Depression, Antidepressants, and Weight Gain in Children.

Abstract

Very few areas in medicine have been as fraught with controversy as the treatment of children with psychotropic medicines. There were significant increases in the prescribing of psychotropic medications in general and antidepressants in particular to children and adolescents since the early 1990s (1). This has led to a common perception that mental disorders are overdiagnosed and that treatment is often unnecessary and even harmful, a view that is amplified in the case of children (2). In fact there is strong evidence that antipsychotics are associated with weight gain and metabolic disease in children (3). However, is the same true for antidepressants? Analyzing these associations is complicated by the observation that depression itself is associated with weight gain, at least in adults (4). Schwartz et al. (5) in this issue of Obesity have addressed the relationships between depression, antidepressant treatment, and increases in BMI using data from a large sample of children extracted from an electronic health record database. The approach modeled the effects of several interactive factors on trajectories of BMI change over time: depression diagnosis, nonmental health comorbidities (e.g., asthma or diabetes), anxiety comorbidity, and antidepressant prescriptions. The study used a mixed-effects linear regression approach that tested for both independent and interactive effects of the predictor variables. Both prior depression diagnosis and antidepressant treatment were significantly associated with increases in BMI, and each showed effects that were independent of the other. From these data the authors concluded that depression itself and treatment with antidepressants were independently associated with increases in BMI. One year of antidepressant exposure was associated with 2.1 kg (4.6 lbs) of weight gain (2.5 kg for SSRIs) at age 18. While this change in weight is by no means trivial and it may be associated with an increased risk of metabolic disease, it is not as high as antipsychotic drugs. A recent article reported that 1 year of exposure with antipsychotics in children and adolescents was associated with a mean weight increase of 11.6 kg (6). Further, attention to the possibility of weight gain and the use of approaches to mitigate risk could prevent this problem. There is a further risk of confounding by indication as noted by the authors. Depressed children treated with antidepressants may not be identical to those not treated. The authors made a good faith effort to control for this by including some children who were treated with antidepressants but were not diagnosed with depression and by adjusting for covariates. However, the possibility of residual confounding remains: that is, the effects of differences in the disease state may persist even after adjusting for covariates. This is suggested by the observation that both comorbid anxiety and longer durations of antidepressant exposure, which might be surrogates for more severe or chronic illness, increased the drug effect. However, overall the results are very similar to findings in adults and suggest that depressed children with or without antidepressant therapy should be carefully monitored for weight gain.