New Diabetes Risk Factor: Antidepressants

If people knew that taking antidepressants for two years or more doubles their risk of diabetes, would they continue to take them? That's what I wondered when I saw the paper by Andersohn et al. in Am J Psychiatry 2009; 166:591–598 called "Long-Term Use of Antidepressants for Depressive Disorders and the Risk of Diabetes Mellitus." It's an eye-opening study, drawing on data from the U.K.'s patient database. The numbers are solid and the results hard to argue with. Even after controlling for body mass index (BMI), hypertension, hyperlipidemia, smoking, age, and other factors, the authors of the study found that long-term use of antidepressants (of any kind: tricyclic, MAOI, SSRI) was associated with an almost two-fold greater risk of diabetes.

This is a shocking result, because it indicates that antidepressants add significantly to the burden of disease. In the U.S., where 27 million people take antidepressants (60% of them for two years or longer), it could mean an extra million cases of diabetes.

From the 1988-1994 time period to the 2005-2008 period, antidepressant usage in the U.S. rose 400%, according to the Centers for Disease Control. This corresponds with an almost-quadrupling of diabetes cases in the same time frame.

Before you start thinking that maybe depression in and of itself is predisposing to weight gain and diabetes (which it is), go read the Andersohn paper. The authors already thought of such things and controlled for them in their study control populations. They found that even after controlling for the usual risk factors, recent long-term (24 months or more) antidepressant usage increased the risk of diabetes by 84%. (Consult the paper for a list of the 29 antidepressants included in the analysis and the individual risk ratios for each.)

The Andersohn study was motivated by an earlier finding that continuous antidepressant use over an average study duration of 3.2 years was associated with an 2.6-fold increased risk of diabetes (95% CI=1.37–4.94) in the placebo arm and 3.39-fold increase in risk (95% CI=1.61–7.13) in the lifestyle intervention arm of the study reported in Diabetes Care. 2008 Mar;31(3):420-6. The Andersohn study confirms the previous finding.

Independent confirmation of the foregoing results can be found in a 2010 cross-sectional study of patients in Finland. Mika Kivimäki et al., writing in Diabetes Care, December 2010 33:12, 2611-261, reported finding a two-fold increased risk of Type 2 diabetes in patients who had taken 200 or more "defined daily doses" (about six months' worth) of antidepressant medication. Stratification by antidepressant type found no significant difference for tricyclics versus SSRIs. Interestingly, diabetes risk was higher for patients who had taken 400 or more daily doses versus those who'd taken 200 to 400 daily doses, indicating a kind of dose-response relationship. The longer you're on meds, the more likely you'll get diabetes.

The graph depicted further above (from CDC's Diabetes Report Card 2012) shows pretty clearly that if there's one thing America doesn't need right now, it's more cases of diabetes. Diabetes is already out of control in the U.S. In many counties (all of those shown in dark red below), diabetes already afflicts more than 11% of the population.

We already know that high body mass index, out-of-band blood lipids, inactivity, and age are important risk factors for diabetes. But we now know a major new risk factor: antidepressants. As Richard R. Rubin writes in US Endocrinology, 2008;4(2):24-7:

Applying current estimates of the number of people in the US who have prediabetes (57 million with impaired glucose tolerance or impaired fasting glucose), and estimates of the prevalence of antidepressant use among adults in the US (at least 10%), it would seem that almost six million people in the US have pre-diabetes and are taking antidepressants. This is a fairly large number of people, and if future research confirms that antidepressants are an independent risk factor for type 2 diabetes, efforts to minimize the potentially negative effects of these agents on glycemic control should be pursued.