The blank page method

(Disclaimer : all newsletter content is for interest and informational purposes only, and should not be considered medical advice. If you are interested in any topics please research further and discuss with your personal physician)

Views & opinions are personal, and do not necessarily reflect Swiss Re's.

Want to know what to eat? Start with a blank page, or a history lesson.

Everyone has a strong view of what is healthy to eat. Much of it misguided, or based on poor or possibly no evidence. (Saying a health authority says one should eat this or that is not evidence)

Almost everything you think you know about what is healthy is an hypothesis.

Epidemiology

Almost everything you think you know about what is healthy is an hypothesis. Created mostly by large observational epidemiology studies.

I have read this recent piece by Gary Taubes on the inherent weakness of nutritional science, which overwhelmingly is based on epidemiology. I strongly suggest you spend a few minutes to read it later.

Realize that all the guidelines, and plates, and pyramids and recommendations are guesses. And many are simply wrong.

We just don't know which ones yet...although interventional trials of metabolically sick people indicate some areas where perhaps we have gone wrong.

So how does nutritional epidemiology work? Ask folks what they eat in detail AT A SINGLE POINT IN TIME (using food frequency questionnaires of which much has been written and criticized), and after a decade or 2, find statistical associations to health outcomes (e.g. heart disease, cancer, mortality).

Associated. Not causal. Associations are not causal and can only create an hypothesis.

Gary writes in details about these weaknesses, and the one I always think about is healthy user effect or bias. A conventional wisdom (based on whatever, and irrelevant whether proven to be true or) indicates THIS (behaviour, food, habit) is healthy and THAT isn't. Those that care about their health do more of THIS, and those that don't do more of THAT.

"They are NOT the same people"

Assuming the only difference between the groups is the THIS or the THAT is fundamentally incorrect. You are not comparing the same people.

The Nurses Health study was studying a single health intervention - hormone replacement therapy (HRT) for post menopausal women. One single health intervention. Observationally those taking it had significantly reduced coronary artery disease. (>50% reduction). WOW.

Randomized trials done later showed the opposite. In RCTs you take out any confounder like healthy user effect. And it completely demolished the Nurses Health Study outcome. Why? We don't really know but we can guess nurses taking HRT cared more about their health, had higher family incomes, better health care, more concerned doctors etc. i.e. They were not the same people.

If we cannot take a single health intervention which should be so easy and make the correct guess, how can we possibly do this in the context of nutrition.

Eating more salad might mean eating more greens, more tomatoes, more olive oil, more balsamic vinegar, more olives, more pickles, more cucumber, more radishes, more egg, more croutons, more salad dressing and the list goes on. And it means eating less of other stuff. Less dessert, less bread, less rice, less pasta, less meat, less pizza, less junk food, less chocolate (and again the list goes on). So how can we possibly try to associate "salad" with a health outcome when the replacement effect creates the same problem of changing one corner square colour of a Rubik's cube and messing up the whole thing. But all we hypothesize on is that one corner square colour, without realizing the huge changes created by that one difference. And no, "adjusting for..." cannot completely correct for the infinite possibilities. It is too complex.

Salt

And then we have salt.

No RCT has ever been done on salt reduction and health outcomes. Not one.

And at our last Food for Thought in 2020 we had world experts debate this. Its not a slam dunk. Currently the European guideline for the upper limit of daily salt intake (~6g per day) is more than double that of the US (2.3g per day).

A 2-3x difference in the maximum recommended intake of a single compound - NaCl - between the US and Europe.

And then we think we can rely on the guidance of complex foods containing 1000's of compounds in varying combinations and proportions made from different quality ingredients with many added chemicals, preservatives, colourants etc.

I don't think we can.

History

So then what should we eat? Anything we want? Possibly not the best idea. Another way I try to fill in the blank page is by considering the past.

First hominids came about 2million years ago. And they continued to hunt, and gather what they could, and be hungry for large periods of time. We evolved before, and then as part of the hominid family during this 2 million year period, only being able to eat what we killed or found.

For 2 million years as Hominids we existed only by hunting and gathering.

100'000 years ago Homo sapiens stabilized as a species. Our evolution essentially halted due to lack of survival pressure. And that evolution was based exclusively on hunting and gathering. And then for another 90'000 years we continued to only hunt and gather as a species.

Since Hominids appeared they have spent 99.5% of their existence only hunting and gathering. Homo sapiens have been hunting and gathering for 90% of their existence.

10'000 ago agriculture helped societies remain geographically stable by having a more reliable food source. It progressed slowly and of course exploded in the last few centuries with the industrial revolution.

Processed and modern foods (common consumption of sugar, flour) did not exist much before the 18th century. Vegetable oil did not exist before 1960.

So for 2 million years our DNA and biology evolved to cope and thrive with what was about naturally. And in the 1960's we told the world that everyone should avoid the fats we had been eating (and had evolved for eating) for 2 million years, and eat an industrially processed oil never part of the natural food system.

So now we have had this new recommended food as a major fat in our diets for perhaps 50-60 years. And this recommendation was based on? Yes, epidemiology, and subsequently found to be of very poor quality, with cherry picked data to try and prove the now debunked diet-heart hypothesis. But still we have the WHO telling us to limit saturated fat.

We have now been eating processed vegetable oil for 0.05% of the time Homo Sapiens have existed. Or 0.0025% of the time Hominids have been on this planet.

And while many will consider the obesity trends and consider the correlation strong, it is just a correlation.

We did other in things in the second half of the 20th century as well, including introducing high fructose corn syrup, explosive sugar consumption, far more processed carbohydrate intake and ultra processed foods have overwhelmed the food system.

A complex Rubik's cube that is hard to solve! But start with that blank page....