Regular alcohol drinkers have lower risk of diabetes, according to a huge new study


There’s a new checkmark in the ‘drinking isn’t all bad for
you’ column.

According to a
new study that looked at more than 70,000 Danish people
those who drink small to moderate amounts of alcohol on a
frequent basis are less likely to develop diabetes than people
who don’t drink at all. 

To be clear, these results shouldn’t be seen as license or
encouragement to drink freely as a health-promoting exercise.

But they do provide further evidence that, for some reason,
people who drink moderately are less likely to suffer from
certain illnesses, including some cardiovascular
and type-2 diabetes.

Regular drinking and diabetes

For the new study, researchers wanted to see how much alcohol
consumption was associated with the lowest diabetes risk, and
determine whether the type of alcohol or the frequency that
people drank mattered.

Using data from the Danish Health Examination Survey, they looked
at the drinking habits of 28,704 men and 41,847 women, and
tracked whether those people developed diabetes within
approximately five years. The researchers excluded
anyone who already had diabetes, was pregnant at the
start of the study, and didn’t provide information on their
alcohol consumption.

The results showed that the study participants least likely to
develop diabetes drank 3-4 days a week. For men, those who drank
14 drinks per week had the lowest risk, as the chart on the left
shows below. For women, those who drank nine drinks per week
had the lowest risk, as the right-hand chart shows.

Diabetes drinking risk curve
for diabetes in 28,704 men (a) and 41,847 women (b) from the
general Danish population according to average weekly alcohol
amount. Dotted lines represent 95% confidence


As the U-shaped risk curve shows, study
participants who didn’t drink at all seemed to
have a higher risk of developing diabetes. People who
drank moderately had a lower risk, up to a certain
point — after that, risk started to rise again. Even heavy
drinkers (up to 40 drinks per week for men and 28 drinks per week
for women), however, still had a lower risk of developing
diabetes than teetotalers.

The lowest risk was associated with drinking that
was spread out throughout the week, rather than occurring in
the same day or two.

The type of alcohol mattered too. Men and women who drank wine
had the lowest diabetes risk. For men, beer was also associated
with a lower risk. Spirits didn’t seem to affect risk for men,
but women who drank seven or more drinks of spirits a week had an
increased risk of developing diabetes.

A brief but important aside on
: The design of this study didn’t allow researchers
to say whether drinkers had a lower risk of developing type-2
diabetes or type 1. Type 2 is generally caused by lifestyle
factors and prevents the body from using insulin, where as type 1
cannot be prevented since the body simply doesn’t produce
enough insulin. The researchers say their
study should refer to type-2 diabetes, since their results
held true even if they eliminated anyone under 40 (by which
point the vast majority of people with type-1 diabetes already
have it).

So what’s going on here?

Tempting as it might be to say that drinking lowers diabetes
risk, we can’t say that. All we know is that people — Danes, at
least — who drink regularly develop diabetes less frequently.

It’s possible that this is because people who drink in moderate
quantities tend to be healthier in the first place than people
who don’t drink at all. The researchers tried to calculate for
these effects — they accounted for things like body
mass index, physical activity, smoking status, and family
history — but it’s always possible that results were still
skewed in some way.

woman sipping drinking wine
woman tastes wine during the Vinexpo Asia-Pacific in Hong Kong,
the region’s largest international wine and spirits


There is a
hypothesis that moderate drinking may improve
some aspects of
health by lowering blood pressure and dilating blood vessels, but
it’s not certain whether that plays a role.

There are a number of other complicating factors, too. On the one
hand, most people under-report their drinking, meaning that
people may actually be drinking more than they reported. Also,
this was a population study in Denmark. Different results might
be found in non-Scandinavian populations (especially non-white
groups, many
which have a higher risk of developing diabetes

When it comes to alcohol and health, we know that drinking too
much isn’t healthy.
Alcohol consumption has been shown to increase the
 of certain cancers — a recent new report found
a link between an increased risk of breast cancer and drinking as
little as one glass of wine or beer each day.

The researchers behind this study aren’t advocating for drinking
as a means of health promotion. But at least in regard to
diabetes, drinking what’s considered a moderate amount throughout
the week seems to be fine.

Do you have an unusual story to tell? E-mail