BACKGROUND:

Pediatric use of second-generation antipsychotics, or neuroleptics, has increased over the past decade. Neuroleptic use can have significant and lasting adverse neurologic, metabolic, and cardiovascular effects. In the current literature, neuroleptic prescribing and monitoring is described in outpatient settings, with little description of inpatient pediatric practice. In this study, we are the first to explore prescribing and monitoring in inpatient pediatrics, highlighting similarities and differences in practice between pediatric medicine and psychiatry.

METHODS:

This retrospective study included patients <18 years of age who received a neuroleptic during inpatient hospitalization between September 2014 and March 2015, within either the pediatric inpatient medical or psychiatric setting. Data collected included sex, age, race, height, weight, length of stay, service providing care, details involving the neuroleptic(s) administered, comorbidities, lipid and glycemic monitoring, and results, monitoring for extrapyramidal symptoms, and mental health consultation.

RESULTS:

Factors associated with improved neuroleptic monitoring included longer length of stay and evidence of an adverse drug effect. Clearly specified indication for use was associated with improved neuroleptic monitoring. Although neuroleptic initiation during admission had improved indication documentation and monitoring practices compared with initiation before admission, a significant gap exists between inpatient psychiatry and medical settings in adverse drug effect monitoring, particularly extrapyramidal symptom monitoring and lipid collection.

CONCLUSIONS:

In our study, we describe current practice in the use and monitoring of neuroleptics in inpatient pediatric medical and psychiatric settings. In this study, we suggest that the use of neuroleptics in inpatient pediatrics, particularly inpatient medical settings, is conducted with suboptimal monitoring and, at times, without clear documented indication.

Neuroleptic (also known as antipsychotic) use in children and adolescents has rapidly increased, with outpatient medical visits by youth involving neuroleptic management nearly tripling over the last decade.13  Neuroleptics have been approved by the US Food and Drug Administration (FDA) for use in children and adolescents to treat certain conditions (Supplemental Table 8).4,5  Much use is off-label, with most outpatient neuroleptic use for indications not approved by the FDA.6,7  Off-label use is associated with rising costs and may have limited evidence-based support.6  Off-label use is often associated with the management of agitation, aggression, and disruptive behaviors, indications that are multifactorial, complex, and require a comprehensive approach to management.8,9  In 2017, the American Society of Clinical Oncology revised its clinical practice guidelines for the management of adults receiving chemotherapy with a high risk of nausea and emesis to propose the addition of olanzapine to the standard regimen.10  This practice is increasing in youth with minimal recognition of the significant metabolic effects of widespread use, despite olanzapine having the highest risk of metabolic adverse effects among second-generation neuroleptics.

Increasing pediatric neuroleptic use is more pronounced for second-generation neuroleptics, which may be related to increasing prevalence, diagnosis, and awareness of mental health conditions in pediatric health care.3,6,11  Although limited information exists on the long-term effects of second-generation neuroleptics on youth, current data suggest youth are at an increased risk of neuroleptic adverse drug effects (ADEs) compared with adults, particularly metabolic effects like increased adiposity, waist circumference, risk of developing Type 2 diabetes mellitus (DM), obesity, dyslipidemia, hypertension, and metabolic syndrome.3,8,1214  There are also considerable neurologic side effects in the form of extrapyramidal symptoms (EPSs), such as dystonic reactions, akathisia, Parkinsonian symptoms, and dyskinesias.15  The risk of EPSs in youth is increased compared with adults, with a relatively reduced risk for olanzapine and quetiapine, which are less dopaminergic.15  Given these considerable risks, there is a need to routinely monitor neurologic, cardiac, and metabolic ADEs in youth prescribed second-generation neuroleptics.3,6,16 

Despite the importance of ADE monitoring with neuroleptic use in youth, there is reduced awareness of these risks and insufficient training among pediatric medical providers, resulting in poor monitoring practices.17  Furthermore, there is poor access to child and adolescent psychiatrists, with primary care providers (PCPs) prescribing nearly 85% of psychotropic medications.18  Communication between PCPs and child and adolescent psychiatrists can be limited, resulting in inadequate coordination of care. A mismatch exists between the rapid rise in neuroleptic use and the relative lack of expertise in managing these medications.

Limited training in the use of neuroleptics is further magnified within inpatient pediatrics. Medically hospitalized youth are often exposed to considerable emotional and physiologic stressors, placing them at higher risk for agitation and aggression, a common indication for inpatient pediatric neuroleptic use.1921  Neuroleptics may also be used in the inpatient setting for the management of disruptive behaviors related to attention-deficit/hyperactivity disorder (ADHD), behavioral symptoms in youth with autism or conduct disorder, delirium, mood disorders, anxiety, sleep problems, or substance use disorders.3,6,7  Increasingly, neuroleptics are being used by PCPs and subspecialists as adjunctive management for physical symptoms such as sleep, nausea, or appetite stimulation, with limited consideration of the downstream metabolic and neurologic effects.7,2224  Transitions between inpatient and outpatient settings can be fraught with poor communication, compounding difficulties with monitoring. Although studies have examined outpatient neuroleptic prescribing and monitoring, few have examined pediatric inpatient use.13  In this study, we describe current prescribing and monitoring of neuroleptic use in a pediatric hospital and a child and adolescent psychiatry hospital (CAPH), both colocated in a large academic hospital setting, to assess overall neuroleptic practice and how neuroleptic prescribing and monitoring compare between inpatient medical and psychiatric settings within the same institution.

This is a single-center study of pediatric youth who received a neuroleptic while either hospitalized in medical or psychiatric settings between September 14, 2014, and March 15, 2015. Pediatric patients were defined as any patient admitted to any pediatric inpatient setting within the larger academic pediatric hospital between the ages of 0 and 17, inclusive. All pediatric patients were prescribed a scheduled or as-needed (PRN) neuroleptic and admitted to the pediatric hospital or CAPH that were included in this study. Within this period, only the first inpatient stay for each patient was included. Neuroleptic prescribing and monitoring were described, including scheduled and PRN use, indications for use, and ADE incidence. These practices were evaluated to determine how certain variables, such as timing of initiation or medical versus psychiatric settings, modulated monitoring. The study was approved by the University of Michigan Health System Institutional Review Board.

The 221-bed pediatric hospital operates independently from the remainder of the academic medical system, which is physically colocated. The hospital houses pediatric general, neonatal, and cardiothoracic ICUs, a pediatric emergency department, a colocated 16-bed CAPH, and access to all approved pediatric subspecialty services. There is an active child life service, pediatric psychology service, pediatric psychiatry consultation service, palliative care service, and adolescent medicine service. CAPH did not have access to child life or psychology consultation for nonpharmacologic psychotherapeutic intervention during this study period. In the larger health system, there is access to a psychiatric emergency department that serves patients across the life span. Each month, the psychiatry consultation service sees ∼35 new consults and a range of 100 to 140 follow-up consultations.

Information was obtained from the electronic medical record with access to all inpatient, outpatient, and urgent care settings within the University of Michigan Health System. The electronic medical record also has a feature termed “Care Everywhere” that provides access to the vast majority of hospitals, clinics, and urgent care settings within the State of Michigan to allow providers access to information from non–University of Michigan resources. Often times this information is incomplete, requires gathering further information from the patient and family, or requires release of records from the outside institution. Inpatient mental health and substance use treatment records from outside institutions are not available through Care Everywhere.

A retrospective chart review was conducted to obtain sex; age at admission; race; baseline comorbidities; height; weight; neuroleptic use on admission and discharge; admission setting; neuroleptic administration details (including frequency and administration count); ADE monitoring; lipid panel; blood glucose (BG) or hemoglobin A1c (HbA1c); documentation of EPSs or Abnormal Involuntary Movement Scale (AIMS) assessment for EPSs; indication for neuroleptic use; length of stay (LOS); use of psychiatry, psychology, and/or child life service; and outpatient follow-up scheduled at discharge. This information was managed by using Research Electronic Data Capture hosted at the University of Michigan.25 

During data collection, baseline comorbidities were limited to DM, hyperglycemia, dyslipidemia, hypertension, and obesity (defined by either explicit documentation or by being above the 85th percentile on the Centers for Disease Control and Prevention BMI-for-age clinical growth chart, as calculated by height and weight on admission). An administration frequency of once was classified as a PRN order.

Monitoring for ADEs was defined as explicit documentation, to capture intentional ADE monitoring rather than completion of monitoring for alternate reasons. For clinical assessment, neurologic assessment was stratified into 2 levels of monitoring: (1) completion of AIMS testing for EPSs or documentation of the presence or absence of EPSs and (2) no documentation of EPSs or AIMS testing. For laboratory collection, ADE monitoring was defined as collection of the appropriate laboratory assessments and documentation in the medical record, demonstrating intent to collect the laboratory assessments appropriate for neuroleptic monitoring. A lipid panel was defined as completed if done 3 months before admission, during the inpatient stay, or 3 months after admission. Per institutional laboratory standards, an abnormal lipid panel was defined as a cholesterol level >200 mg/dL, a triglyceride level >150 mg/dL, a high-density lipoprotein level <40 mg/dL, or a low-density lipoprotein level >160 mg/dL. A BG reading was defined as abnormal if any fasting (before 9 am, when breakfast is typically served) reading recorded was >126 mg/dL. HbA1c readings >6.5% were deemed abnormal.

Indication was defined on the basis of discharge summary because this is the primary means of communication to outpatient providers. If this did not provide sufficient information, indication was derived from the note on the day of neuroleptic initiation (or initial history and physical, if the neuroleptic was prescribed before admission). Indication was unspecified if the reason for use was not explicitly stated in these locations within the electronic medical record. When possible, indications were grouped on the basis of diagnoses from the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition. Use of psychology and/or child life service was used as a measure of nonpharmacological management; psychology consults that were focused on testing and assessment without therapeutic intervention were excluded from the study. Follow-up was defined as having an appointment scheduled per discharge summary.

Descriptive statistics were used to initially explore the data. The χ2 test was used in analyses of the association between monitoring parameters and their potential covariates. For analyses in which 2 cells had an expected count of ∼5, Fisher’s exact test was used. The Mann-Whitney test was used to determine if median LOS was statistically different between 2 groups, defined by potential covariates. All analyses were bivariate and were completed by using IBM SPSS Statistics (version 24; IBM Corporation, Chicago, IL), with significance defined as a P value <.05.

In Table 1, we describe the patients in our study. Among 152 patients, there were 196 unique orders for neuroleptics, most of which were second-generation neuroleptics (n = 192, 98%) and scheduled (n = 164, 84%). In Table 2, we describe agents prescribed on the basis of scheduled or PRN use and type of pediatric inpatient setting. In Tables 3 and 4, we illustrate indication on the basis of setting for scheduled and PRN use, respectively.

TABLE 1

Demographic Characteristics of Patient Population, Including Baseline Neuroleptic Use

CharacteristicCAPH Patients, No. (%)Pediatric Inpatient Medical Patients, No. (%)P
Sex .05 
 Male 42 (46) 37 (62)  
 Female 50 (54) 23 (38)  
Age, ya <.001 
 5–9 2 (2) 7 (12)  
 10–13 16 (17) 25 (42)  
 14–17 74 (80) 28 (47)  
Racea .12 
 White 75 (82) 49 (82)  
 African American 13 (14) 4 (7)  
 Asian 0 (0) 1 (2)  
 Native Hawaiian and/or other Pacific Islander 0 (0) 1 (2)  
 Hispanic or Latino 0 (0) 2 (3)  
 Other 2 (2) 3 (5)  
 Unknown 2 (2) 0 (0)  
Comorbidities 
 Type 1 DM 0 (0) 2 (3) .15 
 Type 2 DM 0 (0) 0 (0)  
 Hyperglycemia 0 (0) 1 (2) .40 
 Dyslipidemia 3 (3) 3 (5) .59 
 Hypertension 0 (0) 5 (8) .009 
 Obesityb 47 (54) 16 (48) .59 
 No comorbiditiesb 40 (46) 13 (39) .52 
Neuroleptic initiation .049 
 Before admission 60 (65) 48 (80)  
 During inpatient stay 32 (35) 12 (20)  
CharacteristicCAPH Patients, No. (%)Pediatric Inpatient Medical Patients, No. (%)P
Sex .05 
 Male 42 (46) 37 (62)  
 Female 50 (54) 23 (38)  
Age, ya <.001 
 5–9 2 (2) 7 (12)  
 10–13 16 (17) 25 (42)  
 14–17 74 (80) 28 (47)  
Racea .12 
 White 75 (82) 49 (82)  
 African American 13 (14) 4 (7)  
 Asian 0 (0) 1 (2)  
 Native Hawaiian and/or other Pacific Islander 0 (0) 1 (2)  
 Hispanic or Latino 0 (0) 2 (3)  
 Other 2 (2) 3 (5)  
 Unknown 2 (2) 0 (0)  
Comorbidities 
 Type 1 DM 0 (0) 2 (3) .15 
 Type 2 DM 0 (0) 0 (0)  
 Hyperglycemia 0 (0) 1 (2) .40 
 Dyslipidemia 3 (3) 3 (5) .59 
 Hypertension 0 (0) 5 (8) .009 
 Obesityb 47 (54) 16 (48) .59 
 No comorbiditiesb 40 (46) 13 (39) .52 
Neuroleptic initiation .049 
 Before admission 60 (65) 48 (80)  
 During inpatient stay 32 (35) 12 (20)  

P values were estimated by using χ2 test or Fisher’s exact test, depending on cell sizes.

a

Because of rounding, percentages may not total 100.

b

Calculated with n = 87 for the CAPH and n = 33 for other pediatric medical inpatient settings because there was a total of 32 patients who did not have both height and wt recorded to determine obesity status.

TABLE 2

Frequency of Use for Individual Neuroleptic Agents

Scheduled Frequency, No. (%)PRN Frequency, No. (%)
CAPHPediatric Inpatient Medical SettingsCAPHPediatric Inpatient Medical Settings
Aripiprazole 41 (25) 13 (8) 1 (3) 0 (0) 
Clozapine 4 (2) 0 (0) 0 (0) 0 (0) 
Haloperidol 0 (0) 1 (1) 0 (0) 3 (9) 
Lurasidone 2 (1) 0 (0) 0 (0) 0 (0) 
Olanzapine 8 (5) 5 (3) 4 (13) 3 (10) 
Paliperidone 0 (0) 2 (1) 0 (0) 0 (0) 
Quetiapine 22 (13) 13 (8) 2 (6) 4 (13) 
Risperidone 16 (10) 28 (17) 2 (6) 0 (0) 
Ziprasidone 7 (4) 2 (1) 13 (41) 0 (0) 
Scheduled Frequency, No. (%)PRN Frequency, No. (%)
CAPHPediatric Inpatient Medical SettingsCAPHPediatric Inpatient Medical Settings
Aripiprazole 41 (25) 13 (8) 1 (3) 0 (0) 
Clozapine 4 (2) 0 (0) 0 (0) 0 (0) 
Haloperidol 0 (0) 1 (1) 0 (0) 3 (9) 
Lurasidone 2 (1) 0 (0) 0 (0) 0 (0) 
Olanzapine 8 (5) 5 (3) 4 (13) 3 (10) 
Paliperidone 0 (0) 2 (1) 0 (0) 0 (0) 
Quetiapine 22 (13) 13 (8) 2 (6) 4 (13) 
Risperidone 16 (10) 28 (17) 2 (6) 0 (0) 
Ziprasidone 7 (4) 2 (1) 13 (41) 0 (0) 

Because of rounding, percentages may not total 100.

TABLE 3

Indications for Scheduled Neuroleptic Use

CAPH, No. (%)Pediatric Inpatient Medical Setting, No. (%)P
Autism spectrum disorder 3 (2) 11 (10) .001 
Bipolar and related disorders 6 (5) 5 (5) .65 
Schizophrenia and other psychotic disorders 11 (9) 3 (3) .16 
Tic disorder 2 (2) 3 (3) .38 
Disruptive, impulse-control, and conduct disorder 7 (6) 6 (5) .58 
Anxiety disorder 4 (3) 10 (9) .009 
Depressive disorder 16 (13) 11 (10) .84 
Feeding and eating disorder 3 (2) 0 (0) .28 
ADHD 0 (0) 15 (14) <.001 
Sleep-wake disorder 6 (5) 2 (2) .48 
Trauma- and stressor-related disorder 0 (0) 2 (2) .15 
Delirium 0 (0) 5 (5) .008 
Mood disorder, NOS 45 (36) 3 (3) <.001 
Agitation 10 (8) 5 (5) .63 
Nausea and/or vomiting 0 (0) 1 (1) .39 
Impulsivity 4 (3) 0 (0) .16 
None of the above 3 (2) 1 (1) 1.00 
Unspecified 4 (3) 28 (25) <.001 
CAPH, No. (%)Pediatric Inpatient Medical Setting, No. (%)P
Autism spectrum disorder 3 (2) 11 (10) .001 
Bipolar and related disorders 6 (5) 5 (5) .65 
Schizophrenia and other psychotic disorders 11 (9) 3 (3) .16 
Tic disorder 2 (2) 3 (3) .38 
Disruptive, impulse-control, and conduct disorder 7 (6) 6 (5) .58 
Anxiety disorder 4 (3) 10 (9) .009 
Depressive disorder 16 (13) 11 (10) .84 
Feeding and eating disorder 3 (2) 0 (0) .28 
ADHD 0 (0) 15 (14) <.001 
Sleep-wake disorder 6 (5) 2 (2) .48 
Trauma- and stressor-related disorder 0 (0) 2 (2) .15 
Delirium 0 (0) 5 (5) .008 
Mood disorder, NOS 45 (36) 3 (3) <.001 
Agitation 10 (8) 5 (5) .63 
Nausea and/or vomiting 0 (0) 1 (1) .39 
Impulsivity 4 (3) 0 (0) .16 
None of the above 3 (2) 1 (1) 1.00 
Unspecified 4 (3) 28 (25) <.001 

P values were estimated by using χ2 test or Fisher’s exact test, depending on cell sizes. Calculated with n = 124 for the CAPH and n = 111 for pediatric medical inpatient settings because some orders could have multiple indications for use. Because of rounding, percentages may not total 100.

TABLE 4

Indications for Neuroleptic PRN Use

CAPH, No. (%)Pediatric Inpatient Medical Settings, No. (%)P
Anxiety disorder 1 (6) 0 (0) 1.00 
Sleep-wake disorder 1 (6) 0 (0) 1.00 
Delirium 0 (0) 7 (64) <.001 
Agitation and/or aggression 14 (88) 4 (36) .03 
CAPH, No. (%)Pediatric Inpatient Medical Settings, No. (%)P
Anxiety disorder 1 (6) 0 (0) 1.00 
Sleep-wake disorder 1 (6) 0 (0) 1.00 
Delirium 0 (0) 7 (64) <.001 
Agitation and/or aggression 14 (88) 4 (36) .03 

P values were estimated by using χ2 test or Fisher’s exact test, depending on cell sizes. Calculated with n = 16 for the CAPH and n = 11 for pediatric medical inpatient settings because some orders could have multiple indications for use.

Initiation During Inpatient Stay

Among all patients studied, initiation during inpatient stay was more common in the CAPH (n = 34, P = .049) and was significantly associated with increased ADE monitoring (88% vs 51%, P < .001), lipid panel collection (74% vs 51%, P = .02), and EPS monitoring (50% vs 22%, P = .002). Among patients who continued taking the neuroleptic at discharge, 71% of patients who were started on a neuroleptic during their inpatient stay had psychiatric follow-up scheduled. Lipid panel collection was significantly associated with CAPH provision of care, with 88% of lipid panel completion of newly initiated neuroleptics done in the CAPH (P = .04). In addition, patients in the CAPH who were started on a neuroleptic during their inpatient stay were more likely to also have a height recorded (96% vs 43%, P = .003) to allow for BMI calculation.

Indication: FDA Approved, Unspecified

Most neuroleptic use was for FDA nonapproved indications (n = 114). Among FDA-indicated use, patients in the CAPH were associated with higher ADE monitoring completion (94% vs 46%, P = .004), lipid panel collection (89% vs 31%, P = .001), BG or HbA1c readings (100% vs 62%, P = .008), recorded height (94% vs 31%, P < .001), and psychiatric follow-up scheduled at discharge (81% vs 23%, P = .002) compared with patients in pediatric medical settings.

Approximately 1 in 5 of all patients had an unspecified indication, which was associated with lower ADE monitoring completion (28% vs 69%, P < .001), EPS monitoring completion (13% vs 34%, P = .02), lipid panel collection (25% vs 65%, P < .001), BG or HbA1c readings (69% vs 94%, P < .001), recorded height (47% vs 81%, P < .001), and psychiatric follow-up scheduled at discharge (24% vs 73%, P < .001). Patients with an unspecified indication for neuroleptic use were also more likely to have received their care on a pediatric inpatient medical setting (P < .001).

Psychiatry Involvement

Among all patients studied, patients in the CAPH had higher ADE monitoring completion (77% vs 33%, P < .001), EPS monitoring completion (Table 5), psychiatric follow-up scheduled at discharge (89% vs 19%, P < .001), lipid panel collection, BG or HbA1c readings, and recorded height (Table 6). Psychiatry was consulted in pediatric inpatient medical settings for ∼one-third of patients using neuroleptics. Psychiatric consultation was associated with higher ADE monitoring completion (74% vs 11%, P < .001), lipid panel collection (47% vs 14%, P = .007), BG or HbA1c readings (95% vs 58%, P = .005), EPS monitoring completion (47% vs 3%, P < .001), and psychiatric follow-up scheduled at discharge (50% vs 6%, P < .001). Of note, AIMS assessment was completed in pediatric inpatient medical settings only when psychiatry was consulted.

TABLE 5

EPS Monitoring Completion Rates for Patients on a Scheduled Neuroleptic

CAPH, No. (%)Pediatric Inpatient Medical Settings, No. (%)P
AIMS assessment 4 (5) 3 (6) 1.00 
Other EPS documentation 27 (31) 7 (13) .01 
Either of the above 31 (36) 10 (18) .02 
CAPH, No. (%)Pediatric Inpatient Medical Settings, No. (%)P
AIMS assessment 4 (5) 3 (6) 1.00 
Other EPS documentation 27 (31) 7 (13) .01 
Either of the above 31 (36) 10 (18) .02 

P values were estimated by using χ2 test or Fisher’s exact test, depending on cell sizes.

TABLE 6

Metabolic Monitoring Completion Rates for Patients on a Scheduled Neuroleptic

CAPH, No. (%)Pediatric Inpatient Medical Settings, No. (%)P
Lipid panel 65 (76) 14 (26) <.001 
BG 85 (99) 39 (71) <.001 
Height 82 (95) 21 (38) <.001 
All of the above 63 (73) 5 (9) <.001 
CAPH, No. (%)Pediatric Inpatient Medical Settings, No. (%)P
Lipid panel 65 (76) 14 (26) <.001 
BG 85 (99) 39 (71) <.001 
Height 82 (95) 21 (38) <.001 
All of the above 63 (73) 5 (9) <.001 

P values were estimated by using χ2 test or Fisher’s exact test, depending on cell sizes.

LOS

LOS was significantly longer for those monitored for ADEs (P = .001), had a completed lipid panel (P = .005), or had a BG or HbA1c completed (P < .001) compared with those who did not. These and other factors significantly associated with LOS are explored in Table 7.

TABLE 7

Factors Significantly Associated With LOS

Patients on Scheduled Neuroleptic, No. (%)LOS in d, Median (Interquartile Range)P
ADE monitoringa <.001 
 Yes 84 (60) 9.7 (11.0)  
 No 57 (40) 3.3 (5.7)  
Lipid panel <.001 
 Yes 79 (56) 9.9 (10.4)  
 No 62 (44) 4.7 (7.1)  
BG .001 
 Yes 124 (88) 8.0 (12.0)  
 No 17 (12) 1.8 (1.4)  
Initiation before admission .002 
 Yes 107 (76) 6.5 (9.4)  
 No 34 (24) 9.5 (16.2)  
ADE experienceda <.001 
 Yes 36 (26) 12.1 (16.3)  
 No 105 (75) 6.3 (8.4)  
Unspecified indication <.001 
 Yes 32 (23) 2.5 (3.6)  
 No 109 (77) 8.8 (11.3)  
Patients on Scheduled Neuroleptic, No. (%)LOS in d, Median (Interquartile Range)P
ADE monitoringa <.001 
 Yes 84 (60) 9.7 (11.0)  
 No 57 (40) 3.3 (5.7)  
Lipid panel <.001 
 Yes 79 (56) 9.9 (10.4)  
 No 62 (44) 4.7 (7.1)  
BG .001 
 Yes 124 (88) 8.0 (12.0)  
 No 17 (12) 1.8 (1.4)  
Initiation before admission .002 
 Yes 107 (76) 6.5 (9.4)  
 No 34 (24) 9.5 (16.2)  
ADE experienceda <.001 
 Yes 36 (26) 12.1 (16.3)  
 No 105 (75) 6.3 (8.4)  
Unspecified indication <.001 
 Yes 32 (23) 2.5 (3.6)  
 No 109 (77) 8.8 (11.3)  

P values were estimated by using Mann-Whitney test.

a

Because of rounding, percentages may not total 100.

Indication for use was clear when a scheduled neuroleptic was started during hospitalization (n = 34), with mood disorder, not otherwise specified (NOS) being the most common indication for patients in the CAPH (n = 13) and delirium being the most common indication for patients in pediatric inpatient medical settings (n = 5). These indications were significantly associated with setting, with the mood disorder, NOS indication being used exclusively at the CAPH and the delirium indication being used only in pediatric inpatient medical settings.

Over 20% of patients experienced an ADE (n = 36), including sedation (n = 11), metabolic changes (n = 8), EPSs (n = 6), and QT prolongation (n = 2). Experiencing an ADE was significantly associated with ADE monitoring (P < .001).

Twenty-six patients were administered a PRN neuroleptic, each with an average of 3 administrations (range: 1–15). There was no significant difference between the percentage of patients administered a PRN neuroleptic when comparing CAPH and pediatric medical settings (14% vs 20%). Only 2 of 10 patients in pediatric medical settings were administered a PRN neuroleptic without psychiatric consultation, and all patients received at least 1 of child life services, psychology consultation, or psychiatry consultation.

In this study, we identified 152 patients who received 196 neuroleptic orders, the majority being scheduled second-generation neuroleptics, within inpatient medical and psychiatry settings over a 6-month period. Clearly specified indication was associated with significantly improved monitoring. Additional factors associated with improved monitoring included neuroleptic initiation during admission, involvement of psychiatry, longer LOS, and evidence of an ADE.

It is unclear whether neuroleptic monitoring may be less likely to be completed when the patient is admitted on a neuroleptic because of an assumption that the patient is stable on the neuroleptic with appropriate use and monitoring by an outpatient provider. The thought may be that this is an “outpatient” concern and should be deferred to the outpatient provider. However, in previous outpatient studies, this assumption has been proved to be false, and greater attention should be called to the continuation of neuroleptics prescribed before admission without discussion with the outpatient provider.7,26  This may also explain the association between poor neuroleptic monitoring and unspecified indication for use because all cases of unspecified indication (n = 32) were for patients admitted on a neuroleptic. Improved diagnostic evaluation, familiarity with best practice, and clear evidence-based indication for use is associated with enhanced monitoring practices. Furthermore, admission of a patient on a neuroleptic is an opportunity to consult psychiatry because the impact of neuroleptics and potential of ADEs can evolve and fluctuate over the course of the hospital stay, making ongoing consultation with psychiatry an important part of care for many patients on neuroleptics. The use of neuroleptics for somatic indications, such as sleep, may be related to the prescribing provider’s lack of familiarity with the risk incurred by neuroleptic use, given the high risk to benefit ratio incurred when using neuroleptics for these indications.

Completed monitoring parameters were significantly associated with a longer LOS. Because of the time sensitivity of hospital stays, it is possible that discussions of monitoring are missed or deferred for patients with shorter stays. The longer the LOS, the more likely the patient was to experience an ADE (Table 7). This may be related to longer exposure to neuroleptics as well as other organic clinical factors influencing neuroleptic usage. In addition, emergence of an ADE would likely result in a prolonged hospital stay to address that ADE. It is unclear from our study which of these plausible explanations had a greater influence on LOS in patients experiencing ADEs. However, the vast majority of hospitalizations are short stays (ie, <72 hours), and in this study, it is indicated that providers may be missing an important opportunity to address the use and monitoring of neuroleptics, which may be impacting the child’s metabolic, neurologic, and cardiovascular health.

Although 76% of patients did not experience ADEs, this percentage may be overstated. Because of the association between experiencing ADEs and ADE monitoring, patients who were not monitored may have experienced ADEs that were not noted or misattributed to another aspect of care. These ADEs were likely not clinically significant to the patient’s acute hospitalization but may have other downstream effects after discharge, particularly if appropriate follow-up and documentation were not provided. Involvement of a psychiatrist, whether through consultation or admission to the CAPH, had a significant impact on improving adherence to monitoring practices. This is especially apparent when considering comorbidities. This was particularly noticeable with EPS monitoring, in which the monitoring or discussion of the potential for EPSs in the medical record was minimal in patients not psychiatrically admitted or not seen by a psychiatric consultant. This is alarming given the significant morbidity associated with EPSs, the importance of addressing EPSs early (including consideration of discontinuation of the neuroleptic), and the potential longer-term effects of EPSs if not addressed.

There were 32 patients whose baseline comorbidities were unknown because their height and weight were not recorded. None of these patients were admitted to the CAPH. Again, the lack of these data is concerning given the considerable metabolic effects associated with neuroleptic use and evidence of poor monitoring practices, particularly in regard to lipid collection. Monitoring may be more likely to be completed while at CAPH because these providers are trained and informed on standard monitoring practices for metabolic and other ADEs related to neuroleptic use.

The choice of a neuroleptic should account for the particular needs of a given patient, including administration route, time to effect, potential ADEs, illness factors, patient’s past experience with neuroleptics, and the specific symptoms being targeted. Neuroleptics also have differential effects on neurotransmitters, with some neuroleptics being relatively more dopaminergic, serotonergic, histaminergic, or anticholinergic.27  Neuroleptic side effects include sedation, metabolic effects, hypotension, tachycardia, QTc prolongation, arrhythmia, and EPSs.27  Additionally, neuroleptics can lower the seizure threshold, impact liver transaminases, or (rarely) precipitate neuroleptic malignant syndrome.27  Cardiac effects are a common concern, although cardiac effects are less common in youth than adults.28  QTc prolongation with neuroleptics is dose related and less common at lower doses.29  Consideration of impacted neurotransmitters and the relative potential of given ADEs should always be considered when choosing a neuroleptic, its use, and its monitoring. For example, the increasing use of olanzapine in antiemetic management of youth receiving high-risk chemotherapy has a significant impact on care. Olanzapine has significant potential for metabolic effects in youth with regular use, including placing patients at higher risk of weight gain, dyslipidemia, and diabetes.15,27 

PRN use was more difficult to characterize because of the smaller sample size. Ziprasidone was the most commonly used PRN neuroleptic, likely because of ziprasidone being included in the CAPH admission order set as well as its common use within the psychiatry emergency service. This practice was adopted for pediatric aggression management in the CAPH at our institution because of ziprasidone’s reduced risk of metabolic ADEs and availability as an intramuscular formulation.30  However, the reduced risk of metabolic ADEs must be weighed against the potential for QT prolongation compared with other agents, such as intramuscular olanzapine.31  In pediatric medical settings, ziprasidone was used much less often as a PRN neuroleptic. It is unclear if this is related to the lack of prescriber familiarity with ziprasidone and its side effects being less favorable in the physically ill child, cultural changes related to psychiatric consultation, input from pharmacy, or a combination of these factors.

Although insight into neuroleptic prescribing and monitoring within inpatient pediatrics was provided in this study, there were some inherent limitations. First, small sample size posed a challenge in reporting meaningful data on items such as PRN use, as highlighted in the previous paragraph. Second, our population had access to an embedded CAPH as well as psychiatry, psychology, and child life consult services. These services vary in structure and availability at similar institutions, limiting generalizability. Third, there was a finite number of data points collected, limiting the following considerations. In this study, alternative practices to neuroleptic use was not accounted for, such as other psychopharmacologic, opioid, sedation, and restraint use. Other concomitant medications were also not considered, which could alter prescribing, monitoring, patient response, or ADE predisposition. In this study, complexity and severity of medical comorbidities during the hospital stay were also not accounted for, which may impact LOS and consideration of neuroleptic use. Only including patients’ first hospitalization during the study period could potentially bias results because it is possible that patients readmitted to psychiatry may be more likely to receive neuroleptics. Furthermore, trying to compare indicated use to patient-specific physical, emotional, and behavioral health factors was challenging because of a lack of background information about the patient. It would have been helpful to have readily available mental health treatment records, past response to neuroleptics, family history of metabolic and/or psychiatric illness, and insurance status. In addition to these aforementioned considerations, authors of future studies could explore more heterogenous populations and gather more data on PRN use.

Within the study design, poor documentation made it difficult to identify intent of parameter assessment or indication for use. In the study, previous mental health provider prescribing and monitoring of neuroleptics by non–health system mental health providers that were not explicitly discussed by care team documentation were also not accounted for. All these limitations are inherent in chart review, in which quality is reliant on documentation. Furthermore, medical provider documentation of ADEs may have been falsely high or low, depending on the over or under attribution of a given neuroleptic to ADE development. Despite these limitations in study design, poor documentation or misattribution of ADEs are still concerning findings that would not detract from the overall finding of this study regarding the use and monitoring of neuroleptics, particularly given the frequency of patient handoffs and transitions of care that occur in the inpatient academic medical and psychiatric settings.

As areas for future study, it would be informative to address low monitoring rates from an institutional level and worthwhile to implement a stewardship program focusing on the proper use of these medications through development and use of a multidisciplinary clinical practice guideline to standardize practices relating to neuroleptic use and monitoring. Other tools, such as an automated messaging program, stopgap measures to require providers to complete certain tasks in the prescription and monitoring of neuroleptics, dissemination of educational resources and tools, or the use of monitoring checklists with neuroleptic use, can prove fruitful in improving monitoring of neuroleptics.32 

Patients treated with a neuroleptic medication with input from psychiatry were more likely to have monitoring completed. This is exemplified in Fig 1, in which we highlight the value of integrated psychiatric consultation within medical settings in a case within our study cohort in which psychiatric consultation had significant impacts on addressing diagnosis, management, and drug monitoring.

FIGURE 1

Impact of psychiatric consultation on inpatient medical practice in neuroleptic use and monitoring. MD, doctor of medicine.

FIGURE 1

Impact of psychiatric consultation on inpatient medical practice in neuroleptic use and monitoring. MD, doctor of medicine.

In this study, we provide insight into current prescribing and monitoring practices within a large academic pediatric hospital. The use of neuroleptics in youth is increasing while metabolic monitoring remains suboptimal. Adequate monitoring is essential to minimize the potential risks associated with long-term use in this population. In this study, it is suggested that pediatric patients admitted for nonpsychiatric reasons or admitted on a preexisting neuroleptic regimen are more likely to be inadequately monitored for metabolic ADEs. Descriptions of use in a single institution may better inform other institutions as to patterns of prescribing and monitoring in neuroleptic use, minimizing risk to youth in these settings. These findings can be used to inform future targets for training, education, collaboration between pharmacy and psychiatry, enhanced nonpharmacologic supports, and system-based efforts to enhance the use and monitoring of neuroleptics in pediatrics. Longitudinal data used to evaluate metabolic monitoring and indications of use for youth treated with neuroleptics across the continuum of care are needed.

Dr Le finalized the design of the data collection instruments, participated in data collection, conducted analyses, and drafted the initial manuscript; Dr Andreasen designed the initial data collection instruments, participated in data collection, and drafted the initial manuscript; Dr Malas conceptualized the study, provided feedback on data to be collected, and provided information about current practices for the manuscript; Dr Bostwick conceptualized and designed the study, provided feedback on data to be collected, and participated in data collection; and all authors reviewed and revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.

FUNDING: Supported by the Michigan Institute for Clinical and Health Research grant (Clinical and Translational Science Award UL1TR000433).

1
Olfson
M
,
Blanco
C
,
Liu
L
,
Moreno
C
,
Laje
G
.
National trends in the outpatient treatment of children and adolescents with antipsychotic drugs
.
Arch Gen Psychiatry
.
2006
;
63
(
6
):
679
685
2
Olfson
M
,
Blanco
C
,
Wang
S
,
Laje
G
,
Correll
CU
.
National trends in the mental health care of children, adolescents, and adults by office-based physicians
.
JAMA Psychiatry
.
2014
;
71
(
1
):
81
90
3
Olfson
M
,
King
M
,
Schoenbaum
M
.
Treatment of young people with antipsychotic medications in the United States
.
JAMA Psychiatry
.
2015
;
72
(
9
):
867
874
4
US National Library of Medicine
.
Daily Med
.
Available at: https://dailymed.nlm.nih.gov/dailymed/. Accessed June 29, 2016
5
Truven Health Analytics
.
MICROMEDEX
.
Available at: www.micromedexsolutions.com. Accessed June 29, 2016
6
Alexander
GC
,
Gallagher
SA
,
Mascola
A
,
Moloney
RM
,
Stafford
RS
.
Increasing off-label use of antipsychotic medications in the United States, 1995-2008
.
Pharmacoepidemiol Drug Saf
.
2011
;
20
(
2
):
177
184
7
Rettew
DC
,
Greenblatt
J
,
Kamon
J
, et al
.
Antipsychotic medication prescribing in children enrolled in Medicaid
.
Pediatrics
.
2015
;
135
(
4
):
658
665
8
De Hert
M
,
Dobbelaere
M
,
Sheridan
EM
,
Cohen
D
,
Correll
CU
.
Metabolic and endocrine adverse effects of second-generation antipsychotics in children and adolescents: A systematic review of randomized, placebo controlled trials and guidelines for clinical practice
.
Eur Psychiatry
.
2011
;
26
(
3
):
144
158
9
Seida
JC
,
Schouten
JR
,
Boylan
K
, et al
.
Antipsychotics for children and young adults: a comparative effectiveness review
.
Pediatrics
.
2012
;
129
(
3
). Available at: www.pediatrics.org/cgi/content/full/129/3/e771
10
Hesketh
PJ
,
Kris
MG
,
Basch
E
, et al
.
Antiemetics: American society of clinical oncology clinical practice guideline update
.
J Clin Oncol
.
2017
;
35
(
28
):
3240
3261
11
Slomski
A
.
Chronic mental health issues in children now loom larger than physical problems
.
JAMA
.
2012
;
308
(
3
):
223
225
12
Verdoux
H
,
Tournier
M
,
Bégaud
B
.
Antipsychotic prescribing trends: a review of pharmaco-epidemiological studies
.
Acta Psychiatr Scand
.
2010
;
121
(
1
):
4
10
13
Correll
CU
,
Manu
P
,
Olshanskiy
V
,
Napolitano
B
,
Kane
JM
,
Malhotra
AK
.
Cardiometabolic risk of second-generation antipsychotic medications during first-time use in children and adolescents [published correction appears in JAMA. 2009;302(21):2322]
.
JAMA
.
2009
;
302
(
16
):
1765
1773
14
McIntyre
RS
,
Jerrell
JM
.
Metabolic and cardiovascular adverse events associated with antipsychotic treatment in children and adolescents
.
Arch Pediatr Adolesc Med
.
2008
;
162
(
10
):
929
935
15
Carbon
M
,
Kapoor
S
,
Sheridan
E
, et al
.
Neuromotor adverse effects in 342 youth during 12 weeks of naturalistic treatment with 5 second-generation antipsychotics
.
J Am Acad Child Adolesc Psychiatry
.
2015
;
54
(
9
):
718
727.e3
16
Dietz
WH
.
Health consequences of obesity in youth: childhood predictors of adult disease
.
Pediatrics
.
1998
;
101
(
3, pt 2
):
518
525
17
Mayne
SL
,
Ross
ME
,
Song
L
, et al
.
Variations in mental health diagnosis and prescribing across pediatric primary care practices
.
Pediatrics
.
2016
;
137
(
5
):
e20152974
18
Goodwin
R
,
Gould
MS
,
Blanco
C
,
Olfson
M
.
Prescription of psychotropic medications to youths in office-based practice
.
Psychiatr Serv
.
2001
;
52
(
8
):
1081
1087
19
Cummings
MR
,
Miller
BD
.
Pharmacologic management of behavioral instability in medically ill pediatric patients
.
Curr Opin Pediatr
.
2004
;
16
(
5
):
516
522
20
Findlay
S
,
Pinzon
J
,
Goldberg
E
, et al
.
Issues of care for hospitalized youth
.
Paediatr Child Health
.
2008
;
13
(
1
):
61
68
21
Deshmukh
P
,
Kulkarni
G
,
Barzman
D
.
Recommendations for pharmacological management of inpatient aggression in children and adolescents
.
Psychiatry (Edgmont)
.
2010
;
7
(
2
):
32
40
22
Honey
BL
,
Ramos
L
,
Brahm
NC
.
Evaluation of monitoring for metabolic effects in children treated with second generation antipsychotics in a pediatric clinic
.
J Pediatr Pharmacol Ther
.
2013
;
18
(
4
):
292
297
23
Thompson
W
,
Quay
TAW
,
Rojas-Fernandez
C
,
Farrell
B
,
Bjerre
LM
.
Atypical antipsychotics for insomnia: a systematic review
.
Sleep Med
.
2016
;
22
:
13
17
24
Dupuis
LL
,
Nathan
PC
.
Optimizing emetic control in children receiving antineoplastic therapy: beyond the guidelines
.
Paediatr Drugs
.
2010
;
12
(
1
):
51
61
25
Harris
PA
,
Taylor
R
,
Thielke
R
,
Payne
J
,
Gonzalez
N
,
Conde
JG
.
Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support
.
J Biomed Inform
.
2009
;
42
(
2
):
377
381
26
Edelsohn
GA
,
Parthasarathy
M
,
Terhorst
L
,
Karpov
IO
,
Schuster
J
.
Measurement of metabolic monitoring in youth and adult Medicaid recipients prescribed antipsychotics
.
J Manag Care Spec Pharm
.
2015
;
21
(
9
):
769
777, 777a–777cc
27
Turkel
SB
,
Jacobson
J
,
Munzig
E
,
Tavaré
CJ
.
Atypical antipsychotic medications to control symptoms of delirium in children and adolescents
.
J Child Adolesc Psychopharmacol
.
2012
;
22
(
2
):
126
130
28
Ben Amor
L
.
Antipsychotics in pediatric and adolescent patients: a review of comparative safety data
.
J Affect Disord
.
2012
;
138
(
suppl
):
S22
S30
29
Haddad
PM
,
Anderson
IM
.
Antipsychotic-related QTc prolongation, torsade de pointes and sudden death
.
Drugs
.
2002
;
62
(
11
):
1649
1671
30
Brook
S
.
Intramuscular ziprasidone: moving beyond the conventional in the treatment of acute agitation in schizophrenia
.
J Clin Psychiatry
.
2003
;
64
(
suppl 19
):
13
18
31
Staller
JA
.
Intramuscular ziprasidone in youth: a retrospective chart review
.
J Child Adolesc Psychopharmacol
.
2004
;
14
(
4
):
590
592
32
Lai
CL
,
Chan
HY
,
Pan
YJ
,
Chen
CH
.
The effectiveness of a computer reminder system for laboratory monitoring of metabolic syndrome in schizophrenic outpatients using second-generation antipsychotics
.
Pharmacopsychiatry
.
2015
;
48
(
1
):
25
29

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data