One nutritional study, for example, shows that eating too many eggs can lead to heart disease, while another study finds the opposite.
The answer to this and other conflicting food studies may lie in the use of statistics, according to a report published today in the American Journal of Clinical Nutrition.
Foods that are clearly harmful one week are clearly good for you the next.
Led by scientists from the University of Leeds and the Alan Turing Institute (the National Institute for Data Science and Artificial Intelligence), the study found that the standard and most common statistical approach for studying food-health relationships is It reveals that it can lead to misleading and nonsensical results. .
Lead author Georgia Tomova, a postdoctoral fellow at the University of Leeds Institute for Data Analytics and the Alan Turing Institute, said: “These findings are a reflection of what we know about the effects of food on health. related to everything that exists.
“It’s well known that different studies on nutrition tend to produce different results. Foods that are clearly harmful one week may look good the next.”
dramatic change
Researchers have found that the widespread practice of statistically controlling, or allowing, someone’s total energy intake can lead to dramatic changes in the interpretation of results.
Controlling other foods eaten can further skew the results, making harmful foods appear beneficial or vice versa.
“Because individual studies vary widely,” Tomova said, “to provide an average estimate of whether and to what extent a particular food causes a particular health condition, review articles tend to rely on
“Unfortunately, most studies take different approaches to controlling for leftovers in the diet, so each study estimates very different amounts, making the ‘average’ rather meaningless. There is a possibility.
The study, funded by the Alan Turing Institute, identified the problem using a novel “causal inference” technique popularized by Judea Pearl, author of “The Book of Why.”
Senior author Dr. Peter Tennant, Associate Professor of Health Data Sciences at Leeds College of Medicine, explains:
“That’s why people say, ‘Correlation is not the same as causation.’ In doing so, we also highlighted quite a few areas that we didn’t fully understand. ”
The authors believe that this new study will help nutritional scientists better understand the problem of poorly managing total energy intake and overall diet, and more clearly understand the impact of diet on health. I hope it will be useful for you.
Dr. Tennant added: “Different studies may provide different estimates for different reasons, but this he believes that one statistical problem may explain many of the discrepancies.” Best of all, this can easily be avoided in the future.”
Further information
Photo: Adobe Stock
The report, entitled “Theory and performance of alternative models for estimating relative causal effects in nutritional epidemiology,” was published October 13 in the American Journal of Clinical Nutrition and can be found here.
For media inquiries please contact Kersti Mitchell, University of Leeds Communications Officer via k.mitchell@leeds.ac.uk.
Comments
Post a Comment