A owner of a natural food store is cryogenically frozen and rekindled two centuries later. To his surprise, he learns that the steak, the cream pies and the hot fudege that he avoided once because unhealthy turned out to be anything but. The film, of course, is the “sleep” of Woody Allen in 1973, and its cinematic shipment highlights a serious problem which has long haunted biomedical science and especially research in nutrition – namely that a surprising number of conclusions are based on very thin evidence, and many are not only unreliable but false.
Some of these problems have been highlighted in a recent Law and freedom Theodore Dalrymple test entitled “The fraudulent laboratory. “Scientific dishonesty is a real threat to the credibility of research, but fraud represents only the tip of the iceberg.
Consider heart disease. In the 1960s, experts began to recommend that Americans reduce saturated food fats and cholesterol, which they promote as the main culprit behind heart disease, the country’s number one killer. However, the reduction of fats and cholesterol has not improved the situation. In fact, there is little robust evidence that low-fat diets improve health, and when the United States has joined this approach, it began to develop an obesity epidemic, with around 40% of adult Americans now qualifying as obesity.
Fifty years ago, it may be meaningful for scientists who found plates containing cholesterol stifling coronary arteries to blame heart disease on excessive consumption of food cholesterol. But this reasoning is simplistic to the extreme. On the one hand, the dicton of Ludwig Feuerbach that we are what we eat is bad. We do not become, for example, more cattle when we consume beef. Consumption of a low fat diet also does not seem to reduce the risk of heart disease. In fact, the much larger threat seems to come from simple sugars, which a nutrition researcher has marked “pure, white and deadly”.
In this case, although there was no attempt to falsify data, the intention to obscure seems to have played an important role. An influential article from 1965 New England Journal of Medicine Review revealed that fats and cholesterol were the main food culprits of coronary disease. It was not until much later that it became clear that this research had been funded by the Sugar Research Foundation, whose main objective seems to have been to exempt sucrose as guilty. If the examination peers and readers had known the source of funding, they could have submitted the report to a more in -depth examination.
However, the deeper problem is not so much a deliberate attempt to mislead, but the methods less than the robust which underlie almost all research in nutrition. John Ioannidis, MD DSC, a much appreciated researcher from the University of Stanford who launched early criticisms with ventures as diverse as the darling of Wall Street, now disappeared, and very generalized locking, helped to explain why it is difficult to base nutritional recommendations on really rigorous research. Perhaps his best known article, published in 2005, is “Why most of the published research results are false. “Summarizing his conclusions, he writes:
A research result is less likely to be true when the studies carried out in a field are smaller; When the effect sizes are smaller; When there is a higher number and a lower preselection of the tested relationships; When there is greater flexibility in conceptions, definitions, results and analytical modes; When there is greater financial interest and other prejudices; And when more teams are involved in a scientific field in pursuit of statistical significance. Simulations show that for most conceptions and study parameters, it is more likely that a research claim is false than true. In addition, for many current scientific fields, the affirmed research results can often be simply precise measures of the dominant bias.
Despite decades of study, there are still a lot of things we think we know but not. A difficulty stems from the fact that so many nutritional studies lack randomization, which means that any result is subject to confusion factors. The differences in results between two groups that we attribute to the regime can actually be explained by something else – factors such as smoking, alcohol consumption or exercise. Unsurprisingly, people who make an effort to consume only healthy foods can take care of other ways that are credited with diet.
This problem stems from the fact that most nutritional studies are observational and non -experimental. They are counting on the self -declared diet and the results for health, wrongly trying to conclude that the latter can be attributed to the first. Popular media are part of the problem in the sense that they are happy to report such results, even if their scientific base is low. As long as observation studies continue to draw such attention, randomized experimental trials are likely to languish.
Another problem is even more fundamental: the difficulty of evaluating the diet rigorously. Many studies depend too much on the recall, asking people what they ate over a long period, which is subject to many kinds of bias. Questioned by a researcher, many of us can tend to unconsciously minimize our food indiscretions and overcome the solidity of our nutritional choices. This applies not only to what we eat, but how much we eat many different types of food.
Like all human efforts, science is subject to biases, and this very responsibility constitutes a dead angle for many people, outside and inside the scientific community.
Another problem concerns the tendency to focus excessively on unique nutrients, such as proteins, foods containing vitamin D or cruciferous vegetables. Ioannidis suggests that the global role of any type of food or nutrient in taking into account human health is relatively low. It is likely that a person’s overall food exerts a much greater influence, but researchers often persist in focusing on individual constituents. In many cases, the “noise” of other factors probably submerges the “signal” of the food ingredient of interest.
Ioannidis compares the number of long nutritional studies that are not very reliable and unreliable to a pandemic. Instead of many small studies that try to answer a variety of questions, fewer targeted studies are needed. Although this would decrease the number of nutritional studies and researchers producing them, this would probably reduce the overall costs of nutrition research and provide much more reliable conclusions. Nutritional advice should be based on robust science, not competing opinions.
Of course, nutrition is not the only area in which the results are questionable or worse. Ioannidis maintains that similar problems fear other areas such as neuroscience and oncology. Summarizing the full extent of the problem, he described what he calls the “disorder of medical disinformation”:
First, highly published medical research is not reliable or is of uncertain reliability, offers no advantage to patients or is not useful for decision -makers. Second, most health professionals are not aware of this problem. Third, they also do not have the skills necessary to assess the reliability and usefulness of medical evidence. Finally, patients and families often lack relevant and precise medical evidence and qualified advice at the time of medical decision -making.
Perhaps the most fundamental problem focuses on the justification underlying these studies. In many cases, the intention is not to elucidate the truth but to advance a program – for example to increase the benefits of a company of pharmaceutical or medical devices or to advance the career of a researcher. Ioannidis characterizes many successful researchers such as “managers absorbing more money”. If a scientifically valid discovery or innovation generates income, so much the better, but income should not be authorized to fold science.
Scientific results are not necessarily true simply because they are supported by a large set of data, have been subjected to a complex statistical analysis or have been published in a journal evaluated by peers. Like all human efforts, science is subject to biases, and this very responsibility constitutes a dead angle for many people, outside and inside the scientific community. We must all remember that science, at the base, is not a corpus of irrefutable facts received, but one means among others by which we are pursuing knowledge. And because it is a human business, it is inevitably subject to a human bias.
Nutrition research funded by the sugar industry justifies the same meticulous examination as studies on the health effects of smoking subscribed by the tobacco industry. With large sums of online money, different and more selfish questions can be asked, research methods can be modified, analyzes can be biased and the results can, where unfavorable, be deleted or spun in the directions deemed more advantageous for the funder. One thing is certain – this low fat yogurt is not as healthy as we have long been led to suppose, especially if it is loaded with sugar.