Wednesday, April 1, 2015

It’s Nuts: The Confusing Findings from Nutrition Research

Is your head spinning with all the confusing and contradictory news reports on diets that prolong life and prevent disease?   Just this week in the New York Times, there were reports that red meat was really not bad for us, that fish oil probably doesn’t do anything to prevent stroke or heart disease and that eating nuts increases our lifespan and prevents Alzheimer’s disease. These articles come in the wake of other reports that suggest that dietary fat, at least from some sources, is not the enemy.  Butter even made a starring appearance on the cover of Time magazine.

Why is there so much contradictory evidence on food?  And what should we believe? The gist of the argument goes this way.  Nutritional research on humans is difficult to do, and much of it has been flawed and of poor quality.  A 2013 paper written by two researchers, Jonathan Schoenfeld and John Ionnidis, evocatively titled, “Is everything we eat associated with cancer? A systematic cookbook review,” found that there were claims that 40 ingredients frequently mentioned in cookbooks were associated either with increased or decreased cancer risk.  In their detailed statistical analysis of the research behind these claims, Schoenfeld and Ionnidis found that most of the evidence was weak.  In fact, the more studies done on an ingredient, the more likely the average effects found trended toward zero. 

How do we make sense of the information presented in the media?  Unfortunately, many health reporters appear more concerned with headlines than understanding science.  An exception is Aaron Carroll, who wrote the “Red Meat Is Not the Enemy,” article and has authored other thoughtful pieces on nutritional research.

There are some criteria you can apply next time you read or hear a story about the nutritional benefits of a food or diet.

1.  Is the evidence from a descriptive study?  What I mean by that is a study that gathered information on large numbers of people and looked for associations between dietary or other factors and a particular disease.  There are two major problems with these kinds of studies.  First, the findings may be due to chance.  If researchers test the association of 100 different foods with a disease such as Alzheimer’s, we know that at least five of those comparisons will be statistically significant by chance.  In other words, these will be false positive findings. Some statisticians argue that the false positive rate may be as high as 13 out of 100 comparisons.  Many descriptive studies have probably done this kind of fishing in their data, looking for the comparisons that are significant and ignoring the ones that are not.

The second problem is these associations are not the same thing as causation.  The association of nuts and health may be due to the fact that some other factor might lead people to eat nuts and also contribute to health.  Maybe nut eaters among us are wealthier, or more health conscious to start with so they exercise more and avoid processed foods like potato chips and French fries.  It may not be the nuts at all that make a difference.

2. In contrast to descriptive studies, researchers sometimes conduct experiments in which people are randomly assigned to the experimental diet or to a control group.  This is potentially a stronger research approach, which can establish that it is the specific diet, and not other things people might be doing, that contributes to a positive outcome.  This approach is the main way that medications are tested, and it works well in the short run.  But there are two problems that plague research on diets.  First, people comply with the diet in varying degrees. Second, we need long time periods to be able to see the effects of diet.  As the study goes on, people are less likely to be compliant and some will begin to drop out.  In the end, the people who stay in the study and stick with a diet will be highly conscientious and so they are probably doing other things to assure good health.

3.  Whatever type of research is used, a single food is not likely to make a difference. As much as we would wish to find foods that have magical abilities to prevent illness or delay aging, there are no superfoods. This is the point of the cancer paper and it’s the point that Aaron Carroll makes in the red meat article.  Carroll states that one food is not “the culprit,” but rather that too many of us eat too much.  In other words, it’s obesity that increases the likelihood of many different diseases and an early death, and not any one specific food that we eat.

What to do?  In truth, we all know what a healthy diet looks like.  A variety of whole, natural foods in reasonable quantities.  Really, it's that simple.  There is a large body of research on diets, whether it's Weight Watchers, Atkins, the Mediterranean diet, or any other diet "prescription." That research tells us that people lose weight while they are on any of these diets, but the majority of people will gain back the weight when they stop adhering closely to the diet.  In fact, the implication of a diet plan is that we "go on" the diet, and eventually "go off."  And that's the problem.  Most of these very rigid and restrictive diets are not sustainable because people feel deprived of many foods they love.  And they find it difficult to eat in restaurants or with friends.  

What makes more sense is to decide to focus on healthy foods and perhaps to make one change at a time.  For instance, deciding to eliminate soft drinks and substitute flavored sparkling water will subtract calories (or subtract artificial sweeteners in the case of diet soft drinks) without decreasing the amount of liquids that you drink.  Eating a piece of fruit at the end of a meal instead of a cookie is another possibility. Having an open-faced sandwich halves the amount of bread used.  Only eating half of your entree in a restaurant and taking the rest home will give you a more reasonable portion in most cases.  It takes several weeks to establish a new habit, so pick one thing (we usually recommend the easiest one first), and continue until you no longer have to think about it.  Then make one more change.  There's no real end point, just a series of changes moving you towards a healthier diet.  And because you aren't "on a diet," if situations arise where you splurge on a treat, it's okay.  Just go back to the good habits you're developing the next day.  One of the advantages of making your own decisions about what you eat rather than following a prescription is that you don't have to feel guilty if you "go off the diet," nor ashamed because you couldn't stay on it.  

After all, enjoying good food is one of life’s pleasures. 

Links to the articles cited here:

Fish Oil Claims Not Supported by Research

Red Meat Is Not the Enemy

Nuts Are a Nutritional Powerhouse

Is Everything We Eat Associated with Cancer? A Systematic Cookbook Review

No comments:

Post a Comment