Issues With Nutrition Recommendations - Part 1

Do you find the world of nutrition and diet confusing or frustrating? Do you struggle with determining the healthiest diet for yourself?

There have been many moments where the world of nutrition has made my head start to spin. You have the government, most doctors, and dietitians recommending the food pyramid. You have scientists and the dietary camps arguing back and forth, all making seemingly good points. But who is right? Vegans/Vegetarians? Keto? Paleo? Carnivores? Every diet in between?

Perhaps this head-spinning madness is due to nutrition research being subject to heavy bias and manipulation, and the difficult (maybe impossible) challenge of extrapolating a health outcome to someone's diet, when health is very multi-factorial. Even when recommendations are based on the bulk of the available evidence, and even supported by various institutions, that doesn't always discount counter evidence or mean there is no issues with the bulk of the evidence. It also doesn't mean it will actually apply to any given individual. A lack of evidence might also mean there has been a lack of interest or gain from doing the research, so evidence may yet to be uncovered, or research may never even be done.

Here are some potential challenges with extrapollating conclusions from nutrition research, and some issues with nutrition recommendations based on them:

• Many of our nutrition recommendations are based on large scale observational studies.

• These are done by observing what people ate or having them recall what they ate in the past (sometimes over several months) then linking that to disease outcomes. What did you have for dinner two weeks ago today? Don’t know? Me neither! People can also be more likely to under and over estimate consumption of certain foods if they feel they are being judged on their food choices, or if they are aware their food choices may be considered unhealthy.

• Researchers then have the very difficult task of trying to control for a vast amount of ever changing lifestyle factors, environmental inputs, and nutrient interactions (some known, some unknown), but they aren’t going to be able to control for everything, creating a bit of a guessing game.

  • The only way that lifestyle factors could be controlled would be to seclude study participants away from their normal daily lives long term, which is unethical. Just as it would be, for example, to feed someone poor quality food you suspect will be detrimental when consumed long term, in order to see how it impacts their health.

• Relating to the above point, nutritional epidemiology and observational studies run a unique risk of being inaccurate due to healthy user bias and unhealthy user bias.

  • Let's say a large scale observational study indicates that a plant based diet results in better health outcomes and that diets including more animal products result in poor health outcomes. The narrative over the last several decades has been that meat and fat are bad for you. So the individuals consuming more plant foods are also more likely to be living healthier lifestyles (healthy user bias), while many individuals consuming higher meat and fat diets are more likely to live unhealthy lifestyles (unhealthy user bias). Comparing a whole food, plant based diet and a healthy lifestyle to a Westernized diet and lifestyle containing animal foods is obviously going to show greater health results than the latter. No one needs to run a study to figure that out. But it by no means shows that the absence or reduction of meat is what produced the favorable end result.

  • For example, some western epidemiology shows meat consumption is associated with poorer health outcomes, but in China the same types of studies indicate the opposite. Which is right then?

• In the past (and present) nutrition science institutions and their philosophies have also been largely influenced by founders personal beliefs or corporate interests. For example, the institutional influence on the anti-meat narrative and the push for the world to go more “plant based”.

  • There is the Seventh Day Advent Church (who are vegan/vegetarian proponents) with strong ties to Loma Linda University research and the Academy of Nutrition and Dietetics, with their broad reaching nutrition guideline influence.

  • There is also the issue of research bias (impacted by both financial and personal interests), data cherry picking (choosing data that supports sought after research conclusions and rejecting the unsupportive data), misinterpretation of data, and even data suppression (like this suppressed research found hidden away in a dead researchers basement), adding further confusion, misinformation, and potentially impacting the health of millions of people.

• Animal food consumption being associated with poor health is a prime example of this. Logically and intuitively does it makes sense that animal foods - whole foods, rich in essential nutrients, that humans have been thriving on throughout our history - are all of a sudden key players in the development of all our modern chronic diseases over the last several decades?

• However, to be fair, when you consider public health nutrition vs individual nutrition, and that most people's meat consumption is in the context of a Westernized diet and lifestyle, the recommendations to reduce consumption might have some validity on paper. However, I would argue that in the context of a Westernized diet and lifestyle that animal foods are not the actually the problem.

• The "meat is bad for you" narrative has been repeated for so long that it's largely assumed to be true. This repeated messaging then becomes very difficult to retract from public knowledge and the belief that it is best practice endures.

• An excellent example of this is the claim that red meat causes cancer from the IARC Report 2015, a consensus statement from scientists in 22 countries. They examined research between red meat intake, cancer and potential risks. They determined that for every 100 grams of unprocessed red meat eaten per day there was a 17% increase in colon cancer risk. When looking at processed meat, the risk was 18% increase per 50 grams. The study declared that red meat was carcinogenic which then became headline news.

HOWEVER - there have been various critiques of this study you likely never heard reported:

• The relative risk (the risk of doing the action and getting the disease) of animal protein and cancer was a score of 1.2-1.3 (below 2 is not considered statistically relevant).

• If you compare that to smoking and cancer, the relative risk is a score of 12, with over a 1000% increased risk.

• Unfortunately, that 18% makes a much sexier headline.

  • This doesn't mean the risk is null, but again, you must consider the context. Do you think a whole food, animal inclusive diet, alongside healthy lifestyle and environmental inputs has a similar risk of an animal inclusive diet that's high in ultra-processed foods with unhealthy lifestyle and environmental inputs?

  • Could it be more likely that animal foods mistakenly got lumped in with the true culprits of poor health - being ultra-processed foods, full of highly refined flours, sugars, seed oils, preservatives, flavor enhancers etc, and the overconsumption of them, in combination with an unhealthy lifestyle?

• When reviewed it was discovered that these scientists used only 14 out of 800 studies in their conclusion and all 14 were observational data. This is an example of cherry picking data to fit their hypothesis.

• The excluded studies also contained interventional studies that clearly did not show a relationship between red meat and cancer.

• In the report, 8 of the 14 studies showed no link to red meat and colon cancer development.

• Of the remaining 6 studies only one showed a statistically significant correlation between meat and colon cancer, meaning it could be a mistake or showing up by chance (an outlier). This study was also from the Seventh Day Adventist’s (a vegan/vegetarian advocate church).

  • The study also indicated that the meat consumers lived unhealthy lifestyles (unhealthy user bias) smoking, drinking, no exercise etc, and noted that the strongest correlation with red meat and cancer was in obese insulin resistant individuals (both strong risk factors for cancer). These results do not show that meat consumption alone was responsible for cancer.

• So what happened with that information? Sweeping proclamations were made that red meat causes cancer to the same degree as cigarettes and the media ran with it.

  • This is an example of why higher quality interventional studies may be needed as they establish mechanisms of two things being causally related rather than simply being correlated (correlation does not equal causation).

Studies omitted from IARC Report included:

• Asian epidemiology showing decreased cardiovascular/cancer mortality with those that consumed more meat.

• Colon cancer rates found to be higher in vegetarians in a UK observational study.

• All interventional animal studies showing no increased risk of colon cancer.

• One study even showed bacon being protective from colon cancer in rats. Is it? That would be nice, but you wouldn't go ahead and base your diet off of that would you?

• With associational data described above, you must ask does this apply to everyone in all situations?

• Was diet the main reason for favorable or unfavorable health outcomes? Or was it the overall lifestyle?

• How do all the countless variables and factors in someone’s life (besides diet and nutrition) impact health? Gender, genetics, prenatal/childhood conditions, marital status, socioeconomic status, education, ethnicity/migrant status, lifestyle, and medical technology access all play large roles in overall health alongside nutrition.

• Perhaps the reason many nutrition recommendations are so contradictory is due to lack of consideration of these factors, or difficulty extrapolating these factors to health outcomes.

  • There are also cell and animal studies sometimes cited that don’t often extrapolate to humans, or short-term intervention studies that look at certain lab markers that experts THINK may link to some diseases, but have not been proven effective in the long-term - but click bait headlines may sensationalize these findings and imply that they can be generally applied to everyone.

• The issues above are likely what has led to a plethora of dietary camps waging war on each other, and people feeling they need to track calories, macros (carbs/fats/protein), micronutrients, and rely on biological tracking devices (not that those things aren't useful).

• The public's confidence in making healthy food choices has been lost and nutrition research and influencers have created a vast amount of confusion and apathy towards what foods we as a population generally choose to consume.

• I feel this apathy and confusion created from the conflicting nutrition recommendations of what to eat is IDEAL for processed food companies who take that confusion and use it to manipulate consumers into eating their disease promoting, food-like products.

• For proof of this I think all you need to do is look around at the general population and how overweight, symptom laden, and sick a disturbing amount of people are, and the massive profit margins of processed food makers. It's estimated in the US that 88% of the population do not have good metabolic health!! And I'm sure Canada wouldn't fare much better if equivalent statistics were determined.

When you come across a nutritional claim or headline, take it with a grain of salt and look into what kind of studies the claim was based on, and try to think about it in a variety of contexts.

• With nutrition research it’s generally best to rely on randomized control trials (RCT’s) with humans that show favorable outcomes (weight loss, improved mood, reduced inflammation, reduced autoimmune reactions etc), rather than using animal studies, cell studies, and observational studies.

• Even better than one or two RCT’s would be a systematic review or meta-analysis summarizing several RCT’s. Although, relying on higher quality data (according to the hierarchy) can also have its downsides, as long term RCT’s would be virtually impossible in the context of nutrition and its impact on health over decades like observational data can show to some degree.

• This is why with nutrition research, one could argue that nutrition related observational data (nutritional epidemiology) could actually paint a more accurate picture of the health impact of foods in some cases, as they can point towards longer term health outcomes if other health outcome variables are well accounted for (although as mentioned above this is a very difficult task).

So what on earth do we do with all of this? How should we eat for optimal health? Who do you trust for nutrition advice? See Part 2 for the answers!