The Big Hammer of Self-Reported Dietary Data

“Give a boy a hammer and everything he meets has to be pounded.” Abraham Kaplan’s Law of the Instrument has some wisdom to offer for some of the contentious arguments ongoing about self-reported, observational dietary data. The latest round in this argument was published by Steven Nissen and Nina Teicholz in the Wall Street Journal. As others have, they take issue with the over-dependence upon such weak data to back up dietary guidelines:

Many of the dietary recommendations are justified by observational studies, using a scientific method known as prospective epidemiology. Researchers send out questionnaires to large numbers of people, asking about diet and lifestyle. They then follow up for years to record health outcomes. This method cannot show causation, only associations.

Ambika Satija and colleagues from Harvard, who built their careers on large observational studies such as the Nurses Health Study, responded to some of these criticisms in Advances in Nutrition. They wrote that the criticisms are “nonconstructive and sometimes naive.” They went on to say:

Some researchers consider RCTs as the be-all and end-all of causal inference. This sentiment may be appropriate in the pharmaceutical industry, but the drug trial paradigm cannot be readily translated for use in the nutritional sciences.

It’s worth noting that experimental methods are not contrivances of the pharmaceutical industry.

The rhetoric has become overheated on both ends of the spectrum. Stubborn insistence that large, prospective cohort studies are “irreplaceable” doesn’t help. Characterizations of people who disagree as dangerous and misleading are unhelpful, too. Likewise unhelpful are arguments that broadly dismiss key nutrition research methods as “pseudoscientific impediments to scientific progress.”

When all is said and done, not every question in nutrition and obesity yields to pounding with the big hammer of observational data. It’s a great hammer that delivers some great discoveries. But it does get misused.

A tremendous opportunity for innovation with new experimental methods lies ahead. Wise researchers will seize it and stop bickering.

Click here for the commentary by Nissen and Teicholz. Click here for the commentary by Satija et al. Click here for more perspective from the Washington Post.

Listen! photograph © Domiriel / flickr

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


February 9, 2016

6 Responses to “The Big Hammer of Self-Reported Dietary Data”

  1. February 09, 2016 at 6:31 am, Joe Gitchell said:

    Amen, Brother Kyle.

    And I’m willing to guess that you call for less heat and more light will likely go about as well as my efforts (so far–hope springs eternal)!



    • February 09, 2016 at 8:29 am, Ted said:

      Thanks, Joe. I think the time is coming when we will move on from this argument. Maybe the National Academy of Medicine will be helpful.

  2. February 09, 2016 at 9:49 am, Erik Arnesen said:

    Hi, BTW, did you see the response from Frank Hu, Walter Willett et al. to Nissen’s editorial in Annals of Internal Medicine? It’s in the comments tab here:

    • February 10, 2016 at 7:14 am, Ted said:

      Thanks! I had not seen those comments — more heat. Need more light.

  3. February 10, 2016 at 11:29 am, Allen Browne said:


    As you say “The rhetoric has become overheated on both ends of the spectrum.” But as someone once said – “Don’t throw the baby out with the bath water”. Discussion and critique is good as long as it stops there.

    Yup – wise researchers will seize the opportunity.and we need to keep our eye on the ball – the patients.

    Keep up the good work

    • February 10, 2016 at 2:31 pm, Ted said:

      Thanks, Allen.