Review of: Burco M et al, JAMA Pediatrics 2017;171(12):1200–1207

Over the last decade, several published studies have reported an increased risk of type 2 diabetes associated with antidepressant use in adults. But does the same hold true for children and adolescents?

This paper is the first population-based study to examine the risk of onset of type 2 diabetes with the use of antidepressants in younger patients.

Medicaid administrative claims data from California, Florida, Illinois, and New Jersey were analyzed in a cohort of 119,608 youths, ages 5–20, who initiated treatment with antidepressants from 2005 through 2009. Regression models were used to analyze the risk of onset of type 2 diabetes relative to antidepressant use, duration, and dosing.

Current use of antidepressants was associated with a 1.92 adjusted relative risk for type 2 diabetes (95% CI, 1.43–2.57). Current users of SSRIs or SNRIs had a 1.88 adjusted relative risk (95% CI 1.34–2.64), and current users of TCAs had a 2.15 adjusted relative risk (95% CI 1.06–4.36). There were no elevated risks in current users of antidepressants other than SSRIs, SNRIs, or TCAs.

For SSRIs or SNRIs, the risk of onset of type 2 diabetes increased with longer durations of exposure and with larger cumulative dosing. Compared to risk for their use for the first 90 days, there was a relative risk of 1.68 (CI 0.83–3.40) for 91–150 days of use, 2.56 (CI 1.29–5.08) for 151–210 days of use, and 2.66 (CI 1.45–4.88) for > 210 days of use.

Compared to risk after a cumulative antidepressant dose of 1 mg–1500 mg of fluoxetine hydrochloride equivalents, there is a relative risk of 1.22 (CI 0.59–2.52) for 1501 mg–3000 mg of dose equivalents, 2.17 (1.07–4.40) for 3001–4500 mg of dose equivalents, and 2.44 (1.35–4.43) for > 4500 mg of dose equivalents.

CCPR’s Take
The study suggests that long-term antidepressant use, particularly with SSRIs or SNRIs, is associated with increased risk of onset of type 2 diabetes mellitus in children and adolescents. This increased risk is particularly prominent with long-term use and higher daily doses. But the study is observational and must be interpreted with caution. Causality cannot be inferred; however, there is a correlation. Type 2 diabetes mellitus represents a rare but serious adverse outcome to discuss with patients and families, and we should vigilantly monitor for it.