Hong Kong in Motion

Data on Bias That Defies an Investigator’s Bias

When does a hypothesis become a bias? One answer can be found in a recent publication about nutrition research in JAMA Internal Medicine. The authors – Nicholas Chartres, Alice Fabbri, and Lisa Bero – surmised that food industry sponsorship of research might generate outcomes that favor the sponsors. They conducted a systematic review and meta-analysis and found “insufficient evidence” to support their hypothesis.

So the authors re-asserted their hypothesis. They concluded that “industry-sponsored studies were more likely to have conclusions favorable to industry than non–industry-sponsored studies” while conceding that “the difference was not significant.” They also called for more research to look for evidence of this problem.

Naturally, reporting on this study was very confused. Medscape reported “Clear Bias in Studies Funded by Soft Drink Industry.” Business Insider reported “Science Can’t Find Evidence of Bias in Food Industry-Funded Research.”

In describing the problem of influence from the food industry, senior author Lisa Breo said:

Conclusions do not always agree with results, but can be “spun” to make readers’ interpretations more favourable. This spin on conclusions has been identified as a tactic used in other industries and can influence how research is understood by the lay community.

Perhaps we have some bias about bias.

Observing a steady stream of such publications in JAMA Internal Medicine, we have come to believe that the editors have a strong point of view on this subject. One might even call it a bias.

Research integrity is critically important. The point of research is to test our suppositions and our biases. Industry, academia, government, and nonprofits all harbor biases. We see some of the strongest biases coming from people with a righteous point of view and a book to sell.

Strong points of view should be aired, debated, and tested through rigorous research. Fallacious arguments against the people advancing an idea have no place in this process. The idea is what matters.

Click here for the study by Chartres et al. Click here for more on nutrition research integrity.

Hong Kong in Motion, photograph © Steve Webel / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


November 7, 2016

4 Responses to “Data on Bias That Defies an Investigator’s Bias”

  1. November 07, 2016 at 9:28 am, James said:

    Well said – We see some of the strongest biases coming from people with a righteous point of view and a book to sell.

  2. November 07, 2016 at 3:43 pm, Allen Browne said:


  3. November 11, 2016 at 7:02 am, Allen Browne said:

    Sounds like studies on the effectiveness of diet-activity-behavior (DAB) for those who already have obesity. “It’s not very often very effective so we need more research to show it’s effectiveness”. Millions of dollars that could be used to help find effective treatment algorithms based on the physiology and biochemistry of the disease.

    • November 11, 2016 at 7:45 am, Ted said:

      Never let facts get in the way of what you know is right.