Atypical Antipsychotic Use Among Medicaid-Insured Children and Adolescents: Duration,Safety,and Monitoring Implications

Abstract

Objective:

Over the last two decades, the increased use of atypical antipsychotic medications, often for unlabeled indications including attention-deficit/hyperactivity disorder (ADHD), has been profound. This study aims to characterize duration of atypical antipsychotic use by age group and Medicaid eligibility category, and among youth with noncomorbid ADHD.

Methods:

Administrative data on 266,590 youth 2–17 years of age, and continuously enrolled in a mid-Atlantic state Medicaid program in 2006, were assessed in terms of median days of atypical antipsychotic use using bivariate analyses and multivariable quantile regression. Also, in a subanalysis of youth diagnosed with ADHD without any reported psychiatric comorbidities (i.e., noncomorbid ADHD), age-specific adjusted odds and adjusted median days of atypical antipsychotic use by Medicaid eligibility category were assessed. Additionally, patterns of use of single atypical antipsychotic regimens and two concomitant atypical antipsychotic regimens were described.

Results:

Overall, the median annual duration of atypical antipsychotic use was 180 days (interquartile range: 69–298 days). Children (2–12-year-olds) had longer durations of use than did adolescents (13–17-year-olds) (median 192 vs. 179 days), respectively. In the absence of any comorbid psychiatric diagnosis, ADHD-diagnosed foster care youth had more than threefold greater adjusted odds of atypical antipsychotic use than did youth enrolled in income-eligible Medicaid categories. Nearly one third of such ADHD-diagnosed foster care youth received atypical antipsychotics regardless of age group, with annual duration of use >250 median days in 2–12-year-olds. In concomitant atypical antipsychotic regimens, risperidone, aripiprazole, and quetiapine were the most common.

Conclusions:

Exposure to atypical antipsychotics in Medicaid-insured youth, in particular for children in foster care and those diagnosed with ADHD, was substantial, warranting outcomes research for long-term effectiveness, safety, and oversight for appropriate cardiometabolic monitoring.

Introduction

Over the last two decades, the increased use of second-generation antipsychotic medications, that is, atypical antipsychotics, has been profound (Olfson et al. 2002; Patel et al. 2005; Cooper et al. 2006; Olfson et al. 2006; Matone et al. 2012; Zito et al. 2013). This increased use has, however, occurred at a greater rate in youth that in adult populations (Olfson et al. 2012), mostly for publicly insured youth and for off-label (not approved by the United States Food and Drug Administration [FDA]) behavioral conditions, such as attention-deficit/hyperactivity disorder (ADHD) and other disruptive behavior disorders (Cooper et al. 2004; Crystal et al. 2009; Constantine et al. 2011, 2012; Zito et al. 2013). Compared with privately insured youth, youth in state Medicaid systems had five to sixfold greater antipsychotic use (Crystal et al. 2009), and by the calendar years 2006–2007, Medicaid-insured youth diagnosed with externalizing behavioral disorders by far represented the largest group of youth receiving antipsychotic medications (Matone et al. 2012; Zito et al. 2013).

The greatly expanded use of these agents in youth for ADHD occurred in the absence of FDA evidence-based labeling. Currently, atypical antipsychotic medications are FDA approved for use only in youth for schizophrenia, bipolar disorder, and irritability associated with autism, based on evidence from short-term randomized clinical trials. However, children and adolescents treated in community settings can vary sizably from clinical drug trial populations, and may often have substantially different psychotropic drug treatment outcomes. Several community-based studies have raised safety concerns for the use of atypical antipsychotic agents in children regarding treatment-emergent adverse events related to weight gain (Calarge et al. 2009; Correll et al. 2009; Andrade et al. 2011). Such treatment-emergent weight gain was found to be substantially greater in youth than in adults, according to secondary analysis of published clinical trial data (Safer 2004). To date, no atypical antipsychotic agent has an FDA-labeled indication for use in behavioral disorders in children and adolescents, underlying the urgent need for careful scrutiny of the duration of atypical antipsychotic use and the reasons for their use, according to clinician-reported psychiatric diagnosis.

In 8 week, open-label, clinical studies of olanzapine, risperidone, and quetiapine among preschool children, significant treatment-emergent cardiac and metabolic abnormalities were identified (Biederman et al. 2005; Joshi et al. 2012). In another study comparing short-term and long-term antipsychotic-emergent safety outcomes in children, children treated with longer durations of antipsychotic agents had greater cardiometabolic adversities and dyskinetic movements (Laita et al. 2007), warranting the need to assess patterns of atypical antipsychotic use in relation to their duration by age group and psychiatric diagnosis. A recent study also pointed out a trend of concomitant use of multiple antipsychotic agents among Medicaid-insured youth in a mid-Atlantic state (dosReis et al. 2011). Of youth with antipsychotic dispensings,>10% used two or more antipsychotic agents concomitantly (dosReis et al. 2011), raising concerns for appropriateness, in view of mounting evidence for antipsychotic treatment emergent cardiometabolic adversities.

The main objectives of this study are 1) to characterize duration of atypical antipsychotic use by age group, Medicaid eligibility category, and among ADHD-diagnosed youth; and 2) to identify the most commonly used atypical antipsychotic agents singly or concomitantly.

Methods

Data source

This study organized and analyzed computerized data from Medicaid administrative claims for youth enrolled in fee-for-service and managed care programs for the calendar year 2006, in a mid-Atlantic state. The claims database included enrollment files at the person level, which were linked to service claims files of Medicaid enrollees, including outpatient and physician claims and prescription drug claims through an encrypted identification number assigned to each enrollee. The services claim files comprised enrollee information at the claim level including the International Classification of Diseases and Related Health Problems, Ninth Revision (ICD-9) diagnosis and procedure codes.

The final analytical files comprised de-identified data. This study was reviewed and approved by the Institutional Review Board (IRB) at the state's Department of Health and Mental Hygiene and was exempt from the IRB review at the University of Maryland, Baltimore.

Study design

The following selection criteria were used to define the study cohort: 1) continuous enrollment for 12 months in the state Medicaid program in 2006; 2) being 2–17 years of age as of January 1, 2006. Continuously enrolled youth represented 72.0% of the state Medicaid program enrollees.

Outcomes

Atypical antipsychotic use was assessed by linking enrollment files and prescription drug files based on the presence of at least one dispensing claim for an atypical antipsychotic medication in the study year. Only atypical antipsychotics available in 2006 in the United States were considered in the study (i.e., risperidone, olanzapine, quetiapine, aripiprazole, ziprasidone, and clozapine). Youth with conventional antipsychotic use accounted for <5% of youth with any antipsychotic use in 2006 and, therefore, were excluded from the analyses. Annual prevalence of atypical antipsychotic use was expressed as the percentage of all eligible youth in the study year.

Duration of antipsychotic medication class use was assessed by a previously described algorithm using dispensing date, days of supply, and national drug code (NDC) information available at the prescription drug claim files (dosReis et al. 2002; Sikka et al. 2005; Constantine et al. 2010; dosReis et al. 2011). Among youth with any atypical antipsychotic dispensing in 2006, cumulative days of supply for dispensed antipsychotic medications were calculated for each youth. Gaps of <15 days between sequential antipsychotic dispensings were assumed to be continuous episodes of antipsychotic use.

Study subjects with ≥31 overlapping days of use of two or more atypical antipsychotic medications were flagged as having concomitant antipsychotic use. In addition, each antipsychotic medication regimen involving two or more atypical antipsychotic entities was identified per subject.

Independent variables

Independent variables included age groups (children [2–12-year-olds]; adolescents [13–17-year-olds]), race/ethnicity groups (white, African American, Hispanic, other), Medicaid eligibility groups (state Children's Health Insurance Program [CHIP]; Temporary Assistance for Needy Families [TANF]; foster care; and Supplemental Security Income [SSI]), gender, and psychiatric diagnostic groups, as detailed subsequently.

Clinician-reported psychiatric diagnoses per enrollee were assessed using ICD-9-Clinical Modification (CM) codes from outpatient and physician files. To identify clinician-reported psychiatric diagnoses from the claims, at least two outpatient physician claims on separate days were required (West et al. 1994). The psychiatric diagnostic groups included schizophrenia and other psychoses, pervasive developmental disorders (PDD) and intellectual disability (ID), bipolar disorder, disruptive disorders, ADHD, depressive disorders, anxiety disorders, adjustment disorder, communication and learning disorders, and any other psychiatric diagnosis. For youth in more than one psychiatric diagnostic category, a hierarchical approach in the order of decreasing severity of diagnosed conditions was adopted from a previously published study (Olfson et al. 2010). The hierarchical approach followed the order indicated.

Statistical analyses

Statistical analyses were conducted at the person level using SAS version 9.2 (SAS Institute, Inc., Cary, NC). Bivariable and multivariable quantile regression models were employed to compare the unadjusted and adjusted median duration of antipsychotic use by demographic, clinical, and administrative characteristics of continuously enrolled Medicaid-insured youth who received atypical antipsychotic medications in 2006.

A subgroup analysis was conducted among youth with clinician-reported ADHD, the largest group of youth receiving antipsychotic medication. To better understand the extent of off-label atypical antipsychotic use in “noncomorbid” ADHD, we excluded youth with any clinician-reported comorbid psychiatric diagnosis. Youth with “noncomorbid” ADHD did not have any clinician-reported psychiatric comorbid condition during the entire study year. Youth with “noncomorbid” ADHD accounted for the majority (n=11,616; 67%) of ADHD-diagnosed youth. Hence, in the absence of any clinician-reported comorbid psychiatric disorder, we conducted multivariable logistic regression and quantile regression modeling to report adjusted odds and median days of atypical antipsychotic medication use by Medicaid eligibility categories in these “noncomorbid” ADHD-diagnosed children and adolescents.

Results

Characteristics of the study population

There were 266,590 continuously enrolled Medicaid-insured youth ages 2–17 years in a mid-Atlantic state in 2006, which largely comprised African American youth (56.0%), and children 6–12 years old (42.8%). Overall, youth who were Medicaid eligible by low family income (i.e., CHIP and TANF) represented the majority of the Medicaid population (88.7%), and ADHD and disruptive behavior disorders were the most common clinician-reported psychiatric diagnoses (data not shown). The annual prevalence of atypical antipsychotic use in 2006 was 2.5% in children (2–12-year-olds) and 5.2% in adolescents (13–17-year-olds).

Duration of atypical antipsychotic use

Table 1 reports unadjusted and adjusted median days of atypical antipsychotic use among continuously enrolled youth in 2006. Overall, the median duration was 180 days (interquartile range: 69–298 days) regardless of age group. Half of the children (2–12-year-olds) had atypical antipsychotic use for ≥192 days in 2006. When 2–5-year-olds were examined separately, atypical antipsychotic duration of use in foster care youth exceeded that of youth identified as disabled (SSI), 180 versus 150 median days, respectively (data not shown).

Table 1.

Median Days of Atypical Antipsychotic (AAP) Use in 2006 by Demographic and Clinical Characteristics of Continuously Enrolled Medicaid-Insured Youth, n =8696^a

Characteristics	AAP n	Unadjusted median days	Interquartile range	Adjusted difference^b in median days	95% confidence interval
Age groups (years)
2–12 years	4775	192.0	80.0–300	22.0	(13.8, 30.2)
13–17 years	3921	179.0	60.0–297.0	Reference
Eligibility
CHIP	1184	171.5	60.0–283.5	24.0	(11.6, 36.4)
TANF	2839	125.0	60.0–248.0	Reference
SSI	2449	210.0	90.0–309.0	75	(64.6, 85.4)
Foster Care	2224	240.0	109.0–325.0	118.0	(108.3, 127.7)
Gender
Female	2693	150.0	60.0–270.0	Reference
Male	6003	201.0	88.0–304.0	39.0	(30.4, 47.6)
Race/Ethnicity
White	4057	210.0	90.0–309.0	Reference
Black	4136	153.5	60.0–277.5	−67.0	(−75.4, −58.6)
Hispanic	227	150.0	60.0–270.0	−66.0	(−96.1, −35.9)
Other	276	181.0	88.0–305.0	−42.0	(−67.6, −16.4)
Region
Northern metropolitan	4763	185.0	73.0–300.0	Reference
Southern metropolitan	1500	178.0	60.0–292.0	5.0	(−5.1, 15.1)
Other	2433	181.0	73.0–296.0	−2.0	(−10.5, 6.5)
Psychiatric diagnosis^c
Schizophrenia/other psychoses & PDD/ID	368	246.0	117.5–330.0	Reference
Bipolar disorder	1254	219.0	95.0–310.0	−26.2	(−45.0, −7.4)
Disruptive disorders and ADHD	4307	180.0	68.0–287.0	−60.2	(−78.4, −42.0)
Other^d	2767	178.0	60.0–300.0	−59.2	(−78.4, −40.0)

There were 8696 atypical antipsychotic-treated youth among 266,590 continuously enrolled Medicaid-insured youth (2–17 years) in 2006. Hence, duration of antipsychotic use was presented only for antipsychotic-treated youth.

Difference relative to designated reference groups, adjusted for age groups, Medicaid eligibility groups, gender, race/ethnicity, region, and psychiatric diagnosis.

Hierarchical approach was applied for psychiatric diagnoses in the decreasing order as shown in the table.

Other includes adjustment disorder, communication and learning disorders, and other psychiatric diagnoses not listed in the table.

CHIP, Children's Health Insurance Program; TANF, Temporary Assistance for Needy Families; SSI, Supplemental Security Income; PDD/ID, Pervasive developmental disorder/Intellectual disability.

Youth who were male, white, and enrolled in “special-needs” eligibility groups (i.e., foster care and SSI) had significantly longer median durations of atypical antipsychotic use than their demographic counterparts. Youth in foster care (240 median days) had the longest median duration of atypical antipsychotic use compared with youth in income-eligibility groups (CHIP: 172 median days; TANF: 125 median days) and those with disability (SSI: 210 median days). After adjusting for sociodemographic and clinical characteristics, children had 22 days longer median duration of atypical antipsychotic use than did adolescents. Notably, a more pronounced longer duration of 118 days in foster care youth was observed compared with youth in income-eligible groups. Schizophrenia/other psychoses/PDD/ID and bipolar disorder led the duration data. However, when 2–5-year-olds were examined separately, bipolar disorder led the group, with 198 median days of atypical antipsychotic exposure (data not shown).

Substantial atypical antipsychotic use for ADHD across Medicaid eligibility groups

Table 2 presents study year data on a cohort of 11,616 youth with noncomorbid ADHD, by age group and Medicaid eligibility group in terms of stratified annual % prevalence, odds of atypical antipsychotic use, and median days of atypical antipsychotic use. In the absence of any comorbid psychiatric diagnosis, ADHD-diagnosed foster care youth had more than threefold greater adjusted odds of atypical antipsychotic use than youth enrolled in income-eligible Medicaid categories (i.e., CHIP and TANF), regardless of age-group. One third of ADHD-diagnosed foster care children (2–12-year-olds) were given atypical antipsychotics, and half of the children who received atypical antipsychotic medications for noncomorbid ADHD used these agents for ≥251 days in the study year. Across age groups, ADHD-diagnosed foster care youth had >100 days of additional exposure to atypical antipsychotic agents compared with ADHD-diagnosed youth in income-eligible groups, controlling for sociodemographic and clinical characteristics.

Table 2.

Age Group Specific Annual Atypical Antipsychotic (AAP) Prevalence, Odds and Median Days of AAP Use by Medicaid-Eligibility Categories Among Continuously Enrolled Youth with “Noncomorbid” Attention-Deficit/Hyperactivity Disorder (ADHD) ^a in 2006, n =11,616

	n	AAP n	AAP prevalence %	Adjusted OR^b	95% CI	Unadjusted median AAP days	Interquartile range	Adjusted difference in median AAP days^c	95% CI
Age^*Eligibility
2–12 years olds
Income eligible(TANF or CHIP)	6640	849	12.8	Reference		152.0	60.0–270.0	Reference
SSI	1235	367	29.7	3.1	(2.7, 3.6)	180.0	90.0–275.0	35.0	(12.5, 57.5)
Foster care	846	268	31.7	3.6	(3.1, 4.2)	251.0	123.0–320.5	105.2	(80.7, 129.7)
13–17 years olds
Income eligible(TANF or CHIP)	1770	241	13.6	Reference		155.0	60.0–280.0	Reference
SSI	633	215	34.0	3.6	(2.9, 4.5)	195.0	86.0–301.0	66.0	(33.3, 98.7)
Foster care	492	140	28.5	3.0	(2.4, 3.8)	222.5	96.5–326.5	101.8	(59.7, 143.8)

Represents only youth with ADHD in the absence of any clinician-reported psychiatric comorbid condition (i.e. children and adolescents with any comorbid psychiatric diagnosis were excluded).

Adjusted for age groups, Medicaid eligibility groups, interaction term of age groups and Medicaid-eligibility groups, race/ethnicity, gender, and region.

Multivariable quantile regression model was conducted among ADHD-diagnosed youth without any comorbid psychiatric diagnosis who received atypical antipsychotic drugs in 2006 (n=2080), adjusting for age groups, Medicaid eligibility groups, interaction term of age groups and Medicaid eligibility groups, race/ethnicity, gender, and region.

CI, confidence interval; OR, odds ratio; TANF, Temporary Assistance for Needy Families; CHIP, Children's Health Insurance Program; SSI, Supplemental Security Income.

Concomitant atypical antipsychotic use

Among youth with atypical antipsychotic dispensings, 7.4% used two or more atypical antipsychotic entities concomitantly, raising major safety and appropriateness concerns in the absence of efficacy or effectiveness data supporting the use of two or more antipsychotics concomitantly in vulnerable youth populations. Among youth who used a single atypical antipsychotic agent (n=8053), the leading atypical antipsychotic entities were risperidone (55.5%), aripiprazole (31.7%), and quetiapine (20.6%) (data not shown). Likewise, these three agents were typically used in concomitant atypical antipsychotic regimens, as illustrated in Figure 1. Ziprasidone was present in almost all the other combinations (17.9%), with clozapine combinations accounting for the remainder (1.0%).

FIG. 1.

Frequency of concomitant atypical antipsychotic regimens among continuously enrolled Medicaid-insured youth with concomitant antipsychotic use, n=643.

When analyzed separately in selected psychiatric diagnostic groups among youth with atypical antipsychotic dispensings (data not shown), youth diagnosed with schizophrenia/other psychoses (16.6%) and with bipolar disorder (13.5%) had greater proportions of concomitant antipsychotic users than ADHD-diagnosed youth (6.7%).

Discussion

In this cross-sectional study, we characterized duration of atypical antipsychotic use by age group and Medicaid eligibility categories, and within a subset of noncomorbid ADHD-diagnosed youth, and identified the most commonly used atypical antipsychotic agents singly or concomitantly. To our knowledge, this is the first study elucidating the greatly expanded use of atypical antipsychotic agents in foster care youth with noncomorbid ADHD (clinician-reported ADHD without any reported comorbid psychiatric diagnosis) compared with youth in family-income eligible Medicaid categories (TANF and CHIP).

In the present study, the major findings are as follows. 1) The median duration of atypical antipsychotic use was 180 days, with slightly longer use in children than adolescents, 192 versus 179 median days, respectively. 2) Atypical antipsychotic duration of use (median days) by Medicaid eligibility group in the descending rank order was foster care (240 days), youth with disabilities, that is, youth in SSI (210 days), and family-income eligible groups CHIP (172 days) and TANF (125 days). 3) In children (2–12-year-olds) with noncomorbid ADHD, that is, the subset of children with ADHD as the only clinician-reported psychiatric diagnosis, foster care children had the largest atypical antipsychotic use in the study year (31.7%). Compared with family-income Medicaid-eligible youth, foster care youth had a threefold greater atypical antipsychotic use across age groups. 4) Among atypical antipsychotic-treated children (2–12-year-olds) with noncomorbid ADHD diagnosis, the duration of atypical antipsychotic use was 3 months longer in foster care than in family-income eligibility groups and 1 month longer in the SSI group. Similar findings were also observed for 13–17-year-old foster care youth. 5) Concomitant atypical antipsychotic use occurred in >7% of atypical antipsychotic-treated youth, with risperidone, aripiprazole, or quetiapine typically used in two atypical antipsychotic combination regimens. The greater annual utilization of concomitant atypical regimens in youth diagnosed with schizophrenia and with bipolar disorder may imply cross-tapering of atypical antipsychotic medications. Nevertheless, our algorithm of ≥31 overlapping days of use of two or more atypical antipsychotic medications is consistent with previously published methods to identify concomitant use of atypical antipsychotic agents (dosReis et al. 2011).

Previous studies have also examined the duration of antipsychotic use in youth populations (Constantine et al. 2011; dosReis et al. 2011; Constantine et al. 2012). Constantine et al. examined a Florida Medicaid cohort of continuously enrolled young children (<6 years old) and reported 173.5 median days of use during the year between July 2003 and July 2004 (Constantine et al. 2011). In a study conducted among Medicaid-insured youth in another mid-Atlantic state in 2003, dosReis et al. observed similar durations of antipsychotic use (mean days) in three Medicaid-eligibility groups in the following rank order: Foster care (222.3 days), SSI (190.3 days), TANF (134.9 days) (dosReis et al. 2011). Harris et al. examined a large 3 year cohort of 6–17-year-old youth with Medicaid-managed behavioral health insurance in counties with mental health therapy available (Harris et al. 2013). Comparing those with and without concurrent mental health therapy, the duration of use of any antipsychotic ranged from an average of 320 days (with) to 184 days (without), respectively. Although their data are not generalizable to typical Medicaid settings, the longer durations for integrated care may imply greater medication adherence. Overall, our findings on duration of atypical antipsychotic use corroborate previous studies and further show the duration of atypical antipsychotic use by Medicaid eligibility groups in relation to age and eligibility, and among noncomorbid ADHD-diagnosed youth across the age groups.

We chose to examine noncomorbid ADHD because it constitutes a common condition for which antipsychotic use is deemed off label, that is, without FDA labeling for specific indications (Roberts et al. 2003). Physicians have the authority to use a marketed drug off label. However, antipsychotic use in youth for ADHD and other behavioral problems underscores the need for oversight and postmarketing research, to assure that benefits outweigh the risks (Varley and McClellan 2009). Atypical antipsychotic use for off-label behavioral conditions such as ADHD now accounts for the majority of antipsychotic-treated youth with Medicaid coverage, raising short- and long-term safety concerns in the absence of sufficient efficacy data to justify potentially severe treatment-emergent adverse events (McIntyre and Jerrell 2008; Calarge et al. 2009; Correll et al. 2009; Andrade et al. 2011). It is probable that these medications are being prescribed to treat highly aggressive and problematic behavior in ADHD, especially when patients fail to respond to evidence-based treatment options. Nevertheless, long-term outcomes assessment of these medications in community youth populations are urgently needed to substantiate their expanded use to manage severe behavioral problems. This is particularly relevant to patient-centered comparative effectiveness research as a national strategy to characterize the long-term effectiveness and safety profile of various treatment modalities.

In this study, we found that the duration of atypical antipsychotic use in Medicaid-insured youth, particularly among young children in foster care, was much longer than that in many clinical studies, in which treatment durations were mostly in the range of 3–12 weeks. In spite of the short treatment exposures in these studies, metabolic abnormalities such as significant weight gain, hypercholesterolemia, sharp increases in triglyceride levels, elevations in blood glucose, and insulin resistance were common across the atypical antipsychotic class (Laita et al. 2007; Calarge et al. 2009; Correll et al. 2009; Cohen et al. 2012; Alan and Kultur 2013). In a meta-analysis of 41 clinical trials of second generation antipsychotics in youth, Cohen et al. (2012) observed significant antipsychotic treatment-emergent adverse events across the studies. Compared with placebo, average antipsychotic treatment-emergent significant weight gain ranged from 1 to 4 kg in 3–12 weeks. Similarly, significant antipsychotic treatment-emergent increases in blood glucose levels ranged from 2.1 to 10.8 mg/dL; average cholesterol level increases ranged from 4.5 to 10.8 mg/dL. Comparing 1 month atypical antipsychotic-exposed youth with >12 month-exposed youth, Laita et al. demonstrated that treatment-emergent cardiometabolic events were common in both groups, but significantly greater in the longer exposure group (Laita et al. 2007).

Taken together, these studies strongly support requirements to monitor the use of atypical antipsychotics, particularly in very young children (Egger 2010), for off-label conditions (Varley and McClellan 2009), and in vulnerable populations, that is, youth in foster care (Naylor et al. 2007; Government Accountability Office [GAO] 2011; Administration for Children and Families 2012) and other Medicaid eligible youth (Government Accountability Office [GAO] 2012). Monitoring of atypical antipsychotic use for appropriateness and safety takes several forms. First, clinicians can adopt epidemiologic monitoring of individual youth following clinical guidelines endorsed by psychiatry and diabetes experts (American Diabetes Association, American Psychiatric Association, American Association of Clinical Endocrinologists, and North American Association for the Study of Obesity 2004). Despite widespread promulgation of these recommendations, there is little evidence of its adoption in clinical settings, although professional guidelines recommending voluntary monitoring are emerging (American Academy of Child and Adolescent Psychiatry 2009; AACAP Work Group on Quality Issues 2010). As child psychiatry experts have recommended, laboratory data on blood glucose levels and lipids before starting atypical antipsychotic treatment and periodically thereafter will help to identify treatment-emergent metabolic abnormalities early, to minimize risk. In clinical care, such practices would assure parents and physicians that discontinuation based on rapid, substantial weight gain or adverse changes in laboratory values will occur.

A second approach to assure appropriate atypical antipsychotic monitoring is population based, and is conducted from administrative claims data by quality assurance programs for Medicaid and privately insured youth. Several studies in other states suggest very low rates of laboratory monitoring during antipsychotic treatment in Medicaid-insured children and adolescents (Morrato et al. 2010a,b). Overall, laboratory monitoring to assure appropriate use of atypical antipsychotics, particularly for behavior and other off-label conditions, whether in clinical studies or in statewide Medicaid program evaluations, has been largely absent.

However, in the past few years, action by state Medicaid agencies regarding oversight of psychotropic drugs, especially antipsychotics, is impressive, spurred by federal legislative mandates for Medicaid-insured youth in general (P.L. 110-351 2008), and foster care youth (GAO 2011; P.L. 112-34 2011; Administration for Children and Families 2012) in particular. One of the most potentially innovative approaches is the use of peer review preauthorization of physicians' requests to initiate antipsychotic agents in children (State of Maryland DHMH 2013). The ages covered are gradually being expanded from very young to youth <18 years old. Peer review by academic-based experts in child psychiatry offers the opportunity to both reduce unnecessary use and increase atypical antipsychotic monitoring. The impact of these programs warrants study.

Limitations

Several limitations of the study must be acknowledged. First, these data sets are extracted from administrative claims for reimbursement in the Medicaid insurance system, and may be subject to unmeasured confounding bias, such as disease severity. Noncomorbid ADHD was defined by excluding ADHD-diagnosed youth with any clinician-reported comorbid psychiatric diagnoses, and such definition may be subject to misclassification bias because of undocumented psychiatric symptoms/comorbidities, such as aggression. Consequently, the information on diagnosis reflects the judgment of the treating physician, and is not as reliable as studies of individuals assessed according to research standards. However, clinician-reported diagnosis sheds light on the nature of presenting problems and the medications used to treat these problems in community populations. This information emphasizes the need for follow-up outcomes research to establish effectiveness, safety, and tolerability for such medication use in “real world” practice settings. Second, dispensed medications may not actually have been consumed, and may not necessarily measure actual medication adherence. However, this study of continuous enrollees permitted us to examine duration across many repeated dispensings (up to 298 days in the year). Long durations of use with repeated sequential dispensing are likely to reflect medication adherence. Third, gaps of <15 days between sequential antipsychotic dispensings were assumed to be continuous episodes of antipsychotic use (dosReis et al. 2002; Sikka et al. 2005; Constantine et al. 2010; dosReis et al. 2011). Although this is a widely utilized algorithm to measure duration of medication use episodes, it is likely to inflate the actual annual duration of atypical antipsychotic use. However, this algorithm is not likely to bias the statistical differences among study covariates in relation to annual duration of atypical antipsychotic use, under the assumption that the inflation will not occur differentially across subsets of the study population. Further, our data avoid the self-report and recall biases of survey methods to assess medication utilization.

Finally, because of the cross-sectional nature of its design, the study had limitations with respect to assessing the temporal relationship of antipsychotic exposure and prior treatment patterns, in particular among youth diagnosed with ADHD. Longitudinal study designs with observation periods >1 year are needed to identify new users of atypical antipsychotics and to subsequently examine their prior medication exposures, such as stimulant or atomoxetine use. Such longitudinal study designs would offer information on the rationale for off-label prescribing of atypical antipsychotics, such as whether off-label antipsychotic exposure occurs after a reasonable duration of first-line medication treatment. In addition, a new-user design with longer observation periods would allow persistence of atypical antipsychotic use to be measured from treatment initiation to discontinuation. Using persistence as a measure of duration in new-user designs would allow for a more uniform distribution of measured and unmeasured clinical factors among foster care youth and income-eligible Medicaid-insured youth and would therefore make stronger inferential comparisons in duration of atypical antipsychotic use. However, the findings from a new-user design would be only generalizable to the much smaller population of new users.

Clinical Significance

The implications of the study findings should be interpreted with caution. For youth with severe and chronic psychiatric problems, such as schizophrenia and pediatric bipolar disorder, long-term adherence to atypical antipsychotic medications is usually desired for clinically meaningful improvement and management. Relatively shorter annual duration of use for youth with behavioral disorders may suggest clinical improvement in the management of behavioral disorders. Nevertheless, it is evident from the study findings that in 2006, one half of Medicaid-insured youth who were treated with an antipsychotic medication remained on the medication for ≥6 months. Long durations increase the risk of cardiometabolic adverse events in the absence of appropriate cardiometabolic monitoring, which is particularly serious in young children and vulnerable populations.

Conclusions

Atypical antipsychotic use for behavioral conditions in youth populations is prominent, with substantially longer durations of exposure in foster care youth and very young children. Exposure to atypical antipsychotics, in particular for children in foster care and those diagnosed with ADHD was substantial, warranting independently conducted outcomes research for long-term effectiveness, safety, and appropriate cardiometabolic monitoring.

Footnotes

Acknowledgment

The authors thank the Maryland Department of Health and Mental Hygiene for generous access to the data.

Disclosures

No competing financial interests exist.

References

AACAP Work Group on Quality Issues: Practice parameter for the use of atypical antipsychotic medications in children and adolescents, 2010. Available at http://www.aacap.org/App_Themes/AACAP/docs/practice_parameters/Atypical_Antipsychotic_Medications_Web.pdf. Accessed January 17, 2014 .

Administration for Children and Families: Information memorandum: Promoting the safe, appropriate, and effective use of psychotropic medication for children in foster care, 2012. Available at http://www.acf.hhs.gov/sites/default/files/cb/im1203.pdf. Accessed January 17, 2014 .

Alan

, Kultur

SEC

: Change in metabolic parameters and weight in response to risperidone monotherapy in young children with nonpsychotic disorders: A prospective open-label study. Child Adolesc Mental Health 15 Jul 2013; DOI: 10.1111/camh.12033.

American Academy of Child and Adolescent Psychiatry: Practice parameter on the use of psychotropic medication in children and adolescents. J Am Acad Child Adolesc Psychiatry, 48:961–973, 2009.

American Diabetes Association, American Psychiatric Association, American Association of Clinical Endocrinologists, and North American Association for the Study of Obesity: Consensus development conference on antipsychotic drugs and obesity and diabetes. Diabetes Care, 27:596–601, 2004.

Andrade

, Lo

, Roblin

, Fouayzi

, Connor

, Penfold

, Chandra

, Reed

, Gurwitz

: Antipsychotic medication use among children and risk of diabetes mellitus. Pediatrics, 128:1135–1141, 2011.

Biederman

, Mick

, Hammerness

, Harpold

, Aleardi

, Dougherty

, Wozniak

: Open-label, 8-week trial of olanzapine and risperidone for the treatment of bipolar disorder in preschool-age children. Biol Psychiatry, 58:589–594, 2005.

Calarge

, Acion

, Kuperman

, Tansey

, Schlechte

: Weight gain and metabolic abnormalities during extended risperidone treatment in children and adolescents. J Child Adolesc Psychopharmacol, 19:101–109, 2009.

Cohen

, Bonnot

, Bodeau

, Consoli

, Laurent

: Adverse effects of second-generation antipsychotics in children and adolescents: A Bayesian meta-analysis. J Clin Psychopharmacol, 32:309–316, 2012.

10.

Constantine

, Boaz

, Tandon

: Antipsychotic polypharmacy in the treatment of children and adolescents in the fee-for-service component of a large state Medicaid program. Clin Ther, 32:949–959, 2010.

11.

Constantine

, Jentz

, Bengtson

, McPherson

, Andel

, Jones

: Exposure to antipsychotic medications over a 4-year period among children who initiated antipsychotic treatment before their sixth birthday. Pharmacoepidemiol Drug Saf, 21:152–160, 2012.

12.

Constantine

, Tandon

, McPherson

, Andel

: Early diagnoses and psychotherapeutic treatment experiences of a cohort of children under 6 years old who received antipsychotic treatment in Florida's Medicaid program. J Child Adolesc Psychopharmacol, 21:79–84, 2011.

13.

Cooper

, Arbogast

, Ding

, Hickson

, Fuchs

, Ray

: Trends in prescribing of antipsychotic medications for US children. Ambul Pediatr, 6:79–83, 2006.

14.

Cooper

, Hickson

, Fuchs

, Arbogast

, Ray

: New users of antipsychotic medications among children enrolled in TennCare. Arch Pediatr Adolesc Med, 158:753–759, 2004.

15.

Correll

, Manu

, Olshanskiy

, Napolitano

, Kane

, Malhotra

: Cardiometabolic risk of second-generation antipsychotic medications during first-time use in children and adolescents. J Am Med Assoc, 302:1765–1773, 2009.

16.

Crystal

, Olfson

, Huang

, Pincus

, Gerhard

: Broadened use of atypical antipsychotics: safety, effectiveness, and policy challenges. Health Aff (Millwood), 28:w770–w781, 2009.

17.

dosReis

, Yoon

, Rubin

, Riddle

, Noll

, Rothbard

: Antipsychotic treatment among youth in foster care. Pediatrics, 128:e1459–e1466, 2011.

18.

dosReis

, Zito

, Buchanan

, Lehman

: Antipsychotic dosing and concurrent psychotropic treatments for Medicaid-insured individuals with schizophrenia. Schizophr Bull, 28:607–617, 2002.

19.

Egger

: A perilous disconnect: Antipsychotic drug use in very young children (Editorial). J Am Acad Child Adolesc Psychiatry, 49:3–6, 2010.

20.

Government Accountability Office: Foster children: HHS guidance could help states improve oversight of psychotropic prescriptions, 2011. Available at http://www.gao.gov/assets/590/586570.pdf. Accessed January 17, 2014 .

21.

Government Accountability Office: Children's mental health: Concerns about appropriate services for children in Medicaid and foster care. GAO Report, GAO-13-15, 1-54, 2012. Available at http://www.gao.gov/assets/660/650716.pdf. Accessed January 17, 2014 .

22.

Harris

, Sorbeno

, Kogan

, Schuster

, Stein

: Concurrent mental health therapy among Medicaid-enrolled youths starting antipsychotic medications. Psychiatr Serv, 63:351–356, 2013.

23.

Joshi

, Petty

, Wozniak

, Faraone

, Doyle

, Georgiopoulos

, Hammerness

, Walls

, Glaeser

, Brethel

, Yorks

, Biederman

: A prospective open-label trial of quetiapine monotherapy in preschool and school age children with bipolar spectrum disorder. J Affect Disord, 136:1143–1153, 2012.

24.

Laita

, Cifuentes

, Doll

, Llorente

, Cortes

, Parellada

, Moreno

, Ruiz-Sancho

, Graell

, Arango

: Antipsychotic-related abnormal involuntary movements and metabolic and endocrine side effects in children and adolescents. J Child Adolesc Psychopharmacol, 17:487–501, 2007.

25.

Matone

, Localio

, Huang

, dosReis

, Feudtner

, Rubin

: The relationship between mental health diagnosis and treatment with second-generation antipsychotics over time: A national study of U.S. Medicaid-enrolled children. Health Serv Res, 47:1836–1860, 2012.

26.

McIntyre

, Jerrell

: Metabolic and cardiovascular adverse events associated with antipsychotic treatment in children and adolescents. Arch Pediatr Adolesc Med, 162:929–935, 2008.

27.

Morrato

, Druss

, Hartung

, Valuck

, Allen

, Campagna

: Metabolic testing rates in 3 state Medicaid programs after FDA warnings and ADA/APA recommendations for second-generation antipsychotic drugs. Arch Gen Psychiatry, 67:17–24, 2010a.

28.

Morrato

, Nicol

, Maahs

, Druss

, Hartung

, Valuck

, Campagna

, Newcomer

: Metabolic screening in children receiving antipsychotic drug treatment. Arch Pediatr Adolesc Med, 164:344–351, 2010b.

29.

Naylor

, Davidson

, Ortega–Piron

, Bass

, Gutierrez

, Hall

: Psychotropic medication management for youth in state care: Consent, oversight, and policy considerations. Child Welfare, 86:175–192, 2007.

30.

Olfson

, Blanco

, Liu

, Moreno

, Laje

: National trends in the outpatient treatment of children and adolescents with antipsychotic drugs. Arch Gen Psychiatry, 63:679–685, 2006.

31.

Olfson

, Blanco

, Liu

, Wang

, Correll

: National trends in the office-based treatment of children, adolescents, and adults with antipsychotics. Arch Gen Psychiatry, 69:1247–1256, 2012.

32.

Olfson

, Crystal

, Huang

, Gerhard

: Trends in antipsychotic drug use by very young, privately insured children. J Am Acad Child Adolesc Psychiatry, 49:13–23, 2010.

33.

Olfson

, Marcus

, Weissman

, Jensen

: National trends in the use of psychotropic medications by children. J Am Acad Child AdolescPsychiatry, 41:514–521, 2002.

34.

P.L. 110-351. Fostering Connections to Success and Increasing Adoptions Act, 2008. Available at https://www.childwelfare.gov/systemwide/laws_policies/federal/index.cfm?event=federallegislation.viewlegis&id=121. Accessed January 17, 2014 .

35.

P.L. 112-34. The Child and Family Services Improvement and Innovation Act, 2011. Available at https://www.childwelfare.gov/systemwide/laws_policies/federal/index.cfm?event=federalLegislation.viewLegis&id=122. Accessed January 17, 2014 .

36.

Patel

, Crismon

, Hoagwood

, Johnsrud

, Rascati

, Wilson

, Jensen

: Trends in the use of typical and atypical antipsychotics in children and adolescents. J Am Acad Child Adolesc Psychiatry, 44:548–556, 2005.

37.

Roberts

, Rodriguez

, Murphy

, Crescenzi

: Pediatric drug labeling. JAMA, 290:905–911, 2003.

38.

Safer

: A comparison of risperidone-induced weight gain across the age span. J Clin Psychopharmacology, 24:429–436, 2004.

39.

Sikka

, Xia

, Aubert

: Estimating medication persistency using administrative claims data. Am J Manag Care, 11:449–457, 2005.

40.

State of Maryland DHMH: Expansion of the peer review pre-authorization program of antipsychotic medications for youth less than 18 years old (Rep. No. PT 26–13), 2013. Available at https://mmcp.dhmh.maryland.gov/pap/docs/PT%2026-13%20FINAL.pdf. Accessed January 17, 2014 .

41.

Varley

, McClellan

: Implications of marked weight gain associated with atypical antipsychotic medications in children and adolescents (commentary). JAMA, 302:1811–1812, 2009.

42.

West

, Strom

, Freundlich

, Normand

, Koch

, Savitz

: Completeness of prescription recording in outpatient medical records from a health maintenance organization. J Clin Epidemiol, 47:165–171, 1994.

43.

Zito

, Burcu

, Ibe

, Safer

, Magder

: Antipsychotic use by Medicaid-insured youths: Impact of eligibility and psychiatric diagnosis across a decade. Psychiatr Serv, 64:223–229, 2013.