisingly, conclusions based on such shoddy scientific research has led to some significant flip-flops on dietary recommendations. Epidemiological data suggested cholesterol might be linked to heart disease, and fat to cancer. The result was decades of physicians recommending avoidance of egg yolks and shellfish. Millions of Americans adopted low-fat diets and ate more carbohydrates. Today, a large body of evidence suggests that eating excessive carbohydrates increases the risk of heart disease, obesity, and diabetes.
Rates of obesity and diabetes remain high, despite federal guidelines aimed at reducing these diseases. But not because the advice is ignored. A 2008 report by the Agriculture Department estimated changes in food consumption form 1970-2005: grains rose 41%; vegetable oils rose 91%; fish and shellfish rose 37%; vegetables rose 23%; and fruits rose 13%. Eggs and red meat both fell 17% and whole milk fell 73%! Yet during this same period the incidence of diabetes doubled.
Clearly we have much to learn about nutrition and the healthiest diets. Fortunately, Congress appropriated $1 million for an independent review of the federal dietary guidelines. New guidelines are expected to come out in 2020 and hopefully these will be based on improved scientific evidence. To ensure unbiased reporting, Congress has asked several members of the 2015 Dietary Guidelines Advisory Committee who are also members of The National Academy of Medicine to recuse themselves from this review.
The public deserves to know the best advice on healthy diets from unbiased, scientific studies – and unbiased, politically independent observers.