Abstract

I read with interest the study of Zhu et al. (2025), which showed an increased risk of type 2 diabetes (T2M) for Medicaid youth taking a stimulant and two or more psychotropic class. I second their call for clinicians to pay attention to this issue and do appropriate clinical and laboratory monitoring for endocrine and cardiovascular risks (Hilt, 2021). At the same time, I did note some positive findings in the data from their study. In 30,000 children with attention-deficit/hyperactivity disorder (ADHD), 75.5% were treated with stimulants alone, and additional 16.4% were treated with a stimulant and one additional class of psychotropic. Only small numbers were treated with three or more medications, which suggests clinicians are heeding guidelines to manage ADHD first before moving to polypharmacy (Blader et al., 2021). Also, of the 7737 treated with polypharmacy, the rate of diabetes was 28 patients or 0.38 compared with 15/22,732 (0.066%) in the control group. The relative risk is high, although the absolute risk increase (0.38–0.066) is 0.31%. Thus, the Number Needed to Harm is 1/333. I am not in any way dismissing the risk, merely pointing out that it is manageable. When faced with a child with ADHD and serious comorbid disorders, the ADHD should be managed pharmacologically first (Pliszka et al., 2006). If appropriate psychosocial interventions do not address the comorbidity (or the condition is too severe to wait), then treatment with other psychotropics is indicated, with all the monitoring that Zhu et al. suggest. A final point I would make is that there is an increased risk of T2D in ADHD and other psychiatric disorders, which could be independent of pharmacotherapy (Landau and Pinhas-Hamiel, 2019). If T2D risk is a function of severity, then it might be confounded with the number of medications prescribed. Perhaps there needs to be greater cardiometabolic vigilance in children with severe psychiatric disorders regardless of medication status.
Footnotes
Disclosures
No competing financial interests exist.
