From the Journals

Reassuring findings on SSRIs and diabetes risk in children


SSRIs are associated with a much lower risk of type 2 diabetes (T2D) in children and adolescents than previously reported, new research shows.

Investigators found publicly insured patients treated with SSRIs had a 13% increased risk for T2D, compared with those not treated with these agents. In addition, those taking SSRIs continuously (defined as receiving one or more prescriptions every 3 months) had a 33% increased risk of T2D.

On the other hand, privately insured youth had a much lower increased risk – a finding that may be attributable to a lower prevalence of risk factors for T2D in this group.

“We cannot exclude that children and adolescents treated with SSRIs may be at a small increased risk of developing T2D, particularly publicly insured patients, but the magnitude of association was weaker than previous thought and much smaller than other known risk factors for T2DM, such as obesity, race, and poverty,” lead investigator Jenny Sun, PhD, said in an interview.

“When weighing the known benefits and risks of SSRI treatment in children and adolescents, our findings provide reassurance that the risk of T2DM is not as substantial as initially reported,” said Dr. Sun, a postdoctoral research fellow in the department of population medicine at Harvard Medical School’s Harvard Pilgrim Health Care Institute, Boston.

The study was published online Sept. 2 in JAMA Psychiatry.

Limited evidence

Previous research suggested that SSRIs increase the risk of T2D by up to 90% in children and adolescents.

However, the investigators noted, the study reporting this finding was too small to draw conclusions about the SSRI class as a whole also did not examine specific SSRIs.

In addition, although “several studies have reported that antidepressant use may be a risk factor for T2D in adults, evidence was limited in children and adolescents,” said Dr. Sun.

“Rapid changes in growth during childhood and adolescents can alter drugs’ pharmacokinetics and pharmacodynamics, so high-quality, age-specific data are needed to inform prescribing decisions,” she said.

For the current study, the researchers analyzed claims data on almost 1.6 million patients aged 10-19 years (58.3% female; mean age, 15.1 years) from two large claims databases.

The analysis focused on those with a diagnosis warranting treatment with an SSRI, including depression, generalized or social anxiety disorder, obsessive compulsive disorder, PTSD, panic disorder, or bulimia nervosa.

The Medicaid Analytic Extract database consisted of 316,178 patients insured through Medicaid or the Children’s Health Insurance Program. The IBM MarketScan database consisted of 211,460 privately insured patients. Patients were followed up for a mean of 2.3 and 2.2 years, respectively.

Patients who initiated SSRI treatment were compared with those with a similar indication but who were not taking an SSRI. Secondary analyses compared new SSRI users with patients who recently initiated treatment with bupropion, which has no metabolic side effects, or with patients who recently initiated psychotherapy.

“In observational data, it is difficult to mimic a placebo group, often used in RCTs [randomized, controlled trials], therefore several comparator groups were explored to broaden our understanding,” said Dr. Sun.

In addition, the researchers compared the individual SSRI medications, using fluoxetine as a comparator.

A wide range of more than 100 potential confounders or “proxies of confounders,” were taken into account, including demographic characteristics, psychiatric diagnoses, metabolic conditions, concomitant medications, and use of health care services.

The researchers conducted two analyses. They included an intention-to-treat (ITT) analysis that was restricted to patients with one or more additional SSRI prescriptions during the 6 months following the index exposure assessment period.


Next Article: