Sultan Lab for Mental Health Informatics

Pediatric Antipsychotic Overuse: A National Public Health Concern

Antipsychotic medications were developed for serious mental illnesses including schizophrenia and bipolar disorder. Over the past three decades, however, their use in American children has expanded dramatically — far beyond the conditions for which the U.S. Food and Drug Administration has approved them. The result is a national prescribing pattern in which antipsychotic medications are routinely given to children for symptoms (aggression, irritability, disruptive behavior) rather than for the diagnoses for which these drugs have demonstrated efficacy.

The Sultan Lab for Mental Health Informatics at Columbia University has conducted some of the most influential national analyses of this prescribing pattern. This page summarizes the key findings, with a focus on the 2019 JAMA Network Open study that documented how children newly diagnosed with ADHD — a condition for which antipsychotics have no FDA-approved indication — are nonetheless frequently prescribed these medications.

Sultan 2019 JAMA Network Open: The Headline Findings

In a landmark study published in JAMA Network Open in July 2019, Dr. Sultan and colleagues analyzed national prescription claims from the Truven Health MarketScan database covering commercially insured children and adolescents from 2010 to 2015. The study followed 187,563 youths aged 3-24 from the moment they received a new ADHD diagnosis, excluding any child who already had a diagnosis for which antipsychotics have an FDA-approved indication (schizophrenia, bipolar disorder, autism spectrum disorder, or Tourette disorder). The study then asked a simple question: how many of these children received an antipsychotic prescription within one year, and what clinical reasoning supported that prescription?

One in Forty Children with ADHD Received an Antipsychotic

The headline finding: 2.6% of newly-diagnosed ADHD youths — approximately one in forty — were prescribed an antipsychotic medication within one year of their ADHD diagnosis. This rate is more than four times higher than the general-population annual antipsychotic prescribing rate in young people of approximately 0.6%. Critically, none of these children had a diagnosis that would have justified antipsychotic treatment under FDA labeling at the time of their ADHD diagnosis.

Preschoolers Had the Highest Rate

The age stratification of antipsychotic initiation reveals a striking pattern that runs counter to widely-held clinical assumptions:

Preschool-aged children — for whom clinical practice guidelines recommend behavioral therapy and parent-training as first-line treatment, and for whom long-term neurodevelopmental effects of antipsychotic exposure remain poorly understood — were the age group most likely to receive these medications following an ADHD diagnosis. This finding raises particular concern given the metabolic, cardiovascular, and developmental risks associated with antipsychotic medications in young children.

Nearly Half Were Never Tried on Stimulants First

Stimulant medications (methylphenidate and amphetamine derivatives) are the evidence-based, FDA-approved first-line treatment for ADHD. Decades of clinical trial evidence support their efficacy and safety. Stimulant medications should typically be tried at adequate dose and duration before considering antipsychotic augmentation in ADHD with aggression or other treatment-refractory symptoms.

Yet the 2019 study found that, of the children with ADHD who initiated antipsychotic medications:

• 47.9% had not received any stimulant prescription before the antipsychotic was started
• 43.8% had received only one stimulant class (methylphenidate or amphetamine, but not both) before the antipsychotic
• Only 8.4% had received both methylphenidate and amphetamine classes — the threshold considered minimally adequate before considering antipsychotic augmentation — before being placed on an antipsychotic

In other words, in nearly half of cases, an American child with newly-diagnosed ADHD was placed on an antipsychotic medication without ever being offered the evidence-based first-line treatment for their condition. This pattern is not consistent with any major American or international clinical practice guideline.

Even Generous Definitions of "Justified" Left Half Without Rationale

The study used a generous definition of what would constitute a clinically defensible reason for antipsychotic prescribing. Beyond the FDA-approved indications (schizophrenia, bipolar disorder, autism spectrum disorder, Tourette disorder), the analysis also counted oppositional defiant disorder and conduct disorder as "evidence-supported" indications — because risperidone in particular has clinical trial evidence for these conditions in stimulant-resistant youth.

Even with this generous definition, the study found:

• 52.7% of antipsychotic-treated ADHD youth received any FDA-indicated OR evidence-supported diagnosis during the follow-up year
• 35.1% received an FDA-approved indication for antipsychotic treatment (most acquired AFTER the antipsychotic was started)
• 26.9% received ODD or conduct disorder diagnoses (the "weak evidence" category)
• 47.3% — nearly half — had no defensible clinical reason to receive an antipsychotic, even using the most generous criteria available

Which Drugs Were Being Prescribed

The three most commonly initiated antipsychotic medications accounted for over 90% of first prescriptions:

Of these, only risperidone has reasonable clinical trial evidence for the off-label use most of these children represented (aggression in ADHD with disruptive behavior). Quetiapine — which accounted for one in five first prescriptions — has the weakest evidence base for pediatric off-label use and one of the most unfavorable metabolic profiles. Its prominent role in pediatric prescribing is difficult to justify on clinical or safety grounds.

Who Receives These Prescriptions: Risk Factors

The 2019 analysis identified clinical and demographic characteristics that substantially increased the likelihood of antipsychotic initiation in children with ADHD. These were not random children — they were the most clinically complex, the most distressed, and the most poorly served by the existing pediatric mental health system.

These data tell a system-failure story rather than a prescriber-failure story. The clinicians prescribing antipsychotics for children with ADHD were often working with the most complex patients in the system — children who had been hospitalized, who had self-harmed, who had multiple co-occurring diagnoses, who had failed other medications. The antipsychotic was frequently a last-resort intervention for clinicians working within fifteen-minute appointments, without access to specialty psychiatric consultation or evidence-based behavioral therapy.

The Vulnerable Populations: Foster Care

While the 2019 Sultan study focused on commercially insured children, an even more concerning pattern emerges in the publicly insured population — particularly children in foster care. Children in foster care represent approximately 3% of Medicaid-insured children but account for approximately 15% of all Medicaid-insured children receiving antipsychotics (Crystal et al., 2016, Health Affairs).

Among foster children, antipsychotic prevalence by age band reaches levels not seen in any other pediatric subpopulation:

By 2010, nearly 14% of foster adolescents were on antipsychotic medications — a rate comparable to adult mental-health-system prevalence. The state-as-parent relationship in foster care has produced the most aggressively medicated child subpopulation in U.S. healthcare. State Medicaid prior authorization programs and foster care monitoring initiatives (in 44+ states by 2013) have produced a modest plateau in this prescribing since 2008, but the underlying prevalence remains alarmingly high.

Why This Pattern Persists: System Drivers

Understanding pediatric antipsychotic overuse requires looking beyond individual prescribing decisions to structural drivers in American pediatric mental health care.

Workforce Shortage

There are approximately 7,000 practicing child and adolescent psychiatrists in the United States serving roughly 70 million children. This shortage means that primary care physicians — pediatricians, family physicians — manage the majority of pediatric mental health concerns. Most of these clinicians work in 15-minute visit slots without access to immediate specialty consultation.

Behavioral Therapy Access

Evidence-based psychosocial interventions for children with disruptive behavior, oppositional defiant disorder, and ADHD-related aggression — including parent-child interaction therapy, parent management training, and multisystemic therapy — are not widely available in most American communities. Reimbursement structures further disincentivize these interventions relative to medication management. When medication is the only available tool, medication becomes the default treatment.

Symptom-Driven Prescribing

Antipsychotic prescribing in children is increasingly driven by symptoms (aggression, impulsivity, mood lability, sleep disturbance) rather than by diagnoses. The drugs are prescribed for behavioral phenomena across diagnostic boundaries, regardless of whether the diagnostic conditions match the FDA-approved indications. This pattern reflects clinical desperation more than diagnostic clarity.

Pharmaceutical Marketing

Pharmaceutical marketing of off-label pediatric uses, especially for second-generation antipsychotics in the 2000s, contributed significantly to the expansion of pediatric antipsychotic prescribing. Several manufacturers paid billions of dollars in legal settlements for off-label pediatric promotion. While marketing has been more constrained in recent years, the prescribing patterns it helped establish have persisted.

Safety Concerns: Why This Matters

Antipsychotic medications, particularly second-generation antipsychotics, carry well-documented safety risks that are especially relevant in pediatric populations:

• Weight gain and metabolic syndrome: Second-generation antipsychotics produce more pronounced weight gain in children than in adults. Olanzapine, quetiapine, and risperidone are among the worst offenders.
• Type 2 diabetes risk: A 2016 meta-analysis (Galling et al., JAMA Psychiatry) found that pediatric antipsychotic exposure is associated with elevated risk of incident Type 2 diabetes.
• Cardiovascular concerns: Recent evidence (Ray et al., 2019, JAMA Psychiatry) suggests that higher-dose antipsychotic treatment in children and youth is associated with increased risk of unexpected death.
• Neurodevelopmental uncertainty: Animal studies and limited human data raise concerns about the long-term effects of antipsychotic exposure on the developing brain, particularly in preschool-aged children.
• Sedation and cognitive effects: Antipsychotic-induced sedation and cognitive blunting can interfere with school performance, social development, and the very symptoms (impulsivity, attention) being targeted.

The risk-benefit calculus for antipsychotic treatment in pediatric populations is therefore highly indication-dependent. For children with severe psychotic disorders, autism-associated severe irritability, or treatment-resistant aggression after adequate stimulant trials, these medications may be appropriate and even life-changing. For children with newly-diagnosed ADHD who have never been offered first-line treatment — the population identified in Sultan 2019 — the risk-benefit calculus is much less favorable.

Policy Implications

The findings from Sultan 2019 and related national studies have several policy implications:

Commercial Insurance Gap

The Sultan 2019 study examined commercially insured children — not Medicaid. While most state Medicaid programs have implemented prior authorization requirements and foster care monitoring programs for pediatric antipsychotic prescribing, commercial insurance plans have largely not followed suit. This creates a regulatory gap in which the commercially insured children who appear in the Sultan study were prescribed antipsychotics without facing the same oversight that publicly insured children would have encountered. Expanding prior authorization and quality measurement to commercial insurance is a logical next policy step.

Stimulant Trial Requirements

Given that nearly half of antipsychotic-treated ADHD youth in the Sultan study had never received a stimulant trial, a clinically reasonable policy intervention would require documentation of adequate stimulant trial (both methylphenidate and amphetamine classes) before antipsychotic initiation for children with ADHD diagnoses. This would not prohibit antipsychotic use in genuinely treatment-resistant cases but would prevent antipsychotic prescribing as a first-line response to ADHD-associated behavioral concerns.

Workforce Investment

The underlying driver of off-label pediatric prescribing is the inadequate supply of child and adolescent psychiatrists and evidence-based behavioral therapy. No prescribing-side regulation will fully address this problem without simultaneous investment in pediatric mental health workforce expansion, telepsychiatry consultation models (e.g., Massachusetts Child Psychiatry Access Project), and reimbursement reform for behavioral interventions.

HEDIS Quality Measures

The National Committee for Quality Assurance Healthcare Effectiveness Data and Information Set (HEDIS) measures for pediatric antipsychotic prescribing now include metrics on metabolic monitoring, psychosocial care, and polypharmacy avoidance. Expanding these measures to address indication appropriateness and stimulant trial documentation would create stronger quality-improvement levers.

Key Publications from the Sultan Lab

The Sultan Lab has published multiple studies on pediatric psychotropic prescribing patterns. The publications most relevant to the antipsychotic overuse story include:

• Sultan RS, Wang S, Crystal S, Olfson M. "Antipsychotic Treatment Among Youths With Attention-Deficit/Hyperactivity Disorder." JAMA Network Open. 2019;2(7):e197850. doi:10.1001/jamanetworkopen.2019.7850 — The foundational study described on this page.
• Sultan RS, Correll CU, Schoenbaum M, King M, Walkup JT, Olfson M. "National Patterns of Commonly Prescribed Psychotropic Medications to Young People." J Child Adolesc Psychopharmacol. 2018;28(3):158-165. doi:10.1089/cap.2017.0077 — National landscape of pediatric stimulants, antidepressants, and antipsychotics.
• Sultan RS et al. 2025 JAMA Psychiatry — ADHD medication protective effects on substance use outcomes.

Dr. Sultan's research collaborators on these studies include Dr. Mark Olfson (Columbia University, the field's leading authority on national psychotropic prescribing trends), Dr. Christoph Correll (Zucker Hillside Hospital, leading researcher on pediatric antipsychotic safety), Dr. Stephen Crystal (Rutgers University, foster care antipsychotic monitoring), and Dr. John Walkup (Northwestern University, pediatric anxiety treatment).

Frequently Asked Questions

What percentage of children with ADHD are prescribed antipsychotic medications?

According to Sultan 2019 JAMA Network Open, approximately 2.6% — or about 1 in 40 — of commercially insured children and adolescents with a new ADHD diagnosis are prescribed an antipsychotic medication within one year. This rate is more than four times higher than the general-population pediatric antipsychotic prescribing rate (~0.6%) and includes children who have no FDA-approved indication for antipsychotic treatment. The rate is highest for preschool-aged children (3-5 years) at 4.3%.

Why are antipsychotics prescribed to children with ADHD if ADHD is not an FDA indication?

Antipsychotics are most commonly prescribed in children with ADHD to manage co-occurring aggressive, impulsive, or oppositional behaviors. Some clinical trial evidence supports risperidone for severe aggression in children with ADHD plus conduct disorder or oppositional defiant disorder, particularly when stimulants have failed. However, Sultan 2019 found that nearly half of children prescribed antipsychotics for ADHD-related concerns had never received a stimulant trial first, contradicting evidence-based prescribing recommendations.

What are the risks of antipsychotic medication in children?

Antipsychotic medications in pediatric populations carry well-documented risks including weight gain, metabolic syndrome, elevated risk of Type 2 diabetes, sedation, and potential effects on the developing brain that are not yet fully understood. Recent evidence also suggests that higher-dose antipsychotic exposure may be associated with increased risk of unexpected death in children and youth. These risks make off-label prescribing without strong clinical rationale particularly concerning, especially in preschool-aged children.

Why are children in foster care more likely to receive antipsychotics?

Children in foster care experience higher rates of trauma, behavioral challenges, and mental health comorbidities than other pediatric populations. They also have less consistent access to evidence-based behavioral therapy and stable adult caregivers who can implement non-pharmacological interventions. The result is that foster children are prescribed antipsychotics at rates 5-10 times higher than other Medicaid-insured children. Foster care has become an "incubator" for state oversight policies including prior authorization requirements and judicial review, though these protections do not extend to commercially insured children.

How does the Sultan 2019 study connect to other pediatric prescribing research?

The Sultan 2019 study builds on a body of research from Columbia, Rutgers, and other institutions documenting the dramatic rise in pediatric antipsychotic prescribing since the late 1990s (Olfson 2012, 2015; Crystal 2009, 2016; Hales 2018). It also connects to Dr. Sultan's earlier 2018 paper on the overall landscape of pediatric psychotropic prescribing and to his more recent work examining the relationship between ADHD medication treatment and real-world outcomes including substance use. Together, these studies form a comprehensive picture of how psychotropic medications are prescribed to American children and what the consequences of those patterns are.

Related Resources

• Pediatric Psychotropic Prescribing Landscape - National patterns across stimulants, antidepressants, and antipsychotics (Sultan 2018)
• ADHD Antipsychotic Research - Sultan Lab antipsychotic and ADHD research overview
• Off-Label Antipsychotic Prescribing - Clinical complexity of off-label use
• Sultan Lab for Mental Health Informatics - Full lab overview
• ADHD Medications Guide - Clinical medication information
• ADHD Stimulant Prescribing Patterns - IQVIA national analysis
• Complete Publication List - All Sultan Lab peer-reviewed research
• Media Coverage - Press coverage of Sultan Lab research