Adverse Events in Very Young Children Prescribed Psychotropic Medications: Preliminary Findings from an Acute Clinical Sample

Abstract

Objective:

The present study used the Pediatric Adverse Events Rating Scale (PAERS) to provide a systematic assessment of adverse events (AEs) related to psychotropic medication use in a clinical sample of young children attending a specialized, early childhood partial hospital program. Study goals were as follows: 1) To describe the frequency and types of specific psychotropic medication-related AEs experienced by very young children (ages 3–7 years) in an acute clinical sample, and 2) to identify the psychotropic medication(s) and/or class(es) associated with the highest frequency of AEs.

Methods:

Results:

The percentages of children who experienced one or more AEs attributed to a psychiatric medication ranged from 0 (sertraline, melatonin) to 41.2% (fluoxetine), with wide variability in the types AEs reported. The overall frequencies of events caused by a stimulant were similar across the two medications examined (21.4% and 27.7% for mixed amphetamine salts and methylphenidate, respectively), with mood-related difficulties and decreased appetite being the most common AEs reported. The frequencies of AEs caused by an α agonist were also similar across the two medications examined (9.8% and 17.2% for guanfacine and clonidine, respectively), with fatigue as the most commonly reported AE. With respect to the selective serotonin reuptake inhibitor (SSRI) class, there was a trend for fluoxetine to be associated with more AEs (41.2%) than sertraline (for which no AEs were reported). The most common AEs reported for fluoxetine were impulsivity and poor concentration.

Conclusions:

The data presented here support existing literature reporting differences in AEs between age groups. More rigorous studies are warranted to further examine the types and frequencies of AEs related to psychotropic medications in very young children.

Introduction

Research focused on the adverse events (AEs) of psychopharmacological interventions for very young children with psychiatric disorders lags behind prescription practice in community and hospital settings. Although recent epidemiological samples have indicated a sharp rise in the rates of medication use in young children (Zito et al. 2007; Huefner and Griffith 2014), limited literature is available about the risks, benefits, and alternatives for medication treatment among preschoolers, toddlers, and young school-age children. As a result, clinicians and families of very young children frequently must make decisions about psychopharmacological treatments based on literature about older populations, or clinical anecdotes.

To date, only a few studies have evaluated AEs associated with psychiatric medication use in very young children (Huefner and Griffith 2014). The majority of published studies involving young children and psychotropic medication primarily address effectiveness. Of notable exception are the Preschool ADHD Treatment Study (PATS) (Wigal et al. 2006), which examines the effectiveness, safety, and tolerability of methylphenidate in preschool-age children, and the Pediatric OCD Treatment Study (POTS) (Pediatric OCD Treatment Study Team 2004) and the Pediatric OCD Treatment Study II (POTS II) (Franklin et al. 2011), which assesses not only the effectiveness of sertraline in young children with obsessive-compulsive disorder (OCD), but also the AEs associated with sertraline in children 7–12 years of age. Additionally, in their 2006 article, Safer and Zito discuss their pooled AE data findings on a review of published double-blind, placebo-controlled selective serotonin reuptake inhibitor (SSRI) studies of children (6–11 years old) and adolescents (12–18 years old) that separated AEs by age group (Safer and Zito 2006). Another review by Safer in 2011 outlines the differences in AEs among children (including some preschool-aged groups), adolescents, and adults (Safer 2011). However, each of these studies was limited to a specific medication or medication class and an older study population. To date, no studies have examined AEs across medication classes with a specific emphasis on very young children.

The present study used the Pediatric Adverse Events Rating Scale (PAERS) (March et al. 2007) to provide a systematic assessment of adverse events related to psychotropic medication use in a clinical sample of young children attending a specialized, early childhood partial hospital program. As such, this study is distinct in several important ways. Whereas the vast majority of the current published literature on psychotropic-related AEs in children uses a variety of measures across providers and sites to assess for AE, this study utilizes the PAERS, which was designed to provide an efficient, yet comprehensive and standardized measure to evaluate for psychotropic- related AEs in children. Preliminary evidence of the reliability and validity of the PAERS is indicated in the POTS II (Franklin et al. 2011). Second, this study examines AEs within a real-world, highly impaired sample of children in a clinical service. Third, this study examines AEs in very young children (3–7 years old), and, therefore, provides additional information on the types and frequencies of AEs in an age group that has been under increasing public scrutiny, but relatively underrepresented in the literature. Study goals were as follows: 1) To describe the frequency and types of specific psychotropic medication-related AEs experienced by very young children (ages 3–7 years) in an acute clinical sample, and 2) to identify the psychotropic medication(s) and/or class(es) associated with the highest frequency of AEs.

Methods

Participants

Participants were 158 children (118 males; ages 36–95 months, mean=66 months, SD=14.6 months) who presented to a hospital-based day treatment program for young children with severe emotional and behavioral problems, and were prescribed a psychotropic medication at any point during the hospitalization, either by the partial hospital program child psychiatrist, or by an outpatient prescriber. Data were collected over a period of ∼3 years, from 2011 to 2014. An additional 215 children, admitted during this same time period, were not prescribed psychotropic medications and were, therefore, not included in the current report. A parent or guardian provided informed consent for child and family participation in the study. The hospital's Institutional Review Board approved all study measures.

Children were referred to the day treatment program for a variety of significant clinical concerns, including serious other-and self-directed aggression, persistent out-of-control tantrums, anxiety and mood dysregulation, feeding and eating problems, noncompliance with directions, and general family dysfunction. For most children, these clinical problems were quite severe, with mean parent-reported behavior problem symptoms well beyond established clinical cutoffs on norm-referenced measures of early childhood behavior problems. Additionally, most children referred to the program had previously failed treatment at a lower level of care. Families of participating children also presented with high levels of general distress and dysfunction. A summary of child and family clinical characteristics, along with sample demographics, is presented in Table 1 (Radloff 1977; Abidin 1995; Achenbach and Rescorla 2000; Achenbach and Rescorla 2001; Dunn and Dunn 2007).

Table 1.

Demographic and Clinical Characteristics of the Sample

Demographic characteristics
Gender of child	74.7% male
Age of child	Mean=65.5 months (SD=14.6)
Child ethnicity (%)	White/Non-Hispanic: n=73 (46.2%)
	Black/African-American: n=4 (2.5%)
	Hispanic/Latino: n=18 (11.4)%
	Other/Unspecified: n=63 (39.8%)
Single parent household (%)	39.9%
Family unemployment (%)	29.1%
Maternal education high school/GED or less (%)	43.2%
Family income	Median=$32, 500

Child and family clinical measures
CBCL Child Internalizing Problems (t score)	CBCL/1.5- 5 years: Mean=66.10 (SD=8.86; 61% above clinical cutoff score)
	CBCL/6−18 years: Mean=65.09 (SD=9.94; 56.1% above clinical cutoff score)
CBCL Child Externalizing Problems (t score)	CBCL/1.5-5 years: Mean=77.13 (SD=9.65; 90% above clinical cutoff score)
	CBCL/6–18 years: Mean=71.28 (SD=8.07; 82.2% above clinical cutoff score)
PPVT-IV Child Receptive Language (t score)	Mean=99.57 (SD=14.96)
CESD Maternal Depression	Mean=22.93 (SD=12.97; 71.2% above clinical cutoff score)
PSI Maternal Parenting Stress	M=107.50; SD=21.85; 84.8% above clinical cutoff score)

CBCL, Child Behavior Checklist (clinical range≥64; Achenbach and Rescorla 2000; Achenbach and Rescorla, 2001); PPVT-IV, Peabody Picture Vocabulary Test (Mean=100; SD=15; Dunn and Dunn 2007); CESD=Center for Epidemiological Studies Depression Scale (clinical cutoff score≥16; Radloff 1977); PSI=Parenting Stress Index (clinical cutoff score≥85th percentile; Abidin 1995).

Measures

The PAERS-Parent Version is a 45 item scale that was completed by each child's caregiver upon admission and every following week (March et al. 2007). The scale is composed of questions about the presence and severity of symptoms (including behavioral, emotional, and physical domains) experienced by the child in the previoue week. The PAERS-Clinician Version was completed by the primary child psychiatrist for each child who entered the program on medications, or started a medication during treatment, and was based on the completed caregiver PAERS. When indicated, the primary child psychiatrist would conduct a brief follow-up interview with the caregiver on responses that were not consistent with side effects elicited during regular open-ended discussions about family observations of the child while their child was on medication. The clinician PAERS also includes questions about the relationship between symptoms reported and medications, and whether the symptom severity has lead to impairment in functioning.

Results

For the purposes of this study, we focused on six commonly prescribed medications (methylphenidate, mixed amphetamine salts guanfacine, clonidine, fluoxetine, and sertraline), across three medication classes (stimulant, α-agonist, SSRI). We also examined melatonin as a commonly used sleep aid. These selected agents represented medications that were frequently prescribed in our sample, and for which the sample size was adequate for analysis. Eight children on whom the PAERS was completed were excluded from these analyses, because of not being prescribed fluoxetine, sertraline, Adderall, methylphenidate, guanfacine, clonidine, or melatonin. Additionally, given the very small number of children in our sample who were prescribed an antipsychotic medication (risperidone, n=11; aripiprazole, n=3; and quetiapine, n=2), we were unable to perform analyses on the frequency of AEs related to an antipsychotic. Descriptive analyses are presented in Table 2. As shown, the percentages of children who experienced one or more AEs attributed to a psychiatric medication ranged from 0 (sertraline, melatonin) to 41.2% (fluoxetine), with wide variability in the types AEs reported. Odds ratios (ORs) were computed to determine whether frequencies of AEs differed across medications, within a medication class.¹ The overall frequencies of events caused by a stimulant medication were similar across the two medications examined (21.4% and 27.7% and for mixed amphetamine salts and methylphenidate, respectively), OR=0.87, 95% confidence interval (CI)=0.26–2.92, with mood-related difficulties (e.g., irritability, anger/lability) and decreased appetite being the most common AEs reported. The frequencies of AEs caused by an α-agonist were also similar across the two medications examined (9.8% and 17.2% for guanfacine and clonidine, respectively), OR=0.63, 95% CI=0.14–2.79 with fatigue as the most commonly reported AE. With respect to the SSRI medications, there was a trend for fluoxetine to be associated with more AEs (41.2%) than sertraline (for which no AEs were reported), OR=17.11, 95% CI=0.85–342.76. The most common AEs reported for fluoxetine were impulsivity and poor concentration.

Table 2.

Frequency of Adverse Events (AE) by Medication

Medication	No AE(s) reported n (%)	One or more AE(s) reported n (%)	Specific AE(s) reported
Methylphenidate (n=47)	34 (72.3%)	13 (27.7%)	Irritability or anger	n=6 (12.8%)
			Lability	n=6 (12.8%)
			Sadness/depression or apathy	n=4 (8.5%)
			Decreased appetite	n=3 (6.4%)
			Tics	n=3 (6.4%)
			Insomnia	n=3 (6.4%)
			Weight loss	n=2 (4.3%)
			Nausea or vomiting	n=2 (4.3%)
			Abdominal pain	n=1 (2.1%)
			Racing thoughts	n=1 (2.1%)
			Motor restlessness:	n=1 (2.1%)
mixed amphetamine salts (n=28)	22 (78.6%)	6 (21.4%)	Decreased appetite	n=5 (17.9%)
			Lability	n=3 (10.8%)
			Sadness/depression or apathy	n=2 (7.1%)
			Abdominal pain	n=1 (3.6%)
			Headache:	n=1 (3.6%)
			Anxiety	n=1 (3.6%)
			Irritability or anger	n=1 (3.6%)
			Assaultive:	n=1 (3.6%)n
			Insomnia	n=1 (3.6%)
Fluoxetine (n=17)	10 (58.8%)	7 (41.2%)	Impulsivity	n=2 (11.8%)
			Trouble concentrating	n=2 (11.8%)
			Increased appetite	n=1 (5.9%)
			Excessive thirst	n=1 (5.9%)
			Weight gain	n=1 (5.9%)
			Bruising/bleeding	n=1 (5.9%)
			Rash	n=1 (5.9%)
			Anxious mood	n=1 (5.9%)
			Racing thoughts	n=1 (5.9%)
Sertraline (n=14)	14 (100%)	0	N/A
Guanfacine (n=41)	37 (90.2%)	4 (9.8%)	Fatigue	n=2 (4.9%)
			Muscle weakness or shakiness	n=1 (2.4%)
			Racing heart	n=1 (2.4%)
			Irritability or angrt/hostility	n=1 (2.4%)
			Excessive thirst	n=1 (2.4%)
			Dizziness	n=1 (2.4%)
Clonidine (n=29)	24 (82.8%)	5 (17.2%)	Fatigue	n=3 (10.3%)
			Stomachache	n=1 (3.4%)
			Nausea	n=1 (3.4%)
			Racing heart	n=1 (3.4%)
			Dizziness	n=1 (3.4%)
Melatonin (n=47)	47 (100%)	0 (0%)	N/A

We were also interested in exploring child age as a factor that might be related to increased risk for AEs, hypothesizing that the very youngest children in our sample might be particularly vulnerable. To examine this possibility, we conducted a series of independent t tests to compare the mean age of children who did or did not experience an AE for each of the medications examined. There were no significant age differences found, across all medications examined.

Discussion

The frequencies of AEs in this sample of very young children, as compared with the frequencies of AEs reported in existing literature for older children and adults, appear to be higher for some medications (methylphenidate, fluoxetine) and lower for others (sertraline). Additionally, the type of AE seen most frequently in each of the studied psychotropic medications seemed to differ between very young children and older children.

Fluoxetine was associated with the highest frequency of AEs (41.2%) in this clinical sample, representing nearly four times the frequency of AEs (11%) reported previously among adolescent populations. The most frequent AEs in this study included trouble concentrating and impulsivity.

An unexpected finding in this study was the absence of AEs associated with sertraline. Frequencies of AEs reported in rigorous studies such as the POTS (Pediatric OCD Treatment Team 2004) range from 4% to 29% of children and adolescents 7–17 years old, with the higher frequency AEs being physical rather than psychiatric. Clearly there are many differences between both the study samples and the design in our study compared with the POTS, making it impossible to make direct comparisons of the results.

Not surprisingly, the frequencies of AEs for very young children on methylphenidate in this study are comparable to those found in similarly aged children in the PATS (Wigal et al. 2006), with the most frequent AEs being mood symptoms such as irritability and emotionality.

There are several important factors that warrant caution in the interpretation of the results in this study. First, the sample size of individual medication classes was relatively small, which limited power to identify AE differences among medication agents within classes. Second, the method of assessing AEs in this study did not permit evaluation for the resolution of, or change in, severity of any reported AE. The PAERS allows for the attribution of AE to a “study drug,” or “other drug” (in our study, this would refer to outpatient medication), but does not account for a change in AE over time or related to dosing. Information on dose-dependent AEs, or side effects that wane with prolonged treatment, is clearly relevant to both prescribers and families during the medication decision-making process. Nonetheless, the PAERS offers a comprehensive and systematic way of evaluating for psychotropic-related AEs in children that is relatively easy to use. As such, it merits consideration as a standard by which all clinicians may assess AEs. Third, the population in this study was far more heterogeneous than those included in the PATS and POTS, with respect to presenting complaints and comorbidities; additionally, the different methodology in collecting data on AEs in each of the studies and the differences in study design may have accounted for any differences in rates of AEs. A final limitation of this study is the relatively brief period of observation and assessment, which may not have been sufficient to evaluate for side effects associated with the SSRI class.

Conclusions

Despite its limitations, this study addresses a critical but underresearched area by providing clinically relevant, systematically collected data on an issue that has long been recognized in both the psychiatric and pediatric community (Zito et al. 2000), and has been increasingly at the forefront of public and media attention. Moreover, as this study was incorporated into an existing treatment program and thus able to utilize a real-world sample of impaired preschoolers, it provides a practical model for the busy practitioner to learn through a systematic evaluation of current clinical practice. The data presented here support existing literature reporting differences in AEs between age groups. On a more positive note, this study affirms melatonin as a well-tolerated option for even very young children, at least in the short term. More rigorous studies are certainly warranted to examine the safety and tolerability of psychotropic medications in very young children.

Clinical Significance

Recent data presented in both mental health publications and public media reflect a significant increase in the use of psychotropic medications in very young children. Research focused on the AEs of psychopharmacological interventions for very young children with severe impairment from psychiatric disorders lags behind prescription practice in the community and hospital settings. Although recent epidemiological samples have indicated a sharp rise in the rates of medication use in young children, providers and families must consider potential risks and benefits of, and reasonable alternatives to, psychopharmacological treatments based on extremely limited available literature and clinical anecdotes. Prescribers need to be aware of the risks of psychotropic medication use in very young children. The present study used the PAERS (March et al. 2007) to provide a systematic assessment of AEs related to psychotropic medication use in a clinical sample of young children attending a specialized, early childhood partial hospital program. It thus provides additional information on the types and frequencies of AEs in an age group that has been under increasing public scrutiny, but relatively underrepresented in the literature.

Footnotes

Disclosures

No competing financial interests exist.

References

Abidin

: Parenting Stress Index, 3rd ed. Odessa, FL: Psychological Assessment Resources; 1995.

Achenbach

, Rescorla

: Manual for the ASEBA Preschool Forms and Profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families; 2000.

Achenbach

, Rescorla

: Manual for the ASEBA Preschool Forms and Profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families; 2001.

Dunn

, Dunn

: Peabody Picture Vocabulary Test, 4th ed. Minneapolis: Pearson Assessments; 2007.

Franklin

, Sapyta

, Freeman

, Khanna

, Compton

, Almirall

, Moore

, Choate-Summers

, Garcia

, Edson

, Foa

, March

: Cognitive behavior therapy augmentation of pharmacotherapy in pediatric obsessive-compulsive disorder: The Pediatric OCD Treatment Study II (POTS II) randomized controlled trial. JAMA, 306:1224–1232, 2011.

Huefner

, Griffith

: Psychotropic medication use with troubled children and youth. J Child Fam Stud, 23:613–616, 2014.

March

, Karayal

, Chrisman

: CAPTN: The pediatric adverse event rating scale. Boston: Scientific Proceedings of the Annual Meeting of the American Academy of Child and Adolescent Psychiatry; 2007.

Pediatric OCD Treatment Study Team: Cognitive-behavior therapy, sertraline, and their combination for children and adolescents with obsessive-compulsive disorder: The Pediatric OCD Treatment Study (POTS) randomized controlled trial. JAMA, 292:1969–1976, 2004.

Radloff

: The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Meas, 1:385–401, 1977.

10.

Safer

: Age-grouped differences in adverse drug events from psychotropic medication. J Child Adolesc Psychopharmacol, 21:299–309, 2011.

11.

Safer

, Zito

: Treatment-emergent adverse events from selective serotonin reuptake inhibitors by age group: Children versus adolescents. J Child Adolesc Psychopharmacol, 16:159–169, 2006.

12.

Wigal

TIM

, Greenhill

, Chuang

, McGough

, Vitiello

, Skrobala

, Swanson

, Wigal

, Abikoff

, Kollins

, McCracken

, Riddle

, Posner

, Ghuman

, Davies

, Thorp

, Stehli

: Safety and tolerability of methylphenidate in preschool children with ADHD. J Am Acad Child Adolesc Psychiatry, 45:1294–1303, 2006.

13.

Zito

, Safer

, Reis

, Gardner

, Boles

, Lynch

: Trends in the prescribing of psychotropic medications to preschoolers. JAMA, 283:1025–1030, 2000.

14.

Zito

, Safer

, Valluri

, Gardner

, Korelitz

, Mattison

: Psychotherapeutic medication prevalence in medicaid-insured preschoolers. J Child Adolesc Psychopharmacol, 17:195–203, 2007.