Thursday, May 23, 2019

An example of how nutrition studies are used to disseminate misinformation

Pam Popper is expert in analyzing nutrition studies for quality, or lack thereof. A study is done, it is published in a journal, and news media goes to press with it. They may be particularly interested in publishing if the study is sure to attract a lot of attention, as was the case with this 2015 study on cholesterol that Popper has a look at for the benefit of us non-experts. She shows us how a study is analysed for quality in an 11 minute presentation:


I'd like to add a few other points about why we the public do not realize these sort of "studies" are misrepresentation, and why we find them so compelling.

1) We do not know, and have no way of knowing, that the "weight of the evidence" lies exactly in the opposite direction. Only legitimate nutrition scientists would happen to know this sort of thing off hand, along with a relatively small number of medical doctors (the study geeks).

What is "weight of the evidence"? It's the aggregate conclusion of all quality studies on one specific topic, after filtering out low quality where study authors are shown to have pre-existing bias, or where honest mistakes have been made in the design of the "model under test".

2) We have our own biases, and one thing that's true about biases is we like to have them confirmed. In the case of health and nutrition studies "we love to hear good things about our bad habits". All things in moderation sounds reasonable until we consider it can mean almost anything depending on the biases of the individual professing that sentiment.

3) Media, where we learn of such studies, has approximately the same biases as the general population. An overweight editor wants to hear "bacon is back" as much as the rest of us.

4) Nutrition is confusing for most of us because interest groups use low quality studies to make their case (and protect their profits) in the media. The studies that are sensational, and confirm public biases, attract the attention of the media. And we have no way of knowing there are frequently dozens (or more) quality studies saying the exact opposite. These are the studies that are rarely published outside of scientific journals.

No comments:

Post a Comment