Tuesday, October 2, 2012

Fat Advantage: 61% Lower Rates of Metabolic Syndrome in High Fat, 101% Higher Rates in Low Fat Dairy Lovers

"Got milk" is not the question health conscious supermen and -women should pose. "Got full fat milk, fermented dairy and cheese" is the line to remember (the original image was part of the "Got Milk Campaign")
While diet fads come and go, the advice the wise (not seldom obese or otherwise sick) experts on the boards and panels of our well-meaning governments is calling "dietary recommendations" is about as resistant to reform as the dreaded MSRA strains are to the antibiotics doctors are throwing at you whenever you sneeze. Against that background the recent trend we are seeing with respect to an increase in the recommended amount of dietary protein does almost amount to a quantum leap; a leap with a significant caveat, however. A fat caveat, so to say:
"A healthy diet includes [...] lean meats, poultry, fish, beans and fat-free or low-fat dairy products" (NIH. 2012).
Luckily, you as a SuppVersity reader do not have to rely on the NHI's thwarted interpretations of the latest research they claim to use, when they are "turning discovery into health" (no joke, this is a literal citation from the footer of the NHI website!), but can compare it to my thwarted interpretations of the latest research and cherry picked data ;-)

Cherry pick of the day: Longitudinal large scale study on dairy intake and metabolic health

 For Today, this means that you get to enjoy the latest results of a large scale observational study from the University of Sydney that's based on datasets from the Blue Mountains Eye Study (BMES) a population-based cohort study of common eye diseases and other health conditions in residents aged 49 years and over in the Blue Mountains area, west of Sydney. A longitudinal study the baseline information was obtained in 1992/1994 from  and complemented by follow-up ten years later.

The data sets included food frequency questionnaires, as well as anthropometric and biochemical assessments all of which were included in the present analysis of the association betweenn dairy consumption with the ten-year incidence of Metabolic syndrome (MetSyn) and type 2 diabetes. What's so special about this dataset, is that the food questionnaires were actually detailed enough to assess the effects of full- and low-fat dairy, separately - a very important advantage, as a cursory glance at the data in figure 1 reveals.
Figure 1: Odds ratios (95% confidence intervals) of incident metabolic syndrome according to quartiles of reduced/low fat,
regular fat and total dairy product intake (data based on Louie. 2012; adjustments for age and sex (basic model), smoking status, physical activity (metabolic equivalents), dietary glycemic load, fibre from vegetables, total energy intake and family history of type 2 diabetes (model 1) and calcium (model 2))
While the standard analysis for total dairy consumption (figure 1, left) yielded neither conclusive, nor statistically significant results (the p-values for the different models can be found in the upper right corner of the respective graphs). The categorization into low- and high fat dairy and the adjustments for age and sex (basic model), smoking status, physical activity (metabolic equivalents), dietary glycemic load, fibre from vegetables, total energy intake and family history of type 2 diabetes (model 1) and calcium (model 2) yields very clear and, after adjustment for calcium intake, pretty unflattering result the formulators of the afore mentioned "dietary recommendations" will probably file in their already bristling "statistical outliers"-folder:
  • after adjustment for calcium intake subject in quartiles 2 / 3 / 4 of low-fat dairy are 50% / 145% / 101% more likely to be struck by metabolic syndrome, than those in the lowest quartile of low fat dairy intake (p = 0.043), while
  • subjects in the highest quartile of full-fat dairy intake are - depending on the adjustments made -  48% / 59% / 61% less likely (base model / model 1 / model 2) to suffer frommetabolic syndrome, than those in the lowest quartile of high fat dairy intake (p-values:  0.018 / 0.004 / 0.004)
Yet while the scientists are well aware, that these results stand in stark contrast to the initially cited dietary recommendations, is it not this contrast that surprises them, but rather the fact that a similar significant benefit was not observed for type II diabetes, which is, after all, one of the hallmark features of the rather loosely defined triad of obesity, insulin resistance and cardiovascular disease(s), we usually refer to as 'metabaolic syndrome':
"Due to its higher saturated fat content, regular fat/high fat dairy products were previously believed to increase the risk of type 2 diabetes as a high saturated fat intake is associated with insulin resistance . However, cohort studies and a meta-analysis now suggest otherwise, with higher regular fat/high fat dairy consumption being considered mostly neutral or protective for type 2 diabetes. The results of the present study are consistent with these findings that higher regular fat dairy consumption may be protective of MetSyn and type 2 diabetes. The potential harmful effects of higher saturated fat from regular fat dairy products may have been offset by the protective components of regular fat dairy such as trans-palmitoleate, a fatty acid nearly unique to ruminant foods. Circulating level of trans-palmitoleate was shown to be significantly associated with reduced risk of type 2 diabetes (Q5 vs Q1: 62% reduced risk, p-trend < 0.001). Moreover, the protective effect of trans-palmitoleate may be exerted via the suppression of hepatic fat synthesis, where the latter was strongly associated with insulin resistance." (Louie. 2012; my emphases)
In view of these mechanism, it is all the more surprising that the study at hand and many previous studies didn't find any significant correlations between (regular fat) dairy intake and type II diabetes.

Reduction in metabolic risk, but no effect on type diabetes? Hold on...

Wolverine could be the only face of the "Got Milk" campaign who does not have to care about potential negative health effects of homogenized milk.
And upon a cursory read of the latest literature it does in fact seem as if "null findings" like this, were nothing special. Only recently by Sluijs et al. who had analyzed datasets from a nested case-cohort within 8 European countries of the European Prospective Investigation into Cancer and Nutrition Study (n = 340,234; 3.99 million person-years of follow-up) includind a random subcohort (n = 16,835) and incident diabetes cases (n = 12,403; cf. Slujis. 2012):
"This large prospective study found no association between total dairy product intake and diabetes risk. An inverse association of cheese intake and combined fermented dairy product intake with diabetes is suggested, which merits further study." (Sluijs. 2012)
If we do yet take a closer look at the actual results the actually not so surprising truth is that there was a statistically significant inverse association with diabetes for cheese (p = 0.01) and fermented dairy (p = 0.02).

An association that suggests a 12% reduction in diabetes risk in those study participants who consumed the most cheese and fermented dairy (cheese, yogurt, and thick fermented milk)

And since you all know your real foods, I guess I don't have to tell you that despite the fact that there are low fat varieties of cheese yogurts and other fermented milk products, 90% of them contain way more than the 1.5% let alone 0.1% fat the allegedly healthy low fat "milk" is boasting of. Mere coincidence? I don't think so. Reason to assume that low-fat milk will make you sick? No, but certainly not an argument to avoid the full-fat variety simply because it contains fat (which is the only argument the average dietitian has to favor low- over full-fat dairy products).

Mutant Milk!? New Research Fuels the Flames on Hushed Up Concerns About Ill Health Effects of Homogenized Milk 
Ask Dr. Andro: Are Colostrum and Milk Products in General Healthy Muscle Builders, a Waste of Money or Toxic Waste?
All about milk: Browse past news and articles at the SuppVersity
^ Suggested reads
Additional recent dairy science: Similar beneficial findings for all-cause mortality and fermented dairy (yet inconclusive results for CVD and diabetes) come from the recently published Whitehall II study (4526 subjects,72 % men, mean age 56 years; Soedamah-Muthu. 2012) and for dairy intake during adolescents and diabetes (-38% risk reduction for 2 servings per day or more) from a reanalysis of somewhat questionable data (who remembers exactly how much dairy he had during his adolescence?) from the Nurses' Health Study II cohort that comprises 37,038 women who completed a food-frequency questionnaire about their diet during high school were followed from the time of return of the questionnaire in 1998-2005 (Malik. 2012).

  • Louie JC, Flood VM, Rangan AM, Burlutsky G, Gill TP, Gopinath B, Mitchell P. Higher regular fat dairy consumption is associated with lower incidence of metabolic syndrome but not type 2 diabetes. Nutr Metab Cardiovasc Dis. 2012 Sep 26. pii: S0939-4753(12)00193-7. 
  • Malik VS, Sun Q, van Dam RM, Rimm EB, Willett WC, Rosner B, Hu FB. Adolescent dairy product consumption and risk of type 2 diabetes in middle-aged women. Am J Clin Nutr. 2011 Sep;94(3):854-61.
  • NIH. Health in the News: Love Your Heart. February 2012. < http://newsinhealth.nih.gov/issue/feb2012/feature1 > retreived Oct 02, 2012.
  • Soedamah-Muthu SS, Masset G, Verberne L, Geleijnse JM, Brunner EJ. Consumption of dairy products and associations with incident diabetes, CHD and mortality in the Whitehall II study. Br J Nutr. 2012 Jun 7:1-9.
  • Sluijs I, Forouhi NG, Beulens JW, van der Schouw YT, Agnoli C, Arriola L, Balkau B, Barricarte A, Boeing H, Bueno-de-Mesquita HB, Clavel-Chapelon F, Crowe FL, de Lauzon-Guillain B, Drogan D, Franks PW, Gavrila D, Gonzalez C, Halkjaer J, Kaaks R, Moskal A, Nilsson P, Overvad K, Palli D, Panico S, Quirós JR, Ricceri F, Rinaldi S, Rolandsson O, Sacerdote C, Sánchez MJ, Slimani N, Spijkerman AM, Teucher B, Tjonneland A, Tormo MJ, Tumino R, van der A DL, Sharp SJ, Langenberg C, Feskens EJ, Riboli E, Wareham NJ; InterAct Consortium. The amount and type of dairy product intake and incident type 2 diabetes: results from the EPIC-InterAct Study. Am J Clin Nutr. 2012 Aug;96(2):382-90.


  1. I would be much happier with a study where the feed rodents high and low fat Milk.

    Studies based on questionnaires about what people eat sound spurious at best.

    Who is most likely to eat "low fat" products? Overweight people.

    Who is most likely to end up with metabolic syndrome?
    Overweight people.

    You would likely find the same result with most "diet" foods.

    I see no evidence of causation between between low fat dairy, and metabolic syndrome from a "study" like this.

    1. And i don't suggest there is one

    2. on a related note: You are certainly right having a rodent study to investigate possible mechanisms would be nice, but if you look at what happens to the odds ratios with different adjustments, it appears that low fat dairy derives much of its advantages from its calcium content

  2. "And i don't suggest there is one"

    From your headline complete with exclamation marks, it certainly looks like you want to imply there is a causal link.

    Also I want a rodent test first to see if there is a causal link. You seem again to assumer there is and say it would be valuable to examine possible mechanisms, instead of see if this relationhip holds with actual testing.

  3. Yeah, my impression too...

    But rodent studies are equally problematic, just not in the same direction - determining how much fat in dairy does something to stressed, genetically selected and artificially kept alive mices probably means next to nothing for humans.

    On the other hand, the is so much theoretical advantage for high fat milk that I don't know where to start.

    1. Yes actual human feeding studies would be more beneficial than rodent. It is studies based on what people remember eating that have so many confounding factors that I consider them useless.

      Here is a human skim vs whole milk study:

      Lower Cholesterol in the skim group.

    2. problem with all those old studies: They simply measure the amount of cholesterol floating around. The advantage of fatty foods does however come from its beneficial effects on the cholesterol profile. Just high LDL levels don't make a reliable marker of CVD risk

  4. ok, ok you won. I did at least remove the exclamation mark :)

    that said. I would not expect any side effects in an appropriately controlled trial with plain low fat dairy. I would however consider it possible to see those in a trial using the low fat dairy products from the supermarket where every 0.1% of fat they remove is made up for with yet another E-number that will eventually give the low fat stuff a similar taste, mouth feeling, processing

    if you add to that the line of thinking: If I take the low fat cheese, I can add some more salami to it, because I saved on the fat with the cheese, we are again entering a territory, where rodent studies are not useful

    1. "If I take the low fat cheese, I can add some more salami to it, because I saved on the fat with the cheese, we are again entering a territory, where rodent studies are not useful"

      Actually isn't this exactly the reason we want to do rodent studies. So we can find out if low fat cheese is really a problem, and not have bad habits that might go with eating low fat cheese confounding the results.

    2. this is, of course, but on the other hand it is a major downside, when it comes to translating those results into practice

  5. Another great SuppVersity article. Very timely for me as I've just added dairy back into my day-to-day diet.

    Speaking of dairy, I was wondering what your thoughts are on this article:


    It looks to me like most or all of the changes in hormones (esp. LH, FSH & T) in the male subjects could be explained by circadian rhythmicity which, as you point out in another fine article, most researchers don't take into account in studies like this.

    What say you?

    1. I don't know, but if I did not miss something upon skimming the study, this study has actually exactly zero value, since there is no isocaloris control. Postprandially testosterone drops anyway... so what?

    2. Thanks, and both excellent points.

  6. "An association that suggests a 12% reduction in diabetes risk in those study participants who consumed the most cheese and fermented dairy (cheese, yogurt, and thick fermented milk)"

    Fermented dairy and cheeses are among the foods which contain relatively high amounts if vitamin K2 (gouda and brie are highest right after natto).

    According to this study (Dietary intake of menaquinone is associated with a reduced risk of coronary heart disease: the Rotterdam Study. J Nutr. 2004;134(11):3100-5.: http://jn.nutrition.org/content/134/11/3100.long) a diet rich in vitamin K2 reduces chances of diabetes type 2 (they also state this correlates to fat consumption).

    I think this is something to take into consideration along with your article :)