Obesity: Debate, Dogma and Harvard Dons

Extract from “The Road Not Taken” By Robert Frost

I shall be telling this with a sigh

Somewhere ages and ages hence:

Two roads diverged in a wood, and I—

I took the one less travelled by,

And that has made all the difference.

Some years ago, I was invited by the President of the University of Auckland to deliver one of the university’s public lectures. I chose the title “Obesity: Down the road less travelled” and I tried to show how some of the conventional wisdom on obesity might be a bit too rigid in that it might chose to ignore some awkward data. My overall feeling was that the lay audience enjoyed the talk but that the public health nutrition experts in the audience were not impressed. They were comfortable in their prevailing paradigm. Throughout my professional career as a Professor of Nutrition, I have found that the excitement in research is not showing yet again, albeit under slightly different conditions (old/ young; fit/sedentary; diet A/B, etc),  that some basic tenet of modern nutritional thinking remains safe and sound. That excitement is to be found in the unexplained, unexpected and unusual finding. In proudly pursuing this vision of science, I have been reproached, ridiculed, slandered and shunned. One learns to have a thick skin. 

I was aware of a spat some years ago between a career researcher at the US CDC and the world-renowned Dons at the Harvard School of Public Health. However, it was only recently that I could read the full account of this very public spat. The CDC researcher in question was Dr Katherine Flegal and in a recent paper entitled “The obesity wars and the education of a researcher: A personal account” she provides a detailed account of how Harvard Dons rushed to the parapets of the print and electronic media to condemn, in no uncertain terms, the work of Dr Flegal (1).  

In 2004, researchers at the US Centre for Disease Control (CDC) published a paper in the Journal of the American Medical Association (JAMA) which examined the main causes of deaths in the US. Tobacco was the primary cause of mortality (435,000) with poor diet second, at 400,000 deaths. The dominant factor associated with poor diet was obesity. The paper attracted significant attention and the then director of the CDC, Julie Gerdering predicted that obesity would soon overtake smoking as the leading cause of death in the US. 

But others within the CDC were of the view that this short paper had some flaws and they set about correcting these. Their paper, led by Dr Katherine Flegal, was vetted internally within the CDC and was published in JAMA in 2005. It showed a relationship between bodyweight and mortality rate which was quite different from the original study and which challenged conventional wisdom. The instant the paper appeared, several professors from the Harvard School of Public Health began what can only be described as an extraordinary tirade of abuse, demonising the author and discrediting the paper. This month, Dr Flegal published a paper in a leading cardiovascular journal, outlining in detail the vicious backlash to her challenging paper.

Flegal’s paper differed from the earlier CDC publication in several respects. She used actual measurements of weight and height as opposed to self-reported data which overestimates height and which underestimates weight. Flegal used three recent datasets which were nationally representative of the US population whilst the earlier paper used only one old nationally representative sample. That earlier paper did not adjust for effects of age, sex, smoking or other factors on mortality. Dr Flegal did. The earlier paper had several calculations errors which were eventually accepted by the authors. 

Flegal’s paper showed that as body mass index (BMI), increased, mortality increased but in a manner quite different to conventional wisdom. That wisdom held that at the level of overweight (BMI 25-30), mortality began to rise. Above 30, it began to soar. Flegal’s data did show that mortality rose as BMI rose but not until a BMI of 30 in young people. In older persons, mortality showed no change across the entire spectrum of BMI. 

One would have imagined that the Harvard dons who were instrumental in constructing this conventional wisdom would have invited Dr Flegal to discuss her paper and to explore how both could move forward in the pursuit of best science. Instead, they hit the media. Journalists were told that Flegal’s study was “really naïve, deeply flawed and seriously misleading”. Flegal points out that Professor Walter Willett, the doyen of these dons, in an interview with NPR quipped: “This study is really a pile of rubbish and no one should waste their time reading it”. Willett raised his concerns directly with the Director of the CDC who stood behind Flegal’s work, and a research fellow at Harvard blogged that the paper had been retracted and the Flegal was demoted. All utterly untrue.

 Within a week of the publication, the Harvard dons hastily organised a special symposium at which expert after expert, mostly from Harvard, slated Flegal’s paper. That same week, Flegal was giving a paper at a symposium at Berkeley. Standing at the door was a young woman handing out sheets of paper  setting out why Flegal’s paper was flawed. The matter reached low levels when, students taking courses at Harvard, tweeted that one of the dons had advised that Dr. Flegal’s scientific analysis shouldn’t be trusted because she was “a little bit plump herself”.  

Flegal responded to these attacks exactly as any scientist should by searching the literature on the topic to see what others had found. She conducted a meta-analysis of all published papers which had examined the relationship between mortality and BMI and had her findings published in JAMA in 2013. Her work examined 141 eligible papers and concluded, as she had before, that rising BMI was associated with all cause mortality. However, she noted again that the strength of that relationship was much weaker than conventional wisdom and, again, showed no association with rising BMI within the “overweight “category. In an interview with the BBC, Professor Willett stated: “This is an even greater pile of rubbish” than the study in 2005 and in a radio interview on NPR he repeated this criticism of the 2013 study and separately noted that “rubbish” was a polite term for what he really wanted to say.

The truth is somewhere out there, be it Flegal’s truth or Willett’s truth. It matters a lot in public health to know where that truth lies. But when science gets closer to policy, be it climate change, vaccination programmes, transport or public heath, then science tends to be simplified into bite size pieces, suitable for policy makers. Shades of grey are out of the question and maybe that is necessary for policy makers. However, unanimity of opinion, as has been said, is fine for religion and politics. It has no place in science which must stick solely to risk assessment and be open to changing advice as new data emerges. The Harvard dons should know better. 

(1)Flegal KM. The obesity wars and the education of a researcher: A personal account. Prog Cardiovasc Dis. 2021 Jul-Aug;67:75-79. doi: 10.1016/j.pcad.2021.06.009. Epub 2021 Jun 15. PMID: 34139265.