Total Pageviews

Sunday 26 April 2015

Reporting Bias: No Love for Real Low-Carb Research

Why are diet studies designed and interpreted so badly?

by Richard Feinman PhD

Richard Feinman, PhD, is a professor of cell biology at SUNY Downstate Medical Center.

Articles in the lay press that put a positive spin on carbohydrate restriction are the persistent nemesis of the nutritional establishment.

Such was the case with a recent study on low carb diets by Bazzano et al., which was funded by the National Institutes of Health. Those of us working in dietary carbohydrate-restriction are continually frustrated at the NIH's unwillingness to fund low-carb studies and the media's reluctance to give its experimental successes appropriate coverage

The research community was eager to understand what the Bazzano paper did that the hundred or so previous studies had not done.

Like so many others, it clearly demonstrated the value of carbohydrate-restriction. This was not a revolutionary finding, since there had been vast amounts of earlier work, some of which had more impressive outcomes.

So, as several people pointed out, the story is in the story. The story is that somebody finally paid attention.

Paradoxically, the Bazzano paper may have perpetuated some questionable ideas and practices that, in my view, impede progress in the field.

First, in the way of precedents, important previous studies were dismissed on the grounds that they were shorter term.

The principle that a long study is inherently better than a short study is an opinion. Science usually judges studies on their quality of the work and the ability to answer the question at hand. Short trials have better control of variables and, in diet studies, better interaction with the participants.

Adherence to a diet in an experimental study, as in a clinical or real-world setting, may be dependent on how successful the diet is, but it is also understood that the encouragement and monitoring of the healthcare provider is the major factor and is likely to have less effect as the duration of the trial increases.

If effectiveness and adherence are not completely independent, the second is under substantial control by the experimenter. This is common sense and has been demonstrated numerous times, but the idea persists that we have to wait for the long-term study.

Also, it is incorrect to interpret a 6-month study as being only a 6-month study. One has to ask whether there is anything in the results that suggests any limitation for the future.

It is good that the effectiveness of dietary carbohydrate restriction is getting attention from the media and the granting agencies, but what is missing in this latest paper is the scientific foundation.
While clear on the practical benefits, the Bazzano paper suffers from a purely phenomenological bent, as if we should just try this diet and compare it to that diet. But "diet" in the sense of what people are told to eat is not a biological parameter.

The content of what they actually consume is the controlling variable. Not all macronutrients are the same, and the principle of removing carbohydrates derives from the fundamental biochemistry.

Dietary carbohydrate -- directly or indirectly, via insulin and other hormones -- is catalytic, controlling the disposition of fat and whether it is stored, incorporated into TAG-rich lipoproteins, or oxidized.

Fat, in this sense, has a passive role. The failure to understand this fundamental principle of metabolism accounts for the continued progression of papers critical of high-fat diets.

A high-fat diet in the presence of low carbohydrates is very different from a high-fat diet in the presence of high carbohydrates. Lack of appreciation of this idea is one barrier to progress in this field.

The Bazzano study made a valuable contribution in showing that low-carbohydrate diets "may be an option for persons who are seeking to lose weight and reduce cardiovascular risk factors and should be studied further."

However, further study will only be profitable if we focus on fundamental biochemical mechanisms and if we cite and critically evaluate the work that has already been done.


Graham

No comments: