Alcohol consumption, gender, and type 2 diabetes: Strange … but true

Let me start this post with a warning about spirits (hard liquor). Taken on an empty stomach, they cause an acute suppression of liver glycogenesis. In other words, your liver becomes acutely insulin resistant for a while. How long? It depends on how much you drink; possibly as long as a few hours. So it is not a very good idea to consume them immediately before eating carbohydrate-rich foods, natural or not, or as part of sweet drinks. You may end up with near diabetic blood sugar levels, even if your liver is insulin sensitive under normal circumstances.

The other day I was thinking about this, and the title of this article caught my attention: Alcohol Consumption and the Risk of Type 2 Diabetes Mellitus. This article is available here in full text. In it, Kao and colleagues show us a very interesting table (Table 4), relating alcohol consumption in men and women with incidence of type 2 diabetes. I charted the data from Model 3 in that table, and here is what I got:


I used the data from Model 3 because it adjusted for a lot of things: age, race, education, family history of diabetes, body mass index, waist/hip ratio, physical activity, total energy intake, smoking history, history of hypertension, fasting serum insulin, and fasting serum glucose. Whoa! As you can see, Model 3 even adjusted for preexisting insulin resistance and impaired glucose metabolism.

So, according to the charts, the more women drink, the lower is the risk of developing type 2 diabetes, even if they drink more than 21 drinks per week. For men, the sweet spot is 7-14 drinks per week; after 21 drinks per week the risk goes up significantly.

A drink is defined as: a 4-ounce glass of wine, a 12-ounce bottle or can of beer, or a 1.5-ounce shot of hard liquor. The amounts of ethanol vary, with more in hard liquor: 4 ounces of wine = 10.8 g of ethanol, 12 ounces of beer = 13.2 g of ethanol, and 1.5 ounces of spirits = 15.1 g of ethanol.

Initially I thought that these results were due to measurement error, particularly because the study relies on questionnaires. But I did some digging and checking, and now think they are not. In fact, there are plausible explanations for them. Here is what I think, and it has to do with a fundamental difference between men and women – sex hormones.

In men, alcohol consumption, particularly in large quantities, suppresses testosterone production. And testosterone levels are inversely associated with diabetes in men. Heavy alcohol consumption also increases estrogen production in men, which is not good news either.

In women, alcohol consumption, particularly in large quantities, increases estrogen production. And estrogen levels are (you guessed it) inversely associated with diabetes in women. Unnatural suppression of testosterone levels in women is not good either, as this hormone also plays important roles in women; e.g., it influences mood and bone density.

What if we were to disregard the possible negative health effects of suppressing testosterone production in women; should women start downing 21 drinks or more per week? The answer is “no”, because alcohol consumption, particularly in large quantities, increases the risk of breast cancer in women. So, for women, alcohol consumption in moderation may also provide overall health benefits, as it does for men; but for different reasons.