Thursday, July 6, 2017

Red Light for Red/Processed Meat Actually a Red Light for Obesity? Body Fat Mediates Link of Meat Intake & Health

Mediation analysis suggests: It's not the amount of Cantonese Roast Pork Belly (recipe) you eat that increases the level of inflammation in your body, but the belly you get if you eat too much of it or non-red/processed meat foods.
You will be aware of the fact that even the best epidemiological study cannot account for all correlations; correlations that mess with the results of the studies; correlations like the one between generally healthier lifestyles and the consumption of a vegetarian diet, which have been thwarting the results of epidemiological studies on the effects of meat intake on our health for decades.

To get to the bottom of one of these spurious correlations, scientists from the University of Hawaii Cancer Center (Chai 2017), examined the associations of dietary red and processed meat intake with serum levels of CRP, TNF-a, IL-6, leptin, and adiponectin among control participants in 2 nested case–control studies of cancer in the Multiethnic Cohort.
Looking for more ways to improve your diet? Increase your potassium (K) intake!

Potassium vs. Diet-Inducded Insulin Resis.

In the Lime Light: The Ill Effects of Low K Intakes

Bad News: Most Americans are Sign. K Deficient

Lean, Healthy ... Correlates of High Hair Potassium

Eating a High Protein Diet? Better Watch K!

Potassium Bicarbonate = Anabolic!?
In that, they also sought to determine whether the associations between red and processed meat intake and biomarker levels, if present, were mediated by BMI and were consistent across race/ethnic groups in the MEC.
"The study's biospecimen subcohort of 67,594 cohort members was established in 2001–2006 by asking surviving cohort members to provide blood and urine specimens. When com paring the characteristics at cohort entry of individuals who provided specimens with those who did not, there were no substantial differences between the 2 groups in BMI, dietary fat and vegetable intake, physical activity, and family history of cancer, suggesting that the biospecimen repository partic ipants are broadly representative of the cohort members" (Chai 2017).
The subjects' dietary intakes were calculated based on a self-reported food frequency questionnaires. The subjects reported their average intake in 7 categories and serving sizes of specific foods during the last year; dietary intakes were calculated using a food composition table that included habitual foods for all race/ethnic groups in the study.
Table 1: Characteristics of study participants (Chai 2017).
Multivariable linear models were applied to assess the association between red and processed
meat intake at cohort entry and serum biomarker levels measured 9.1 years later after adjusting for
covariates and to determine the mediator effect of BMI.
Figure 1: The mediator effect of BMI on associations between red and processed meat intake (red meat) and serum leptin, adiponectin, and CRP levels (see Figure 2 for a more comprehensible illustration| Chai 2017).
Let's first take a closer look at the statistical shenanigan the authors used to come up with the results they illustrated in four small diagrams (Figure 1 A-D).
Minimally Processed and Eaten as Part of a Mediterranean-Style Diet, Red Meat Augments MED's Heart Health Benefits | read more
"Three multivariable linear models were applied: (1) In model 1, biomarker levels (outcome variable) were regressed on red meat (predictor variable); (2) in model 2, BMI (mediator) was regressed on red meat (predictor variable); (3) in model 3, biomarker levels (outcome variable) were regressed simultaneously on red meat (predictor variable) and BMI (mediator).

The mediator effect of BMI was estimated as follows: (b for red meat in model 1 [model not including BMI]) ¡ (b for red meat in model 3 [model including BMI]). The percentage change in effect for the mediator effect of BMI was estimated as follows: % change in effect D [exp (b in model 1) ¡ exp (b in model 3)]/exp (b in model 3) x 100%. Variables for red meat, BMI, leptin, adiponectin, and CRP were log-transformed in all analyses."
I guess this leaves you with the question: What the f*** does all that mean? Well, in simplified terms, the authors ran their data through three different models.
  • Model one was based on the assumption that its the red and processed meat (predictor variable) mess with your metabolic health by an inertly 'meaty' mechanism.
  • Model two, on the other hand, built around the assumption that the correlation between high red and processed meat intakes and high levels of serum markers of inflammation are mediated by the subjects' BMI (predictor variable).
  • Model three, finally, incorporated both, meat intake (predictor variable) and BMI (mediator), simultaneously to predict the biomarker levels
In line with previous research that used this mediation analysis (which are an extension of the regular "adjustments" scientists make to their data) to successfully investigate the true underpinnings of psychological health issues (Baron 1986), mediation was established if...
  • significant  associations  (p < 0.05) are observed in the first and second models and 
  • two additional criteria are met in the third model
Said two additional criteria were (a) the BMI (mediator) must significantly predict outcome variable biomarker levels (p < 0.05) and (b) the direct relationship between the predictor and the outcome variables (as indicated by as estimated slope for red and processed meat in model 3) decreases to 0 in case of full mediation or is reduced substantially but different from 0 in case of partial mediation. In less sciency terms: If you see no systematic increase in the biomarkers of inflammation with higher meat intakes you have been able to prove that it's all down to obesity.
Figure 2: Many (Leptin in both, CRP and adiponectin in women), but not all of the inflammatory markers lose their statistical significance when the scientists compensated for a mediating effect of BMI (Chai 2017).
If, on the other hand, there is still a significant increase in said biomarkers w/ increased meat intakes this would suggest that only part of the ill health effects of red meat are mediated by the big bellies of meat lovers ... or, in other words, red and processed meat is bad, but it's not so bad for those who maintain a healthy body fat level (despite eating it regularly).

The study at hand does not fully acquit red and processed meat, but...

... the loss of statistical significance and the relevant reduction of the beta-coefficients, i.e. the link between increased meat intakes and leptin, adiponectin, CRP, TN, -alpha and IL-6, still suggest that the ill health effects of red meats are, at least partly, mediated by increased body fatness.
Table 2:  Red and processed meat intake (g/day) at cohort entry and serum biomarkers after 9 years by sex (Chai 2017).
In that, it is important to note that the absence of adiponectin and CRP in men and TNF-alpha and IL-6 in women from the list of predictors that are no longer predictive of inflammation in Figure 1 & 2 does not imply that these predictors were not mediated by obesity. In fact, the data in Table 2 tells you that red and processed meat intakes are no statistically significant predictors of said biomarkers in the first place.

Especially, whether and to which extent that invalidates the potential ill health effects of TMAO formation (Trimethylamine-N-oxide is formed by bacteria in your gut, so any studies on the health effects of TMAO would have to investigate interactions with the microbiome, as well) and other bioactive substances that are formed during the digestion or processing of meat still deserves further investigation in controlled clinical trials.
This is not an outlier study, but it's news: The findings of the study at hand are not without parallels in previous studies. As Chai et al. point out,  a cross-sectional study, Ley et al. (2014) reported that red meat intake was positively associated with serum CRP and inversely associated w/ adiponectin levels in 3690 diabetes-free females. However, the associations were attenuated for CRP after adjusting for BMI and for adiponectin after adjusting for medical and lifestyle factors. In a similar vein, the association between circulating high-sensitivity CRP levels and red meat intake in Montonen et al. was lost after adjustment for BMI. In contrast to Chai et al. none of the previous studies conducted a thorough mediation analysis to evaluate the role of BMI, though.
Yes, you read that right: There was no statistically significant link between baseline meat intake in grams per kg body weight and adiponectin, CRP, TNF-alpha and IL-6 in men or CRP, TNF-alpha and IL-6 in women to begin with - even before the mediating effects of BMI were investigated.

After adjusted for body mass index at cohort entry, even the existing links between meat intake and leptin in men and women and adiponectin and CRP in women vanished (compare Table 2, Model 1 vs Model 2 for the respective biomarkers).

All that is not "hard" (=experimental) evidence, like the recently discussed Med-Diet study or Hodgson's 2007 study that found an increased intake of lean red meat to leave CRP unaffected in healthy individuals, but the scientists' mediation analysis for BMI (basically an advanced form of adjusting for BMI) seems to confirm what critics have previously highlighted: The ill effects of a high red and processed meat intake scientists have repeatedly observed in epidemiological studies may potentially be mediated by the obesity of meat lovers - an obesity that may, in turn, be related if not triggered by a high intake of processed foods with red and processed meats in them... but I guess that's something that would need investigation in a follow-up study | Comment!
References:
  • Chai, Weiwen, et al. "Dietary Red and Processed Meat Intake and Markers of Adiposity and Inflammation: The Multiethnic Cohort Study." Journal of the American College of Nutrition (2017): 1-8.
  • Hodgson, Jonathan M., et al. "Increased lean red meat intake does not elevate markers of oxidative stress and inflammation in humans." The Journal of Nutrition 137.2 (2007): 363-367.
  • Ley, Sylvia H., et al. "Associations between red meat intake and biomarkers of inflammation and glucose metabolism in women." The American journal of clinical nutrition (2014): ajcn-075663.