Harvard Food Law Society "Forum on Food Policy" TEDx Conference

Last Friday, it was my pleasure to attended and present at the Harvard Food Law Society's TEDx conference, Forum on Food Policy.  I had never been to Cambridge or Boston before, and I was struck by how European they feel compared to Seattle.  The conference was a great success, thanks to the dedicated efforts of the Food Law Society's presidents Nate Rosenberg, Krista DeBoer, and many other volunteers. 

Dr. Robert Lustig gave a keynote address on Thursday evening, which I unfortunately wasn't able to attend due to my flight schedule.  From what I heard, he focused on practical solutions for reducing national sugar consumption, such as instituting a sugar tax.  Dr. Lustig was a major presence at the conference, and perhaps partially due to his efforts, sugar was a central focus throughout the day.  Nearly everyone agrees that added sugar is harmful to the nation's health at current intakes, so the question kept coming up "how long is it going to take us to do something about it?"  As Dr. David Ludwig said, "...the obesity epidemic can be viewed as a disease of technology with a simple, but politically difficult solution".


Taxes/regulations are vigorously opposed by the processed food industry, and also (more understandably) by people who don't want to have their food choices legislated.  Children in particular should be federally protected from predatory food industry practices.  Personally, I'm in favor of legislation that de-incentivizes added sugar consumption.  What if we had a sugar tax that paid for some of the obesity and diabetes-related expenditures that taxpayers currently shoulder through Medicare and Medicaid?  That would simply balance the "externalized" cost of health problems that are caused by sugar in the first place.   

The first panel of speakers on Friday was on nutrition and health, and set the tone for the rest of the conference, which was focused on policy.  Dr. Walter Willett was the first speaker.  Dr. Willett is the chair of the Harvard Nutrition department and the second most cited scientist in clinical medicine.  He reviewed the evidence from his observational studies supporting his vision of nutrition and health.  His optimal diet focuses on whole grains, vegetables and fruit, legumes, fish and poultry, unsaturated fats from nuts and seed oils, limited whole eggs, and avoids refined carbohydrate/sugar, red meat, and replaces saturated fats with unsaturated fats.  He suggests that a fat intake of ~40% of calories is healthy as long as the fat is unsaturated. 

I spoke second, giving a talk titled "The American Diet: a Historical Perspective".  The talk was about the US diet, so apologies to non-US residents of North and South America for the imprecise terminology.  The talk began by describing briefly how certain aspects of health have changed in the US over the last 120 years (increased obesity, diabetes, heart attacks, and other "diseases of civilization"), then focused on the major changes in diet over the last 200 years or so.  The main point I kept coming back to is that the diet has shifted dramatically from simple, home-cooked food to commercially prepared food, although I covered many other aspects of dietary change as well.  I think it was an informative talk, and it was well received.  Volunteers told me it will be freely available on the TED website, as well as their Food Law Society website, within 2-3 weeks. 

Dr. David Ludwig spoke third.  He's an accomplished researcher at Harvard and Boston Children's hospital who focuses on childhood obesity.  His talk focused on the danger of excessively processed/refined food and I thought it dovetailed nicely with what I presented.  He started out with a perspective on the evolution of the human diet from early hominid times-- it's nice to see this line of thinking become more prevalent, although his concept of the Paleolithic diet was not entirely consistent with the evidence I've seen.  I thought one of the most memorable moments of the conference was when he showed a picture and described a meal marketed to children called "Mess With Your Mouth Tacos".  He must have bought one of these and carefully laid out and photographed everything the box contained, which included various "pastes and concoctions", refined tortilla shells, a Capri Sun sweetened beverage, and Nerds (candy).  Whether or not the Nerds go inside the tacos wasn't clear.  The point is that when you stop and think about it, the idea that this stuff is food and should be eaten by children is so ironic it had us snickering in our seats.  They should rename it "Mess With Your Pancreas Tacos".  The talk ended with Dr. Ludwig recommending a "Mediterranean"-type diet for health.  Although it was largely crafted by American nutrition researchers and does not adhere very closely to actual traditional Mediterranean diets (it's loosely based around the traditional diet of Crete), I nevertheless believe it is far superior to the typical American diet.

After our talks, Drs. Willett, Ludwig, Lustig and I participated in a panel discussion about nutrition and health, moderated by Corby Kummer, an accomplished food writer and senior editor at The Atlantic.  We covered many topics, including various aspects of sugar consumption (which were not particularly controversial), meat, and the glycemic index.  I wasn't able to participate as much as I would have liked, due to the assertiveness of the other panelists and the fact that the moderator essentially ignored me.  But I got a few points in related to red meat consumption and traditional grain processing.

Dr. Willett made a few remarks that I wasn't able to address effectively at the time but that I think are worth discussing further.  One of the first things he did was express his dim view of diet-health information on the Internet.  I actually agree with that to some extent.  The Internet has democratized information, for better and worse.  On the positive side, it has given people unprecedented access to information that has allowed them to self-experiment and connect with interesting new ideas, which has often been helpful.  On the other hand, it had allowed the proliferation of "instant experts" who are long on advice and short on rigor, and who manage to convince others to adopt dietary changes that aren't necessarily in their best interest.  But these people existed in the popular press long before the Internet went mainstream. 

Another topic that came up, but which I wasn't able to comment on was the glycemic index.  Support for the importance of this concept comes almost entirely from observational studies, where people who eat high-glycemic foods tend to have worse health outcomes (fat gain, diabetes) than those who eat lower-glycemic foods.  I have three difficulties with this idea: 1) the highest-glycemic foods in the US diet are white flour products and sugar, so how do we know the glycemic index is the relevant factor?  2) controlled trials lasting from 10 weeks to 18 months overall have shown no meaningful effects of glycemic index on total food intake, fat gain, insulin sensitivity or any other marker of health in non-diabetics (1, 2), 3) many non-industrial cultures eat diets that rely heavily on high glycemic carbohydrate such as cassava, taro, partially milled rice, and milled millet and corn, and they generally don't become overweight, diabetic or have heart attacks (3).  I readily admit that could also have to do with other lifestyle factors such as exercise, sleep, etc., but that is true of any ancestral lifestyle.

This brings us to another thing Dr. Willett seems to have a dim view of, which is historical diet-health evidence.  I partially agreed with him, in that this kind of evidence is the least well controlled, and just because we were doing something 100 years ago and weren't fat, doesn't mean everything about our diet and lifestyle was optimal.  However, what it gives us is an archetype for a diet/lifestyle that does not lead to a high chronic disease burden over generations, which is valuable.  I was disappointed to hear Dr. Willett deploy the "people didn't live past 50" argument that is often used to hastily discredit historical/anthropological evidence.  I wasn't able to get this point in, but the fact is that we have age-adjusted data from the US, the UK, and a number of non-industrial cultures, suggesting that at the same age, the chronic disease burden for certain diseases was very low compared with today.  Furthermore, if you compare disease precursors that are present in young people in our society, such as obesity and high blood pressure, the prevalence of these is often radically lower as well.

Another point that I didn't make is that each form of evidence has its limitations.  While historical/anthopological evidence is the least well controlled, the second least well controlled is observational evidence.  Its principal limitations are 1) it is intrinsically incapable of identifying cause and effect relationships; 2) it often suffers from substantial inaccuracy in food assessment (such as that inherent in food frequency questionnaires), which is unevenly distributed across food groups (4); 3) one can never measure and adjust for all relevant confounding factors, and people who live a healthy overall lifestyle will differ from those who don't in ways that are probably relevant to health but can't be corrected for, and this injects a significant risk of spurious results into observational studies, particularly when they're studying factors that carry a cultural stigma and tend to be avoided by health-conscious people (e.g., red meat, 5, 6).  For these reasons, observational studies shouldn't be used as the sole basis for public health recommendations.  A better approach is to look broadly at historical/anthropological evidence, observational evidence, controlled trials, and basic science/mechanism, and see if a coherent pattern emerges.  That allows us to increase our degree of confidence, since the strengths of each method will buttress the weaknesses of the others.  

The rest of the conference focused on food policy from a health, susctainability and marketing perspective.  There were some great speakers, whose talks will be available online shortly.  Policy isn't my specialty so I'm not going to comment on these much, but it was good for me to be exposed to the ideas.  One interesting point that came up is the fact that children don't have the ability to determine persuasive intent in advertising-- i.e., they can't tell when someone is trying to convince them of something that isn't necessarily in their best interest.  They take things that people tell them as fact.  Therefore, marketing to children is inherently deceptive.  One of the highlights of the conference was the quote (I'm paraphrasing) "we need to de-normalize the fact that a clown is telling our children how to eat".  Apologies to the author of that quote whose name I don't recall.  

Overall, an excellent conference.  Thanks to everyone who made it happen.