Dairy and (Pre-)Diabetes - Re-Evaluated: It's Complicated -- Generally Protective Effects, Though (Risk Reduction ~40%)

If you lump all forms of dairy together into the "total dairy" category, US citizens have a ~40% reduced risk of prediabetes when they consume more than 14 servings of dairy per week. And that's by no means the only novel insight from a recent study by USDA researchers...

When women tell you "it's complicated" that's usually a sign you're in trouble. When scientists tell you "it's complicated", it's usually a smart version of telling you: "We don't know nothing, bro..." With that being said, I didn't find the line "it's complicated" in the latest paper by scientists from the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University (Hruby 2017) and still I cannot shake the feeling that... you guessed it: "It's complicated" ;-)

Yet even though it is obviously "complicated", the US dietary guidelines have consistently recommended 2–3 servings dairy/d for adolescents and adults (Office of Disease Prevention). It is thus no wonder that many individuals may perceive dairy products as health foods.... low fat that is; after all, we've been told that fat makes you fat for decades.
Have we been duped for decades? Not really, dairy products may be, as Hruby et al. point out, an important source of protein and several shortfall nutrients (nutrients that may be under-consumed relative to the Dietary Reference Intakes), including calcium, magnesium, potassium, and vitamin D.

In spite of that, the evidence in favor / against the anti-diabetic effects of dairy are... yeah, you heart it already "complicated". As the Tufts researchers point out in the introduction to their recently published paper "The Journal of Nutrition", existing observational evidence suggests a generally small, equivocal, or U-shaped relation between total dairy or specific dairy product consumption and the risk of T2D and related factors. What has been largely ignored, hitherto, is whether dairy is also associated with the incidence of impaired glucose or hyperglycemic stages preceding T2D. Reason enough to do yet another study to... make things more complicated ;-)

"Therefore, the primary aim of this study was to investigate the relations between consumption of dairy and milk-based products and the long-term risk of prediabetes among initially healthy individuals, and the risk of T2D among individuals with prediabetes in a cohort of middle-aged adults, the Framingham Heart Study Offspring Cohort. As a secondary aim, we examined the effect modification of dairy intake and risk of T2D by baseline glycemic status in the combined study population. We hypothesize that associations with incident outcomes differ by dairy product and type, and that these associations may further differ in initially healthy compared with unhealthy participants" (Hruby 2017).

The data stems from The National Heart, Lung, and Blood Institute's Framingham Heart Study Offspring Cohort, a community-based longitudinal study of cardiovascular disease that began in 1971. The scientists used data from the fifth examination cycle (1991–1995) to the eighth exam (2005-2008) of the Offspring Cohort, in which 3799  participants underwent a standard medical examination consisting of laboratory and anthropometric as well as dietary intake assessments and here's what they found:

(Pre-)Diabetes is RAMPANT - 48 in 100 people affected!

Initially, 1867 participants were free of prediabetes, of these almost every second participant, i.e. 902 (48%), developed prediabetes. Only 21% of these developed full-blown T2DM, though.

Figure 1:Diabetes is the world's eighth biggest killer, accounting for some 1.5 million deaths each year. A major new World Health Organization report has now revealed that the number of cases around the world has nearly quadrupled to 422 million in 2014 from 108 million in 1980. The Eastern-Mediterranean region had the biggest increase in cases. Diabetes now affects one in 11 adults w/ high blood sugar levels linked to 3.8 million deaths per year. (statista.com)

Against that background, it's no wonder that the diabetes rates worldwide keep exploding, as you can see in Figure 1 from one of my favorite websites, i.e. statista.com, which also tells you that the T2DM drug market is going to be worth 42.2 billion(!) $US in 2020 ... and here's the shocker $38.8 billion will be paid by the US, alone (learn more)!

Cheese, cream,  butter are not associated w/ pre-diabetes, protect against full-blown T2DM

While cheese, cream, and butter are still cognitively linked to metabolic problems in many mainstream articles, neither cheese nor cream and butter was associated with prediabetes in the study at hand. In fact, when the scientists looked at those 21% of pre-diabetics that transitioned into full-blown T2DM, the intake of high-fat dairy and cheese products (not sign. for cream and butter, alone | p = 0.18 for the fully adjusted model - for adjustments see caption of Figure 2) protected people from losing the last insulin sensitivity they'd left.

Figure 2: Differential risk (vs. 0 intake) of becoming a diabetic with high intakes of cream and butter, cheese (was evaluated only for 1-4 and ≥4) and total high-fat dairy intake; data fully adjusted for r age, sex, and energy intake, family history of diabetes, smoking, dyslipidemia, hypertension, other dietary characteristics, BMI, and weight change (Hruby 2017).

In other words, the study at hand is yet another study (of the more recent ones) showing a dose-response, inverse associations with incident T2D, with 70% and 63% lower risk, respectively, of incident T2D between the highest and lowest intake categories (≥14 compared with <1 serving/wk for high-fat dairy, ≥4 compared with <1 serving/wk for cheese).

Dairy is generally protective

Overall, it can be said that The subjects' total dairy intake - for both, low-fat, and high-fat dairy - was associated with a 39%, 32%, and 25% lower risk of incident prediabetes, respectively, in the highest compared with the lowest intakes (≥14 compared with <4 servings/wk).

Figure 3: Differential risk (vs. zero intake) of prediabetes with high intakes of total dairy (could not be evaluated for <4) and total high-fat and low-fat dairy intake; data fully adjusted for r age, sex, and energy intake, family history of diabetes, smoking, dyslipidemia, hypertension, other dietary characteristics, BMI, and weight change (Hruby 2017).

As you can see in Figure 2 total dairy, high-fat dairy and low-fat dairy show the same trend: the more you eat, the less likely you're to develop problems with your glucose management.

Cheese is actually a healthy food! In case you still have your doubts that real cheese (the one made from dairy instead of vegetable fat and E-numbers) is a healthy food, (re-)read my article "Cheese & Your Health: CVD, Cancer & Metabolic Syndrome - Cheesy Science or Scientific Revelation? A Brief Review" from July 2016, here. These trends, however, are not all of the same significance; with p = 0.002, the chance that the association the scientists observed for total dairy is 15x less likely to be random than the one for either high-fat or low-fat dairy - while this may be a logical consequence of the size of the dataset, it could also be interpreted as further evidence that the type of dairy doesn't really matter.

Yogurt doesn't have its super rep for nothing 

In fact, the differential analysis for milk products from Hruby et al. supports the notion that yogurt is particularly (pre-)diabetes-protective, but total, low-fat and skim milk, and whole-milk are associated with significantly reduced risks of prediabetes, as well.

Figure 4: Differential risk (vs. zero intake for skim milk+yogurt) of prediabetes with moderate intake of total milk, skim milk and yogurt; whole milk intake left out because there was only one intake category; for adj. see Figure 2 & 2 (Hruby 2017).

In that, it should be noted, however, that these risk reductions do not increase linearly; this means, it's not "the more the better". Rather than that, a moderate intake of all four was associated with the greatest relative risk reduction, namely -30% for total milk, -17% for skim milk, -16% for whole milk and -24% for yogurt.  In that, I find it quite telling that the overall intake of whole milk in the US (or, rather, in this cohort) is so low that the scientists could compare data of subjects consuming zero to one vs. those consuming more than one serving per week, only. The anti-fat scare obviously continues to have an effect.

What do previous studies show?

If you take a look at the most recent reviews/meta-analysis, overall decreases in diabetes risk (Gijsbers 2016, esp. yogurt (-14% w/80g/d); Khoramdad 2017 (-8-12% for total dairy); Guo 2017 (-15% all dairy; also -27% for metabolic disease)), reduced CVD and stroke risk (Alexander 2016; de Goede for milk (-7% per 200g); Pimpin 2016 for butter) are the norm, not the exception. The study at hand should thus be understood as part of the accumulating evidence in favor of dairy products for metabolic disease and its unhealthy consequences on the heart.
Where are the experimental trials? Good question: compared to the unlimited number of observational studies, reviews and meta-analyses, there are only very few intervention studies. This wouldn't be the SuppVersity, though, if we didn't, at least, take a peek at the underlying mechanisms and you know that you won't get to those with epidemiological guesswork.

"Emerging evidence indicates that dairy components that alter mitochondrial function (e.g., leucine actions on silent information regulator transcript 1 (SIRT1)-associated pathways), promote gut microbial population shifts, or influence inflammation and cardiovascular function (e.g., Ca-regulated peptides calcitonin gene-related peptide [CGRP] or calcitonin) should be considered as possible mechanistic factors linking dairy intake with lower risk for T2D" (my emphasis in Hirahatake 2014). 

Furthermore, there's the obvious: the potential to increase muscle mass and thus the ability to burn and store glucose that comes with the high BCAA content of dairy proteins (also prevents sarcopenia | McGregor 2013). What relatively few people know, though, is that dairy also contains functional, bioactive peptides, i.e. amino acid chains that activate or block certain receptors in your body:

"Whey protein and casein both contain the bioactive peptides lactokinins or caseinkinins respectively. In-vitro experiments indicate these peptides can inhibit ACE activity. ACE is a rate limiting enzyme in the conversion of angiotensin I to angiotensin II responsible for vasoconstriction. Therefore, lactokinins or caseinkinins both have potential to lower blood pressure. Vascular reactivity is important for glucose disposal and regulation of blood flow, but is impaired in patients with metabolic syndrome due to decreased insulin-stimulated vasodilation via the eNOS pathway and suppressed NO production in the vascular endothelium" (McGregor 2013). 

Now you may say that this is not about diabetes... however, ACE inhibitors will improve your glucose metabolism - in fact so much so that diabetics end up in the emergency room with hypoglycemia (Morris 1997).

Figure 5: The renin-angiotensin system controls much more than just your blood pressure (Henriksen 2007).

In addition to that, dairy protein-derived peptides may also contribute to the insulinotropic effect via dipeptidyl peptidase-4 (DPP-4) inhibitory activity (Bjørnshave 2014).

Another, but IMHO overrated contributor to dairy's health effect is the CLA content of dairy. In particular with respect to diabetes, the focus of this SuppVersity article, CLA is yet rather problematic.  That's in contrast to dairy's effects on the "satiety" or rather incretin hormones glucagon-like peptide-1 (GLP-1) and gastric inhibitory polypeptide (GIP), of which we know from various studies that they help maintain glucose homeostasis, in part by enhancing pancreatic beta-cell responsiveness to glucose (Freeman 2009).

As much as 25% of the postprandial insulin response has been suggested to be attributable to the effects of incretins and dairy is a rockstar, when it comes to their release.

As Hirahatake et al. point out, "[v]arious lines of evidence indicate that GLP-1 production by small intestinal and colonic L-cells is influenced by specific nutrients, and may respond to foods that provide a substrate for certain species of colonic microbiota". Next to dairy, you'll also find prebiotic fibers and resistant starches on the list of GLP1 promoters. This points towards the last, but not least important factor that may contribute to the glucose-normalizing prowess of dairy: the microbiome; and I guess I don't have to tell you that some dairy products have the potential to alter gut microbiota through probiotic effects of live cultures used during fermentation, and others may do so through the prebiotic effects of dairy oligosaccharides (Faye 2012).
References:

• Alexander, Dominik D., et al. "Dairy consumption and CVD: a systematic review and meta-analysis." British Journal of Nutrition 115.4 (2016): 737-750.
• Bjørnshave, Ann, and Kjeld Hermansen. "Effects of dairy protein and fat on the metabolic syndrome and type 2 diabetes." The review of diabetic studies: RDS 11.2 (2014): 153.
• Chen, Mu, et al. "Effects of dairy intake on body weight and fat: a meta-analysis of randomized controlled trials." The American journal of clinical nutrition 96.4 (2012): 735-747.
• de Goede, Janette, et al. "Dairy consumption and risk of stroke: a systematic review and updated dose–response meta‐analysis of prospective cohort studies." Journal of the American Heart Association 5.5 (2016): e002787.
• Faye, Tamburello, et al. "Survival of lactic acid bacteria from fermented milks in an in vitro digestion model exploiting sequential incubation in human gastric and duodenum juice." Journal of dairy science 95.2 (2012): 558-566.
• Freeman, Jeffrey S. "Role of the incretin pathway in the pathogenesis of type 2 diabetes mellitus." Cleveland Clinic journal of medicine 76 (2009): S12-9.
• Guo, Jing, et al. "Milk and dairy consumption and risk of cardiovascular diseases and all-cause mortality: dose–response meta-analysis of prospective cohort studies." (2017): 1-19.
• Harrison, Leonard C., and Margo C. Honeyman. "Cow's milk and type 1 diabetes: the real debate is about mucosal immune function." Diabetes 48.8 (1999): 1501-1507.
• Henriksen, Erik J. "Improvement of insulin sensitivity by antagonism of the renin-angiotensin system." American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 293.3 (2007): R974-R980.
• Hirahatake, Kristin M., et al. "Associations between dairy foods, diabetes, and metabolic health: potential mechanisms and future directions." Metabolism 63.5 (2014): 618-627.
• Hruby, Adela, et al. "Associations of Dairy Intake with Incident Prediabetes or Diabetes in Middle-Aged Adults Vary by Both Dairy Type and Glycemic Status." J.Nutr. First published August 2, 2017, doi: 10.3945/ jn.117.253401
• Khoramdad, Malihe, et al. "Dairy Products Consumption and Risk of Type 2 Diabetes: A Systematic Review and Meta-Analysis of Prospective Cohort Studies." diabetes 12 (2017): 13.
• McGregor, Robin A., and Sally D. Poppitt. "Milk protein for improved metabolic health: a review of the evidence." Nutrition & metabolism 10.1 (2013): 46.
• Office of Disease Prevention and Health Promotion, U.S. Department of Health and Human Services. 1990 Dietary guidelines [Internet]. 1990. [cited 12 Sep 2016]. Available from: https://health.gov/dietaryguidelines/1990.asp
• Office of Disease Prevention and Health Promotion, U.S. Department of Health and Human Services. Dietary guidelines for Americans 2015–2020 [Internet]. 8th ed. 2015. [cited 2016 Jan 7]. Available from: http:// health.gov/dietaryguidelines/2015/guidelines/.
• Pimpin, Laura, et al. "Is butter back? A systematic review and meta-analysis of butter consumption and risk of cardiovascular disease, diabetes, and total mortality." PLoS One 11.6 (2016): e0158118.
• Zivkovic, Angela M., and Daniela Barile. "Bovine milk as a source of functional oligosaccharides for improving human health." Advances in Nutrition: An International Review Journal 2.3 (2011): 284-289.