Characteristics of Youths Treated With Psychotropic Polypharmacy in the United States, 1999 to 2015

The use of medication from 2 or more psychotropic classes, ie, polypharmacy,¹ has increased among US youths² despite limited evidence of efficacy and mounting safety concerns.³ The US Food and Drug Administration (FDA) approved several psychotropic classes, eg, selective serotonin reuptake inhibitors and second-generation antipsychotics, for expanded use in pediatric populations,^4,5 but it is unknown whether this corresponds with the observed increase in psychotropic polypharmacy and who is more likely to receive psychotropic polypharmacy. Our goal was to investigate differences during a 17-year study period in the characteristics of US youths treated with psychotropic polypharmacy and in the psychotropic classes used in combination.

This is a cross-sectional study using the Medical Expenditure Panel Survey (MEPS) data from 1999 to 2015. The MEPS is a national household survey that generates sampling weights to obtain nationally representative estimates on health care use and expenditures in the United States. Sampling design and weighting details are described elsewhere.⁶ The study included youths younger than 18 years who reported psychotropic polypharmacy. Each study year consists of 3 interview rounds. To define psychotropic polypharmacy in a study year, 3 or more psychotropic classes must be reported together in at least 1 of the 3 interview rounds in the year. Our definition identified psychotropic classes taken together and avoided the capture of clinically recommended 2-class regimens (ie, use of α-agonists with stimulants). We used the American Hospital Formulary System to categorize medications by psychotropic class: stimulants, antidepressants, mood stabilizers, antipsychotics, anxiolytics/hypnotics, sedatives, and α-agonists. Atomoxetine was included in the stimulant class because this is a second-line treatment for attention-deficit/hyperactivity disorder (ADHD). The study received an exemption as non–human participants research from the University of Maryland institutional review board. We used publicly available data; therefore, no patient consent was needed in our study.

Individual study years were pooled into 3 periods (1999-2004, 2005-2010, and 2011-2015). We examined youth characteristics (age, sex, race/ethnicity, individual psychotropic class use, and psychiatric disorders) and the psychotropic polypharmacy combinations in each period. Weighted frequency analysis obtained nationally representative estimates and 95% confidence intervals in each period. Separate crude weighted logistic regression models were used to compare youth characteristics between adjacent periods. All raw sample sizes met the MEPS reporting standard. A domain analysis was conducted for subgroup analyses.

Table 1 shows the number of US youths treated with psychotropic polypharmacy increased from 101 836 (1999-2004) to 222 955 (2005-2010) to 293 492 (2011-2015). Youths who were male and White composed the largest proportion in all periods. The proportion who were age 13 to 18 years increased significantly from 2005 through 2010 to 2011 through 2015 (17.2%; 95% CI, 16.9%-17.5%). In 1999 to 2004, the most common psychotropic classes were stimulants/atomoxetine (77.8%), antidepressants (71.3%), and mood stabilizers (61.2%). In the subsequent periods, stimulants/atomoxetine remained the predominant class. Antipsychotics significantly increased 30.7% (95% CI, 30.0%-31.4%) from 1999 through 2004 to 2005 through 2010, while the modest increase from 2005 through 2010 to 2011 through 2015 was not significant. We observed a significant decrease of 20.6% (95% CI, 20.2%-21.0%) in mood stabilizers from 1999 through 2004 to 2005 through 2010 followed by a nonsignificant decrease from 2005 through 2010 to 2011 through 2015. More than 80% of US youths with reported psychotropic polypharmacy had an ADHD diagnosis. There was a 10.1% (95% CI, 9.9%-10.3%) increase from 2005 through 2010 to 2011 through 2015 in the proportion who had 3 or more documented diagnoses.

Table 2 shows the psychotropic class combinations reported in 10% or more of youths who received psychotropic polypharmacy. Stimulants were present in nearly all regimens, and antipsychotics became prominent in combinations after 2004.

Attention-deficit/hyperactivity disorder is the predominant diagnosis among youths who received psychotropic polypharmacy. The findings suggest an increase in concurrent use of antipsychotics with other psychotropics. Use of mood stabilizers decreased possibly as more youths received antipsychotics for mood disorders. Evidence of the efficacy and safety is needed to guide psychotropic polypharmacy practice. A study limitation is that the definition of psychotropic polypharmacy may have obscured medication switching.

Corresponding Author: Chengchen Zhang, MPH, Department of Pharmaceutical Health Services Research, University of Maryland School of Pharmacy, 220 Arch St, 12th Floor, Baltimore, MD 21201 (chengchen_zhang@umaryland.edu).

Accepted for Publication: April 9, 2020.

Published Online: November 2, 2020. doi:10.1001/jamapediatrics.2020.4678

Author Contributions: Ms Zhang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: Zhang, Spence, dosReis.

Drafting of the manuscript: Zhang, Spence.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Zhang, dosReis.

Administrative, technical, or material support: Zhang, Spence, dosReis.

Supervision: Reeves, dosReis.

Conflict of Interest Disclosures: Dr Spence reported other support from Pharmaceutical Research Manufacturers of America (PhRMA) and the US Food and Drug Administration (FDA) outside the submitted work. Dr Reeves reported grants from the National Institute of Mental Health (NIMH), FDA, PCORI, and SAMHSA during the conduct of the study. Dr dosReis reported other support from GSK and PhRNA and grants from NIMH, the Patient Centered Outcomes Research Institute (PCORI), and the PhRMA Foundation outside the submitted work. Ms Zhang is a Maryland Center of Excellence in Regulatory Science and Innovation (CERSI) Scholar with the FDA. Dr Spence is a recipient of a Maryland CERSI Scholar award from the FDA and is currently a PhRMA Foundation predoctoral fellow in health outcomes. No other disclosures were reported.

Funding/Support: Part of the analysis of the research in this article was conducted at the CFACT Data Center, and the support of Agency for Healthcare Research and Quality is acknowledged.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The results and conclusions in this paper are those of the authors and do not indicate concurrence by AHRQ or the Department of Health and Human Services.

Meeting Presentation: This paper was presented at 35th International Conference of Pharmacoepidemiology; August 26, 2019; Philadelphia, Pennsylvania.

Additional Contributions: We acknowledge Ray Kuntz from the Agency for Healthcare Research and Quality for coordinating our data analysis at AHRQ CFACT Data Center. Mr Kuntz did not receive compensation for his role in the study.