Association Between Abnormal Metabolic Parameters and Receiving Subsequent Interventions in Children and Adolescents Initiating Second-Generation Antipsychotics

Abstract

Objectives:

This study aimed to examine the association between abnormal readings of metabolic parameters detected during second-generation antipsychotic (SGA) treatment and the likelihood of receiving subsequent adverse drug event interventions.

Methods:

This was a nested case–control study conducted on patients 1–17 years of age with at least two prescriptions of SGAs between January 2010 and January 2019 using TriNetX EMR data. Following an incident density sampling procedure, patients who received the SGA metabolic adverse event intervention (mAEI) (case) were matched with three nonrecipients (controls). The abnormal readings of metabolic parameters within 30 days before the initiation of mAIEs were further identified. These metabolic parameters include body mass index (BMI) and laboratory parameters such as cholesterol, high-density lipoprotein, low-density lipoprotein, triglycerides, blood glucose, HbA1c, thyroid hormones, liver enzymes, and prolactin. The association of abnormal metabolic parameters with subsequent mAEIs was assessed using a conditional logistic regression model, after adjusting for demographic and other clinical risk factors.

Results:

One thousand eight hundred eighty-four children and adolescents met the inclusion criteria and were prescribed SGA mAEIs. The most common types of mAEIs prescribed were weight management pharmacotherapy (40.6%), switching from a high or medium metabolic risk profile SGA to a low-risk one (30.9%), nonpharmacological treatment (25.4%), and switching from SGA polytherapy to monotherapy (11.7%). The conditional logistic regression analysis on matched mAEI recipients and nonrecipients showed that patients with an abnormal BMI had 43% higher odds of receiving mAEI (odds ratio [95% confidence interval]: 1.43 [1.13–1.79]). However, the presence of an abnormal laboratory reading was not associated with the initiation of mAEIs.

Conclusions:

The prescribing of mAEIs were associated with the presence of obesity, but not with abnormal readings of other metabolic parameters, suggesting that additional data are needed to clarify the long-term implication of SGA metabolic adverse events other than weight gain and to inform the appropriate timing for interventions.

Background

A diverse array of side effects has been associated with second-generation antipsychotics (SGAs) (Carli et al., 2021; Libowitz and Nurmi, 2021; McIntyre and Jerrell, 2008; Pillinger et al., 2020). Other than antipsychotics-induced weight gain (AIWG) (Dayabandara et al., 2017), SGAs can disrupt glucose metabolism, induce dyslipidemia (Chaggar et al., 2011; Grajales et al., 2019), and cause asymptomatic elevation of liver enzymes (Atasoy et al., 2007), thyroid function disturbance (Vedal et al., 2018), and prolactin elevation (AACAP, 2011).

Due to concerns about SGA-associated side effects, the American Academy of Child and Adolescent Psychiatry (AACAP) practice parameter for the use of atypical antipsychotic medications in children and adolescents and the Canadian Alliance for Monitoring Effectiveness and Safety of Antipsychotics in Children (CAMESA) recommend baseline and periodic monitoring of body mass index (BMI), blood glucose, and lipid profile (high-density lipoprotein [HDL], low-density lipoprotein [LDL], total cholesterol, and triglycerides) (AACAP, 2011). In addition, the CAMSEA guidelines also recommend monitoring for prolactin (baseline, annually, or with clinical symptoms), thyroid-stimulating hormones (TSH; baseline and annually only for quetiapine) and liver enzymes (baseline, semiannually, and annually) (Pringsheim et al., 2011).

In addition to the monitoring recommendation, CAMSEA also recommends management strategies for SGA-associated side effects (Ho et al., 2011). These interventions include nonpharmacological interventions such as behavioral counseling and lifestyle modifications, and pharmacological interventions that include adding a weight neutralizing agent, switching to a lower metabolic risk profile SGA, avoiding multitherapy, and tapering off SGA dose if feasible (Dayabhandara et al., 2017; Ho et al., 2011).

However, the guidelines do not provide a clear guidance on when these SGA adverse effect mitigating interventions should be initiated. A study by Morrato et al. (2009) found that the presence of abnormal metabolic parameters in adults on SGAs were not consistently associated with the selection of an SGA with lower metabolic risk. The association of the presence of abnormal metabolic parameters and the SGA metabolic adverse event interventions (mAEIs) in children and adolescents has not been examined.

The objective of our study was to characterize the mAEIs implemented during SGA treatment in children and adolescents and to examine the association of mAEIs (outcome) with the presence of abnormal metabolic parameters (exposure) using a national electronic medical records (EMR) database.

Methods

Data source

This study was conducted using the TriNetX EMR database from January 1, 2009, till January 31, 2019. This database includes 60 million patients from 50 large healthcare organizations in the United States. The health care organizations include a mixture of hospitals, primary care, and specialist providers and contribute data from insured and uninsured patients alike. The database records patient information such as detailed patient demographics, diagnoses, procedural information, vital signs, and laboratory test results. Medication list entries include their RxNorm codes and start dates. The database may include multiple data points, making it a rich source for longitudinal analyses. The database is de-identified in accordance with the Health Insurance Portability and Accountability Act and has been used widely in published literature (Firl et al., 2021; Freedman et al., 2021; Harrison et al., 2021 ; Harrison et al., 2020; Mukherjee et al., 2021; Sultan et al., 2021; Taquet et al., 2021).

Study population

Individuals who were 1–17 years of age and had at least two prescription orders of SGAs were identified from the TriNetX EMR database. The SGAs included were aripiprazole, asenapine, brexpiprazole, cariprazine, clozapine, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine, risperidone, and ziprasidone. To minimize the potential confounding effect of chronic SGA treatment, we used a new-user design. We defined the index SGA as the first SGA prescription patients received after a 6-month period without an SGA prescription order. Patients were excluded if they had <30-day SGA exposure or if they received a diagnosis of dyslipidemia, diabetes, or eating disorder at SGA initiation.

Because many SGA recipients were obese before SGA initiation, abnormal BMI detected during SGA treatment may simply be the continuation of baseline weight status. To understand how the weight status change could affect the downstream regimen adjustment, a subcohort, including patients who were not affected by obesity (n = 607) before SGA initiation, was also constructed. A child was defined as “not affected by obesity” if their body mass index-for-age (or BMI-for-age) percentile is <95%.

Study design

As presented in Figure 1, a nested case–control design was employed to assess the association of the presence of abnormal metabolic monitoring results with the receipt of mAEIs after controlling for other clinical and demographic factors.

FIG. 1.

Nested case–control (1:3 match) study design.

Cases

The mAEI recipients were defined as patients who received the first mAEI beyond 30 days after SGA initiation. The mAEIs included both nonpharmacotherapy treatments and pharmacotherapies. The nonpharmacotherapy comprised dietary interventions, obesity counseling, and psychotherapy. Pharmacological interventions consisted of weight management pharmacotherapy and SGA regimen adjustment, including switching from a high or medium metabolic profile SGA to a low metabolic risk profile SGA or switching from SGA polytherapy to monotherapy. SGAs such as aripiprazole, asenapine, lurasidone, and ziprasidone are considered having low metabolic risk (Carli et al., 2021; Pillinger et al., 2020). Thus, our definition of switching included switching from any other SGA to these four low-risk SGAs. Detailed diagnosis, procedural, and medication RX-Norm codes used to identify mAEIs are described in Supplementary Appendix Table SA1.

Case–control matching

The date of mAEI initiation was defined as the “event date.” Each mAEI recipient was matched at a 1:3 ratio with individuals who had not received an intervention by the case event date, using an incidence density sampling procedure (Grandits, 2010; Richarson, 2004). The incidence sampling procedure allows cases to serve as potential controls for other cases and to be matched based on their SGA follow-up duration, thus eliminating potential immortal time bias. For each matched case–control set, the event date of the case was assigned to the controls.

Observation period

Factors associated with the initiation of mAEIs were observed within a 30-day window before the event date.

Main risk factor

The main risk factor of interest was the presence of abnormal metabolic parameters within the 30-day observation period. According to the AACAP and CAMESA guidelines, the metabolic parameters considered were BMI and laboratory measures, including blood glucose, HbA1c, HDL cholesterol, LDL cholesterol, total cholesterol, triglycerides, liver enzymes, prolactin, and thyroid hormones. BMI was considered an independent measure, and all laboratory parameters were grouped under the measure “Labs.” Both measures (BMI and Labs) were categorized as normal, abnormal, or absent. Abnormal BMI was defined as ≥95 percentile of age- and gender-adjusted BMI values according to the CDC weight chart. Laboratory results were flagged as abnormal or normal according to standard pediatric guidelines (ADA, 2022; Mayo Clinic Laboratories, 2022; NHLBI, 2012; Whitelock et al., 2010). Detailed threshold cutoff values for normal and abnormal BMI and Labs are presented in Supplementary Appendix Tables SA2 and SA3.

Other potential risk factors

Other potential risk factors identified during the 30-day observation period were age, gender, race, SGAs used, use of multiple SGAs, mental disorder diagnoses including attention-deficit/hyperactivity disorder (ADHD), mood and depressive disorders, anxiety disorders, psychotic disorders, and schizophrenia, other mental health disorders, and comedications including antidepressants, psychostimulants, and α-2 agonists, mood stabilizers and anticonvulsants, benzodiazepines and other anxiolytics, and other psychotropic medications.

These factors were selected because literature has reported that gender, race, age group, and preexisting conditions are risk factors associated with SGA use and some of these risk factors can contribute to changes in treatment strategies such as antipsychotic switching or discontinuation (Fulone et al., 2021; Stroup and Gray, 2018). Age and gender could also modify the risk of some adverse drug events (ADEs) (e.g., QT prolongation) associated with SGAs (He et al., 2022). Mental health comorbidities may be associated with the degree of weight gain associated with SGAs (Doane et al., 2022). It has also been shown that concurrent use of SGAs with other psychotropic medication drug classes is associated with a higher risk for SGA-associated metabolic adverse effects (Burcu et al., 2017; Zhang et al., 2021).

Statistical analysis

Paired t-test was conducted to compare continuous variables and Cochran-Mantel-Hansel test was conducted to compare categorical variables between matched cases and controls. A conditional logistic regression model was fitted to determine the association between the presence of abnormal BMI or laboratory values and the initiation of mAEIs after controlling for other potential risk factors.

The study was approved by the institution review board of the University of Houston. All analysis was done using SAS 9.4 (NC).

Results

Incidence and subtypes of mAEIs

The cohort identification schematic is presented in Figure 2. There were 1884 patients who met inclusion criteria and received an mAEI at least 30 days after SGA initiation, of which 607 (32%) had a baseline BMI <95 percentile. The median time to mAEIs from SGA initiation was 64 (interquartile range: 44–98) days. The cohort characteristics are presented in the Supplementary Appendix Table SA4.

FIG. 2.

Schematic diagram for the identification of study cohort. *mAEI, SGA metabolic adverse event intervention. SGA, second-generation antipsychotic.

Table 1 presents the types of SGA mAEIs prescribed. In the full cohort, 40.55% of those initiated mAEIs were prescribed weight management pharmacotherapy, followed by switching from a high or medium metabolic profile SGA to a low metabolic profile (as described earlier) SGA (30.89%), adding nonpharmacotherapy treatment (25.37%), and switching from SGA polytherapy to monotherapy (11.73%). Among those prescribed weight management pharmacotherapy, 74% received adjuvant metformin. While in the subcohort of patients who were not affected by obesity, nonpharmacotherapy treatments (32.12%) were the second most common choice of MAEIs, following weight management pharmacotherapy (36.57%). The third most commonly prescribed mAEI was switching from high- or medium-risk profile SGAs to low-risk profile SGAs (27.18%).

Table 1.

Subtypes of Second-Generation Antipsychotic Metabolic Adverse Drug Event Interventions (Second-Generation Antipsychotic Metabolic Adverse Event Interventions)

Types of SGA mAEIs	Full cohort (N = 1884), N (%)	BMI <95 percentile subcohort (N = 607), N (%)
Nonpharmacotherapy treatment	478 (25.37)	195 (32.12)
Medication nutrition therapy	6 (0.31)	3 (0.49)
Obesity counselling	392 (20.80)	175 (28.83)
Psychotherapy	45 (2.38)	11 (1.81)
Weight management counseling diagnosis	35 (1.85)	6 (0.99)
Weight management pharmacotherapy	764 (40.55)	222 (36.57)
Amantadine	1 (0.05)	1 (0.16)
Metformin	566 (30.04)	166 (27.34)
Phentermine	2 (0.10)	NA
Topiramate	167 (8.86)	43 (7.08)
Zonisamide	28 (1.48)	12 (1.98)
SGA polytherapy at initiation to monotherapy (first switch)	221 (11.73)	69 (11.36)
SGA high or medium metabolic profile to low metabolic profile switching (first switch)	582 (30.89)	165 (27.18)
Switched to aripiprazole	515 (27.33)	145 (23.89)
Switched to asenapine	1 (0.05)	NA
Switched to lurasidone	14 (0.74)	4 (0.65)
Switched to ziprasidone	52 (2.76)	16 (2.63)

Obesity counseling: obesity counseling identified through specific procedural (CPT) codes.

Weight management counseling diagnosis: dietary management and counseling identified through ICD diagnosis code Z71.3 (ICD-10 code).

BMI, body mass index; mAEI, metabolic adverse event intervention; SGA, second-generation antipsychotic.

Incidence and subtypes of SGA ADE monitoring

There were 1882 out of 1884 mAEI patients in the full cohort and 606 out of 607 mAEI patients in the subcohort were successfully matched with controls at a 1:3 ratio. Table 2 presents the monitoring frequency of SGA metabolic parameters in the matched event and control groups. During the 30-day period before the event date, less than half of the patients (41.92%) in the full cohort had BMI monitoring and only a fraction of patients (17.43%) had laboratory monitoring for SGA-associated ADEs. The most monitored laboratory parameters were liver enzymes (8.98%), followed by blood glucose (4.25%) and cholesterol (3.40%). Other metabolic parameters such as HbA1c, TSH, and prolactin were only monitored in a few patients.

Table 2.

Body Mass Index and Lab Monitoring Frequency

Monitoring parameter	Full cohort		BMI <95 percentile subcohort
	Monitoring frequency, N (%)
	Patients with an mAEI (N = 1882)	Matched patients without a mAEI (N = 5644)	SGA MAEIs (N = 606)	Matched patients without a mAEI (N = 1816)
BMI	789 (41.92)	1196 (21.19)	438 (72.28)	1257 (69.22)
Labs (total)	328 (17.43)	725 (12.85)	83 (13.70)	278 (15.31)
Cholesterol	64 (3.40)	124 (2.20)	19 (3.14)	68 (3.69)
HDL	NA	1 (0.12)	NA	NA
LDL	NA	1 (0.12)	NA	NA
Triglycerides	NA	NA	NA	NA
Blood glucose	80 (4.25)	183 (3.24)	21 (3.47)	63 (3.25)
HbA1c	5 (0.27)	3 (0.05)	1 (0.17)	1 (0.06)
TSH	8 (0.43)	11 (0.19)	3 (0.50)	11 (0.61)
Liver enzymes	169 (8.98)	400 (7.09)	38 (6.27)	132 (7.27)
Prolactin	2 (0.11)	2 (0.04)	1 (0.17)	3 (0.11)

BMI, body mass index; HDL, high-density lipoprotein; LDL, low-density lipoprotein; mAEI, metabolic adverse event intervention; SGA, second-generation antipsychotic; TSH, thyroid-stimulating hormones.

In the nonobese subcohort (BMI <95 percentile), almost two-thirds of the patients had BMI monitoring (72.28%) and only 13.70% of patients had any kind of laboratory monitoring. Among the laboratory parameters monitored, the most frequent were alanine aminotransferase (ALT) liver enzymes (6.27%), followed by blood glucose (3.47%) and cholesterol (3.14%).

Characteristics of matched cases and controls

The distribution for the main and other potential risk factors associated with mAIEs for the full cohort is presented in Table 3. The mAEI recipients had significantly higher BMI and laboratory monitoring rates compared with the matched controls (BMI: 41.92% vs. 21.19%, p < 0.001 and Labs: 17.43% vs. 12.9%, p < 0.001). The proportion of patients with abnormal BMI and laboratory reading was also significantly higher among the mAEI group (BMI_abnormal: 10.52% vs. 4.04%; p < 0.001 and Lab_abnormal: 5.79% vs. 4.64%; p < 0.001) than that in the controls. Of children identified with abnormal Labs, 89.90% had elevated liver enzymes. The mean dysregulated Lab value for ALT liver enzyme was 72 U/L, which is 16–26 U/L points (34.7%–46.43%) above the threshold value. The mAIEs recipients also had higher usage of multiple SGAs (12.75% vs. 5.01%; p < 0.001), higher usage of psychotropic comedications such as antidepressants (30.02% vs. 20.84%; p < 0.001), psychostimulants and α-2 agonists (33.95% vs. 26.24%; p < 0.001), mood stabilizers and anticonvulsants (11.42% vs. 7.32%; p < 0.001), benzodiazepines, and other anxiolytics (8.71% vs. 6.25%; p = 0.003), and had more psychiatric comorbidities, including ADHD (36.93% vs. 23.33%; p < 0.001), mood and depressive orders (42.51% vs. 27.02%; p < 0.001), anxiety disorders (25.29% vs. 17.24%; p < 0.001), psychosis, and schizophrenia (6.32% vs. 3.38%; p < 0.001).

Table 3.

Distribution of Risk Factors Between Matched Metabolic Adverse Event Intervention Recipients and Nonrecipients in the Full Cohort

Covariates	Full cohort
	SGA mAEIs (N = 1882)	No SGA mAEIs (N = 5644)	CMH (p)
	N (%)	N (%)	CMH (p)
BMI only	789 (41.92)	1196 (21.19)	<0.001
Normal	591 (31.40)	968 (17.15)
Abnormal	198 (10.52)	228 (4.04)
Lab only	328 (17.43)	725 (12.85)	<0.0001
Normal	219 (11.64)	466 (8.26)
Abnormal	109 (5.79)	262 (4.64)
Cholesterol	3 (0.16)	2 (0.03)
Blood glucose	6 (0.32)	15 (0.26)
Liver enzymes (ALT and AST)	98 (5.20)	242 (4.29)
BMI and lab	307 (16.31)	490 (26.04)	<0.0001
Either BMI or lab abnormal	293 (15.57)	479 (8.49)
Both BMI and lab abnormal	14 (0.74)	11 (0.19)
Age, years			0.0296
13–17	944 (50.16)	2779 (49.24)
6–12	861 (45.75)	2545 (45.09)
1–5	77 (4.09)	320 (5.67)
Gender			0.7773
Male	1149 (61.05)	3947 (61.96)
Female	729 (38.74)	2136 (37.85)
Unreported	4 (0.21)	11 (0.19)
Racial group			0.9366
White/Caucasian	1364 (72.48)	4048 (71.72)
Black/African American	261 (13.87)	803 (14.23)
Asian/PIs/NAs	30 (1.59)	90 (1.59)
Unknown/not reported	227 (12.06)	703 (12.46)
SGA usage			<0.001
Risperidone	707 (37.57)	2468 (43.73)
Aripiprazole	630 (33.48)	1560 (27.64)
Quetiapine	270 (14.35)	888 (15.73)
Olanzapine	178 (9.46)	522 (9.25)
Ziprasidone	57 (3.03)	132 (2.34)
Others	40 (2.13)	74 (1.31)
Multiple SGA usage	240 (12.75)	283 (5.01)	<0.0001
Mental disorder diagnoses
ADHD	695 (36.93)	1317 (23.33)	<0.0001
Mood disorders	800 (42.51)	1525 (27.02)	<0.0001
Anxiety disorders	476 (25.29)	973 (17.24)	<0.0001
Psychosis and schizophrenia	119 (6.32)	191 (3.38)	<0.0001
Other mental health disorders	772 (41.02)	1493 (26.45)	<0.0001
Other psychotropic medications
Antidepressants	565 (30.02)	1176 (20.84)	<0.0001
Psychostimulants and α-2 agonist	639 (33.95)	1481 (26.24)	<0.0001
Anticonvulsants	215 (11.42)	413 (7.32)	<0.0001
Anxiolytics	164 (8.71)	353 (6.25)	0.0003
Other psychotropic medications	64 (3.40)	157 (2.78)	0.1685

ADHD, attention-deficit/hyperactivity disorder; ALT, alanine aminotransferase; AST, aspartate transaminase; BMI, body mass index; CMH, Cochran-Mantel-Hansel; mAEI, metabolic adverse event intervention; PIs, Pacific Islanders; NAs, native Americans; SGA, second-generation antipsychotic.

The distribution for the main and other potential risk factors associated with mAIEs for the subcohort is presented in Table 4 Unlike the full cohort, statistically significant differences between mAEI recipients and matched controls were not observed in BMI monitoring rate. The proportion of patients who had abnormal BMI and Labs was also not statistically different between mAEI group (BMI_abnormal: 6.11% vs. 4.46%; p = 0.1301 and Labs_abnormal 3.96% vs. 3.91%; p = 0.5260) and matched non-mAEI group.

Table 4.

Distribution of Risk Factors Between Matched Metabolic Adverse Event Intervention Recipients and Nonrecipients in the Subcohort of Patients Not Affected by Obesity at Second-Generation Antipsychotic Initiation

Covariates	BMI <95 percentile subcohort
	SGA mAEIs (N = 606)	No SGA mAEIs (N = 1816)	CMH (p)
	N (%)	N (%)
BMI only			0.1301
Normal	401 (66.17)	1176 (64.76)
Abnormal	37 (6.11)	81 (4.46)
Total	438 (72.28)	1257 (69.22)
Lab only			0.5260
Normal	59 (9.74)	207 (11.40)
Abnormal	24 (3.96)	71 (3.91)
Cholesterol	1 (0.16)	1 (0.05)
Blood glucose	1 (0.16)	4 (0.22)
Liver enzymes (ALT and AST)	21 (3.46)	65 (3.58)
Prolactin	1 (0.16)	1 (0.05)
Total	83 (13.70)	278 (15.31)
BMI and lab			0.2263
Either BMI or lab abnormal	58 (9.57)	147 (8.09)
Both BMI and lab abnormal	3 (0.50)	5 (0.08)
Total	61 (10.07)	152 (8.37)
Age, years			0.6279
13–17	298 (49.17)	903 (49.72)
6–12	281 (46.37)	816 (44.93)
1–5	27 (4.46)	97 (5.34)
Gender			0.3879
Male	384 (63.37)	1115 (61.40)
Female	222 (36.63)	701 (38.60
Racial group			0.7095
White/Caucasian	424 (69.97)	1267 (69.77)
Black/African American	87 (14.36)	280 (15.42)
Asian/PIs/NAs	10 (1.65)	38 (2.09)
Unknown/Not reported	85 (14.03)	231 (12.72)
SGA usage			<0.001
Risperidone	219 (36.13)	816 (44.93)
Aripiprazole	208 (34.32)	444 (24.44)
Quetiapine	99 (16.33)	330 (18.17)
Olanzapine	43 (7.09)	152 (8.37)
Ziprasidone	24 (3.96)	39 (2.15)
Others	13 (2.14)	35 (1.93)
Multiple SGA usage	83 (13.70)	144 (7.93)	<0.0001
Comorbid conditions
ADHD	314 (51.82)	832 (45.81)	0.0104
Mood and depressive disorders	292 (48.18)	788 (43.39)	0.0399
Anxiety disorders	225 (37.13)	625 (34.42)	0.2258
Psychotic disorders and schizophrenia	37 (6.11)	94 (5.18)	0.3812
Other mental health disorders	348 (57.43)	944 (51.98)	0.0201
Psychotropic comedications
Antidepressants	251 (41.42)	771 (42.46)	0.6546
Psychostimulants and α-2 agonists	257 (42.41)	770 (42.40)	0.9971
Mood stabilizers and anticonvulsants	65 (10.73)	162 (8.92)	0.1868
Benzodiazepines and other anxiolytics	69 (11.39)	204 (11.23)	0.9181
Other psychotropic medications	31 (5.12)	91 (5.01)	0.9189

The association of BMI and lab monitoring results with the likelihood of receiving SGA mAEIs

The estimates for the risk factors associated with the likelihood of receiving SGA mAIEs from the conditional logistic regression analysis are presented in Table 5. In the full cohort, compared to patients with normal BMI, those with abnormal BMI had 40% higher odds of receiving mAEIs (odds ratio, OR [95% confidence interval, CI]: BMI_{abnormal vs. normal}: 1.43 [1.13–1.79]), and those with absent BMI had 51% lower odds of receiving mAEIs (OR [95% CI]: BMI_{absent vs. normal}: 0.49 [0.43–0.57]).

Table 5.

Conditional Logistic Regression Results

Covariates	Full cohort, OR [95% CI]	BMI <95 percentile subcohort, OR [95% CI]
BMI (ref: Normal)
Abnormal	1.430 [1.139–1.796]^*	1.139 [0.746–1.738]
Absent	0.498 [0.434–0.570]^*	0.964 [0.772–1.203]
Lab (ref: Normal)
Abnormal	1.096 [0.814–1.475]	1.303 [0.731–2.320]
Absent	0.958 [0.787–1.165]	1.303 [0.925–1.834]
Age group (ref: Adolescents: 13–17 years)
School going (6–12 years)	1.098 [0.969–1.245]	1.093 [0.874–1.368]
Toddlers and preschool (1–5 years)	0.901 [0.677–1.197]	0.977 [0.602–1.587]
Gender (ref: Male)
Female	1.012 [0.899–1.141]	0.934 [0.757–1.151]
Unknown/not reported	1.408 [0.433–4.578]	NA
Racial group (ref: Whites)
Black/African American	0.915 [0.776–1.078]	0.990 [0.749–1.308]
Asian/PIs/NAs	1.041 [0.667–1.625]	0.810 [0.389–1.687]
Unknown/not reported	0.933 [0.784–1.111]	1.102 [0.830–1.464]
SGA usage (ref: Aripiprazole)
Olanzapine	0.584 [0.469–0.726]^*	0.563 [0.379–0.837]
Risperidone	0.744 [0.650–0.851]^*	0.497 [0.391–0.630]^*
Quetiapine	0.707 [0.593–0.844]^*	0.539 [0.398–0.730]^*
Ziprasidone	1.088 [0.773–1.533]	1.107 [0.635–1.928]
Others	1.239 [0.807–1.901]	0.565 [0.282–1.129]
Multiple SGA usage (ref: No)	2.083 [1.701–2.551] ^*	2.375 [1.705–3.307]^*
Comorbid conditions (ref: Absence)
ADHD	1.374 [1.192–1.584]^*	1.291 [1.025–1.626]^*
Mood and depressive disorders	1.502 [1.315–1.716]^*	1.169 [0.944–1.448]
Anxiety disorders	0.884 [0.759–1.029]	1.026 [0.825–1.277]
Psychotic disorders and schizophrenia	1.389 [1.070–1.802]^*	1.157 [0.757–1.769]
Other mental health disorders	1.241 [1.089–1.414]^*	1.158 [0.945–1.420]
Psychotropic comedications (ref: Absence)
Antidepressants	1.085 [0.945–1.245]	0.869 [0.703–1.074]
Psychostimulants and α-2 agonists	1.030 [0.899–1.180]	0.903 [0.720–1.131]
Mood stabilizers and anticonvulsants	1.236 [1.019–1.501]^*	1.172 [0.847–1.621]
Benzodiazepines and other anxiolytics	1.069 [0.860–1.329]	1.046 [0.759–1.441]
Other psychotropic comedications	0.661 [0.481–0.935]^*	0.883 [0.552–1.414]

p < 0.05-significant.

ADHD, attention-deficit/hyperactivity disorder; BMI, body mass index; CI, confidence interval; NAs, native Americans; OR, odds ratio; PIs, Pacific Islanders; SGA, second-generation antipsychotic.

Having abnormal laboratory readings was not associated with the likelihood of receiving mAEIs (OR [95% CI]: Lab_{abnormal vs. normal}: 1.09 [0.81–1.47]). Other factors associated with the likelihood of receiving mAEIs included receiving SGAs associated with moderate metabolic risk (risperidone and quetiapine), multiple SGAs (≥2 different SGAs at index), mood stabilizers, and anticonvulsants and diagnosis of comorbid conditions such as ADHD, mood and depressive disorders, psychotic disorders, and schizophrenia, during the 30-day period before the event date.

The results of conditional logistic regression analysis on the subcohort of patients who were nonobese at the baseline (BMI <95 percentile) showed that having weight increase (BMI ≥95 percentile) or abnormal labs during SGA treatment was not associated with higher odds of receiving mAEIs compared to without (OR [95% CI]: BMI_{abnormal vs. normal}: 1.13 [0.74–1.73] and Labs_{abnormal vs. normal}: 1.30 [0.73–2.32]) (Table 5).

Discussion

The purpose of the study was to evaluate any association between abnormal metabolic parameters and the likelihood of receiving mAEIs. The principal finding of this study is that, as expected, abnormal BMI is an important factor associated with the initiation of mAEIs. Patients with an abnormal BMI value had 40% higher odds of receiving any interventions for SGA-induced metabolic adverse effects (mAEIs) within 30 days. AIWG is the most important adverse metabolic effect of SGAs, and it is evident in more than 80% of children during the first few months of SGA therapy (Dayabandara et al., 2017). AIWG can also impact the pathway of glucose dysregulation or cause lipid abnormalities and metabolic syndrome (Chaggar et al., 2011; Dayabandara et al., 2017; Grajales et al., 2019; Libowitz and Nurmi, 2021).

However, our findings also showed that a statistically significant association was not seen for the abnormal reading of other metabolic parameters and subsequent mAEIs. The descriptive statistics showed that although blood glucose and cholesterol were among the most commonly monitored metabolic parameters for pediatric SGA recipients, abnormal readings for these two parameters were rare. The most common abnormal reading for laboratory tests found in our study was liver enzymes. It is known that SGAs commonly cause asymptomatic elevation of liver enzymes such as ALT (Atasoy et al., 2007). Earlier literature has also shown that ∼20% of patients on SGAs have significantly higher elevated liver enzymes (Meiyanti et al., 2022), thus corroborating our findings.

In a prospective study conducted by Erdogan et al. (2008), it was found that 52.5% of children had clinically nonsignificant elevation of liver enzymes with risperidone usage. The elevation of liver enzymes was associated with the type of SGA use (more prone to increase with olanzapine and clozapine), duration of therapy, and combination of SGAs with psychotropic medications (Gonzalez-Heydrich et al., 2003; Meiyanti et al., 2022). LFT are recommended by CAMSEA guidelines in Canada; however, in the United States, there is no consensus for the treatment of abnormal liver function tests (ALT/aspartate transaminase), and not recommended by guidelines that explain the lack of association between elevated laboratory parameters and SGA mAEIs.

In addition, the findings of the subgroup analysis showed that, for patients with abnormal BMI, physicians might not consider mAIEs when the abnormality was newly detected. In the subgroup analysis on patients who were not affected by obesity before SGA initiation, a significant association between abnormal BMI during SGA treatment and subsequent mAIEs was not detected. The finding is different from the results of the main analysis, implying that there might be competing priorities, such as the effectiveness of the treatment, making prescribers decide not to intervene in the ADE immediately.

Another interesting finding is that patients had 50% lower odds of receiving mAEIs when BMI was not monitored compared to those who had a normal BMI measure during SGA treatment. This could be because having any BMI measure may imply that patients had a follow-up visit in which some other changes in the patient's condition have led to the decision to prescribe an mAEI. There might also be a possibility that patients who did not have BMI measurements were likely to have clinicians who did not consider evaluating and managing AIWG. It is important to point out that some mAEIs, such as switching from SGA polypharmacy to monotherapy and prescribing a comedication with weight loss potential, may not necessarily be due to the concern on SGA-associated metabolic side effects. It could be a regimen adjustment due to the improvement of patients' psychiatric symptoms.

Among those who received the mAEIs, the descriptive statistics showed that the selection of interventions for metabolic side effects was mostly evidence based and guideline concordant. For instance, metformin has the most evidence of efficacy among the different interventions to counter AIWG (De Hert et al., 2011). Our results showed metformin was the most prescribed pharmacological mAEIs. The IMPACT Trial showed that switching to a lower risk SGA is as effective as adding metformin (Correll et al., 2020). Our findings revealed that switching to a lower risk SGA was the next commonly prescribed mAEIs. Nonpharmacological strategies like behavioral interventions are often used as frontline therapy for AIWG (Dayabhandara et al., 2017). This was corroborated in our study that patients in the subcohort who were not affected by obesity at the baseline used more nonpharmacotherapy compared to the full cohort.

This large national retrospective study is critically important because, to our knowledge, it is the first to evaluate the association of metabolic risk factors and receiving subsequent SGA mAEIs in children and adolescents. Our previous research has found that the SGA metabolic monitoring recommendation has been modestly followed. The findings of this study suggest that some monitoring recommendations, especially the laboratory monitoring for cholesterol and blood glucose, should be reconsidered, given the low risk of having abnormalities in the population and its limited impact on subsequent interventions.

Our study likely has several limitations. One inherent limitation is that prescriptions in an EMR database are tracked by medication orders/lists, not prescriptions filled at the pharmacy. There can be some misclassification of exposure if the patients were not filling or taking the prescriptions. Another limitation is that due to data unavailability, we were not able to capture certain ADE management strategies, such as dose reduction. In addition, the metabolic parameters captured could be both fasting and nonfasting due to incomplete fasting designation in the EMR dataset. However, perhaps the greatest limitation would be the lack of SGA effectiveness measure, which has prevented us from understanding how physicians made clinical decisions by weighing the risks and benefits of SGAs.

Conclusions

The prescribing of SGA metabolic ADE interventions was associated with the detection of abnormal BMI during the SGA treatment, but was not associated with abnormal Lab readings, which predominantly comprised abnormal liver function tests. The findings suggest that additional data are needed to clarify the long-term implication of SGA metabolic adverse events other than weight gain and to inform the appropriate timing for ADE interventions.

Clinical Significance

The findings of this EMR-based research on children and adolescents prescribed SGAs showed that having an abnormal BMI measure during SGA treatment was associated with a higher chance of receiving interventions for metabolic ADEs. However, a similar association was not found between abnormal lab results detected during the monitoring of SGA-associated metabolic ADEs and the ADE interventions. The findings indicate that more data is needed to help clinicians understand the long-term implications of abnormalities other than weight gain detected during SGA metabolic monitoring and to inform appropriate timing for the ADE intervention in children and adolescents.

Footnotes

Disclosures

No competing financial interests exist.

Supplementary Material

Supplementary Appendix Table SA1

Supplementary Appendix Table SA2

Supplementary Appendix Table SA3

Supplementary Appendix Table SA4

References

AACAP. Practice parameter for the use of atypical antipsychotic medications in children and adolescents; 2011. Available from https://www.aacap.org/ [Last accessed: April 21, 2023].

American Diabetes Association Professional Practice

. 14 (ADA). Children and Adolescents: Standards of Medical Care in Diabetes-2022. Diabetes Care, 2022; 45:S208–S231; doi: 10.2337/dc21-S013

Atasoy

, Erdogan

, Yalug

, et al. A review of liver function tests during treatment with atypical antipsychotic drugs: A chart review study. Prog Neuropsychopharmacol Biol Psychiatry, 2007; 31:1255–1260; doi: 10.1016/j.pnpbp.2007.05.005

Burcu

, Zito

, Safer

, et al. Concomitant use of atypical antipsychotics with other psychotropic medication classes and the risk of type 2 diabetes mellitus. J Am Acad Child Adolesc Psychiatry, 2017; 56:642–651; doi: 10.1016/j.jaac.2017.04.004

Carli

, Kolachalam

, Longoni

, et al. Atypical antipsychotics and metabolic syndrome: From molecular mechanisms to clinical differences. Pharmaceuticals (Basel), 2021; 14:030238; doi: 10.3390/ph14030238

Chaggar

, Shaw

, Williams

. Effect of antipsychotic medications on glucose and lipid levels. J Clin Pharmacol, 2011; 51:631–638; doi: 10.1177/0091270010368678

Correll

, Sikich

, Reeves

, et al. Metformin add-on vs. antipsychotic switch vs. continued antipsychotic treatment plus healthy lifestyle education in overweight or obese youth with severe mental illness: Results from the IMPACT trial. World Psychiatry, 2020; 19:69–80; doi: 10.1002/wps.20714

Dayabandara

, Hanwella

, Ratnatunga

, et al. Antipsychotic-associated weight gain: Management strategies and impact on treatment adherence. Neuropsychiatr Dis Treat, 2017; 13:2231–2241; doi: 10.2147/NDT.S113099

De Hert

, Dobbelaere

, Sheridan

, et al. 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; 26:144–158; doi: 10.1016/j.eurpsy.2010.09.011

10.

Doane

, Bessonova

, Friedler

, et al. Weight gain and comorbidities associated with oral second-generation antipsychotics: Analysis of real-world data for patients with schizophrenia or bipolar I disorder. BMC Psychiatry, 2022; 22:114; doi: 10.1186/s12888-022-03758-w

11.

Erdogan

, Atasoy

, Akkurt

, et al. Risperidone and liver function tests in children and adolescents: A short-term prospective study. Prog Neuropsychopharmacol Biol Psychiatry, 2008; 32:849–857; doi: 10.1016/j.pnpbp.2007.12.032

12.

Firl

, Raiker

, Pakchanian, et al. Associations between a Strabismus diagnosis and subsequent mental health disorders in children and adolescents. Invest Ophthalmol Vis Sci, 2021; 62(8):74.

13.

Freedman

, Kane

, King

, et al. The effect of prescribing antibiotics with opioids on the development of opioid use disorder: A national database study. J Addict Dis, 2022; 40(1):62–70; doi: 10.1080/10550887.2021.1926889

14.

Fulone

, Silva

, Lopes

. Switching between second-generation antipsychotics in patients with schizophrenia and schizoaffective disorder: 10-Year cohort study in Brazil. Front Pharmacol, 2021; 12:638001; doi: 10.3389/fphar.2021.638001

15.

Gonzalez-Heydrich

, Raches

, Wilens

, et al. Retrospective study of hepatic enzyme elevations in children treated with olanzapine, divalproex, and their combination. J Am Acad Child Adolesc Psychiatry, 2003; 42:1227–1233; doi: 10.1097/00004583-200310000-00014

16.

Grajales

, Ferreira

, Valverde

. Second-generation antipsychotics and dysregulation of glucose metabolism: Beyond weight gain. Cells, 2019; 8:1336.

17.

Grandits

GNJ.

Using SAS^® to Perform Individual Matching in Design of Case-Control Studies. In: SAS Global Forum 2010. Paper 061-2010; 2010.

18.

Harrison

, Buckley

BJR

, Lane

, et al. Associations between COVID-19 and 30-day thromboembolic events and mortality in people with dementia receiving antipsychotic medications. Pharmacol Res, 2021; 167:105534; doi: 10.1016/j.phrs.2021.105534

19.

Harrison

, Luciano

. Incidence of Parkinson's disease, dementia, cerebrovascular disease and stroke in bipolar disorder compared to other psychiatric disorders: An electronic health records network study of 66 million people. Bipolar Disord, 2021; 23(5):454–462; doi: 10.1111/bdi.13022

20.

, Yu

, Zhang

, et al. A real-world study of risk factors for QTc prolongation in schizophrenia patients receiving atypical antipsychotics. J Clin Psychopharmacol, 2022; 42:71–74; doi: 10.1097/JCP.0000000000001501

21.

, Panagiotopoulos

, McCrindle

, et al. Management recommendations for metabolic complications associated with second-generation antipsychotic use in children and youth. Paediatr Child Health, 2011; 16:575–580.

22.

Libowitz

, and Nurmi

. The burden of antipsychotic-induced weight gain and metabolic syndrome in children. Front Psychiatry, 2021; 12:623681; doi: 10.3389/fpsyt.2021.623681.

23.

Mayo Clinic Laboratories. Pediatric Reference Values; 2022. Available from: https://www.mayocliniclabs.com/test-info/pediatric/refvalues/reference.php [Last accessed: April 21, 2023 ].

24.

McIntyre

, Jerrell

. Metabolic and cardiovascular adverse events associated with antipsychotic treatment in children and adolescents. Arch Pediatr Adolesc Med, 2008; 162:929–935; doi: 10.1001/archpedi.162.10.929

25.

Meiyanti

, Mulia

, and Pusparini

. The relationship between antipsychotic drug use and liver enzyme levels in people with mental disorders. J Drug Deliv Ther, 2022; 12:7–11.

26.

Morrato

, Cuffel

, Newcomer

, et al. Metabolic risk status and second-generation antipsychotic drug selection: A retrospective study of commercially insured patients. J Clin Psychopharmacol, 2009; 29:26–32; doi: 10.1097/JCP.0b013e31819294cb

27.

Mukherjee

, Pakhchanian

, Raiker

, et al. Burden of thrombotic events in coronavirus disease-19 (COVID-19) patients and effect on outcomes (from a multicenter electronic health record database). Am J Cardiol, 2021; 147:150–152; doi: 10.1016/j.amjcard.2021.03.003

28.

NHLBI. Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents, volume 2022. NIH National Heart Lung and Blood Institute; Bethesda, MD; 2012.

29.

Pillinger

, McCutcheon

, Vano

, et al. Comparative effects of 18 antipsychotics on metabolic function in patients with schizophrenia, predictors of metabolic dysregulation, and association with psychopathology: A systematic review and network meta-analysis. Lancet Psychiatry, 2020; 7:64–77; doi: 10.1016/S2215-0366(19)30416-X

30.

Pringsheim

, Panagiotopoulos

, Davidson

, et al. Evidence-based recommendations for monitoring safety of second-generation antipsychotics in children and youth. Paediatr Child Health, 2011; 16:581–589.

31.

Richardson

. An incidence density sampling program for nested case-control analyses. Occup Environ Med, 2004; 61:e59; doi: 10.1136/oem.2004.014472

32.

Stroup

, Gray

. Management of common adverse effects of antipsychotic medications. World Psychiatry, 2018; 17:341–356; doi: 10.1002/wps.20567

33.

Sultan

, James

, Wang

, et al. Incidence of uncommon clinical events in USA patients with dialysis-dependent and nondialysis-dependent chronic kidney disease: Analysis of electronic health records from TriNetX. Nephron, 2021; 145(5):462–473; doi: 10.1159/000516280

34.

Taquet

, Geddes

, Husain

, et al. 6-month neurological and psychiatric outcomes in 236 379 survivors of COVID-19: a retrospective cohort study using electronic health records. The Lancet, 2021; 8(5):416–427; doi: 10.1016/S2215-0366(21)00084-5

35.

Vedal

TSJ

, Steen

, Birkeland

, et al. Free thyroxine and thyroid-stimulating hormone in severe mental disorders: A naturalistic study with focus on antipsychotic medication. J Psychiatr Res, 2018; 106:74–81; doi: 10.1016/j.jpsychires.2018.09.014

36.

Whitlock

, O'Connor

, Williams

, et al. Effectiveness of weight management interventions in children: A targeted systematic review for the USPSTF. Pediatrics, 2010; 125(2):e396–418; doi: 10.1542/peds.2009-1955

37.

Zhang

, Spence

, Reeves

, DosReis

. Cardiovascular risk of concomitant use of atypical antipsychotics and stimulants among commercially insured youth in the United States. Front Psychiatry, 2021; 12:640244; doi: 10.3389/fpsyt.2021.640244

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB