Prescription Stimulant Use During Pregnancy and Risk of Neurodevelopmental Disorders in Children

Elizabeth A Suarez; Brian T Bateman; Sonia Hernandez-Diaz; Loreen Straub; Christopher J McDougle; Katherine L Wisner; Kathryn J Gray; Page B Pennell; Barry Lester; Yanmin Zhu; Helen Mogun; Krista F Huybrechts

doi:10.1001/jamapsychiatry.2023.5073

. 2024 Jan 24;81(5):477–488. doi: 10.1001/jamapsychiatry.2023.5073

Prescription Stimulant Use During Pregnancy and Risk of Neurodevelopmental Disorders in Children

Elizabeth A Suarez ^1,^2,^3,^✉, Brian T Bateman ⁴, Sonia Hernandez-Diaz ⁵, Loreen Straub ¹, Christopher J McDougle ^6,⁷, Katherine L Wisner ^8,⁹, Kathryn J Gray ¹⁰, Page B Pennell ¹¹, Barry Lester ^12,^13,¹⁴, Yanmin Zhu ¹, Helen Mogun ¹, Krista F Huybrechts ¹

¹Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts

²Center for Pharmacoepidemiology and Treatment Science, Institute for Health, Health Care Policy and Aging Research, Rutgers University, New Brunswick, New Jersey

³Department of Biostatistics and Epidemiology, Rutgers School of Public Health, Piscataway, New Jersey

⁴Department of Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine, Stanford, California

⁵Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, Massachusetts

⁶Lurie Center for Autism, Massachusetts General Hospital, Lexington

⁷Department of Psychiatry, Harvard Medical School, Boston, Massachusetts

⁸The Asher Center for the Study and Treatment of Depressive Disorders, Department of Psychiatry and Behavioral Sciences, Feinberg School of Medicine, Northwestern University, Chicago, Illinois

⁹Department of Obstetrics and Gynecology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois

¹⁰Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Brigham and Women’s Hospital, Boston, Massachusetts

¹¹Department of Neurology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania

¹²Center for the Study of Children at Risk, Warren Alpert Medical School of Brown University and Women & Infants Hospital, Providence, Rhode Island

¹³Department of Psychiatry and Human Behavior, Warren Alpert Medical School of Brown University, Providence, Rhode Island

¹⁴Department of Pediatrics, Warren Alpert Medical School of Brown University, Providence, Rhode Island

Accepted for Publication: November 8, 2023.

Published Online: January 24, 2024. doi:10.1001/jamapsychiatry.2023.5073

^✉

Corresponding Author: Elizabeth A. Suarez, MPH, PhD, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital & Harvard Medical School, 1620 Tremont St, Boston, MA 02120 (h4p@bwh.harvard.edu).

Author Contributions: Drs Suarez and Huybrechts had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Suarez, Bateman, Straub, Wisner, Pennell, Huybrechts.

Acquisition, analysis, or interpretation of data: Suarez, Bateman, Hernandez-Diaz, McDougle, Gray, Pennell, Lester, Zhu, Mogun, Huybrechts.

Drafting of the manuscript: Suarez, Bateman, Huybrechts.

Critical review of the manuscript for important intellectual content: Bateman, Hernandez-Diaz, Straub, McDougle, Wisner, Gray, Pennell, Lester, Zhu, Mogun, Huybrechts.

Statistical analysis: Suarez, Lester, Mogun.

Obtained funding: Bateman, Hernandez-Diaz, Huybrechts.

Administrative, technical, or material support: Straub, Wisner, Zhu, Huybrechts.

Supervision: Bateman, Hernandez-Diaz, Gray, Huybrechts.

Conflict of Interest Disclosures: Dr Bateman reported grants from the National Institutes of Health during the conduct of the study. Dr Hernandez-Diaz reported grants from Takeda and personal fees from Johnson & Johnson and Moderna outside the submitted work. Dr Straub reported grants from the National Institute of Child Health and Human Development outside the submitted work. Dr McDougle reported consulting fees from Acadia Pharmaceuticals and Sage Therapeutics, royalties from Oxford University Press and Springer Publishing, and a lecture honorarium from American Psychiatric Association outside the submitted work. Dr Gray reported personal fees from BillionToOne, Roche, and Aetion outside the submitted work. Dr Pennell reported personal fees from Harvard T. H. Chan School of Public Health for serving on the BRAINS scientific advisory board during the conduct of the study and grants from the National Institutes of Health outside the submitted work. Dr Zhu reported grants from Takeda outside the submitted work. Dr Huybrechts reported grants from Takeda (paid to institution) and UCB (paid to institution) outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by the National Institute of Mental Health (R01 MH116194).

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentation: This work was presented at the 38th International Conference on Pharmacoepidemiology; August 26, 2022; Copenhagen, Denmark.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC10809143 PMID: 38265792

Key Points

Question

Is prescription stimulant treatment for attention-deficit/hyperactivity disorder during pregnancy associated with neurodevelopmental disorders in children?

Findings

In this cohort study including 7065 pregnancies exposed to amphetamine/dextroamphetamine and 1123 pregnancies exposed to methylphenidate in the second half of pregnancy, neither medication was associated with autism spectrum disorder, attention-deficit/hyperactivity disorder, or a composite of any neurodevelopmental disorder in children after careful adjustment for confounding by indication through various design and analytic approaches.

Meaning

These findings suggest that amphetamine/dextroamphetamine and methylphenidate exposure during pregnancy are not likely to meaningfully increase the risk of childhood neurodevelopmental disorders.

This cohort study evaluates the association between prescription stimulant use during pregnancy and neurodevelopmental disorders in childhood.

Abstract

Importance

Use of medications for attention-deficit/hyperactivity disorder (ADHD) during pregnancy is increasing in the US. Whether exposure to these medications in utero impacts the risk of neurodevelopmental disorders in children is uncertain.

Objective

To evaluate the association of childhood neurodevelopmental disorders with in utero exposure to stimulant medications for ADHD.

Design, Setting, and Participants

This cohort study included health care utilization data from publicly insured (Medicaid data from 2000 to 2018) and commercially insured (MarketScan Commercial Claims Database data from 2003 to 2020) pregnant individuals aged 12 to 55 years in the US with enrollment from 3 months prior to pregnancy through 1 month after delivery, linked to children. Children were monitored from birth until outcome diagnosis, disenrollment, death, or end of the study (December 2018 for Medicaid and December 2020 for MarketScan).

Exposures

Dispensing of amphetamine/dextroamphetamine or methylphenidate in the second half of pregnancy.

Main Outcomes and Measures

Autism spectrum disorder, ADHD, and a composite of any neurodevelopmental disorder were defined using validated algorithms. Hazard ratios were estimated comparing amphetamine/dextroamphetamine and methylphenidate to no exposure.

Results

The publicly insured cohort included 2 496 771 stimulant-unexposed, 4693 amphetamine/dextroamphetamine-exposed, and 786 methylphenidate-exposed pregnancies with a mean (SD) age of 25.2 (6.0) years. The commercially insured cohort included 1 773 501 stimulant-unexposed, 2372 amphetamine/dextroamphetamine-exposed, and 337 methylphenidate-exposed pregnancies with a mean (SD) age of 31.6 (4.6) years. In unadjusted analyses, amphetamine/dextroamphetamine and methylphenidate exposure were associated with a 2- to 3-fold increased risk of the neurodevelopmental outcomes considered. After adjustment for measured confounders, amphetamine/dextroamphetamine exposure was not associated with any outcome (autism spectrum disorder: hazard ratio [HR], 0.80; 95% CI, 0.56-1.14]; ADHD: HR, 1.07; 95% CI, 0.89-1.28; any neurodevelopmental disorder: HR, 0.91; 95% CI, 0.81-1.28). Methylphenidate exposure was associated with an increased risk of ADHD (HR, 1.43; 95% CI, 1.12-1.82]) but not other outcomes after adjustment (autism spectrum disorder: HR, 1.06; 95% CI, 0.62-1.81; any neurodevelopmental disorder: HR, 1.15; 95% CI, 0.97-1.36). The association between methylphenidate and ADHD did not persist in sensitivity analyses with stricter control for confounding by maternal ADHD.

Conclusions and Relevance

The findings in this study suggest that amphetamine/dextroamphetamine and methylphenidate exposure in utero are not likely to meaningfully increase the risk of childhood neurodevelopmental disorders.

Introduction

The prevalence of attention deficit/hyperactivity disorder (ADHD) in adults is estimated to be approximately 5% in the US.^1,2 Multiple studies have shown an increase in the prevalence of adult ADHD diagnoses in recent years.^3,4,5,6,7 Consequently, treatment with prescription medications for ADHD in adults has also markedly increased.^3,8,9 Among commercially insured female individuals of reproductive age in the US, stimulant medication use increased more than 3-fold between 2003 and 2015, from 0.8% to 3.9%, while nonstimulant ADHD medication use remained steady at approximately 0.2%.⁹ Similar increases in use of ADHD medications have been observed among pregnant individuals in the US; by 2013, approximately 1.2% reported ADHD medication use.¹⁰ amphetamine/dextroamphetamine is by far the most commonly used stimulant medication for ADHD in adults in the US followed by methylphenidate, whereas nonstimulant medications are less common.^9,11

Stimulant medications cross the placenta and can increase the concentrations of norepinephrine and dopamine, which may play an important role in fetal neurodevelopment.^12,13 A recent qualitative review including 9 cohort studies concluded that data were insufficient to draw conclusions regarding the safety of stimulant medication use in pregnancy with respect to long-term adverse outcomes in the offspring, including neurodevelopmental outcomes.¹⁴ Studies of children exposed to illicit methamphetamine use in pregnancy have reported increased anxiety, depression, and emotional and developmental issues compared to unexposed children.^{12,15,16,17,18} However, the potential for neurodevelopmental effects with prescription stimulant use for ADHD has been reported in only 2 studies, including 898 and 118 exposed pregnancies from Denmark and Canada, respectively.^19,20 Findings were inconsistent across studies, and methylphenidate was the primary medication used to treat ADHD. This leaves a large gap in knowledge for pregnant individuals treated with amphetamines.

The objective of this study was to assess the association between stimulant use for ADHD during pregnancy and neurodevelopmental outcomes in the offspring among publicly and commercially insured pregnancies, stratified by medication type and timing of exposure. In addition to neurodevelopmental outcomes overall, we focus on autism spectrum disorder (ASD) and ADHD.

Methods

Data Sources and Study Cohorts

We defined pregnancy cohorts nested in national Medicaid data from 2000 to 2018, which includes health care utilization data for Medicaid beneficiaries, and the MarketScan Commercial Claims Database (MarketScan) from 2003 to 2020, which includes commercial health insurance claims data. The development of the Medicaid pregnancy cohort has been previously described,²¹ and similar methods were used in the development of the MarketScan cohort. Both data sources include information on demographics, diagnoses, and procedures received during all health care encounters (inpatient, outpatient, or emergency department visits) and dispensed outpatient prescription medications. Pregnant individuals aged 12 to 55 years with live-birth deliveries linked to children were required to have insurance coverage from 3 months before the date of the estimated last menstrual period to a minimum of 1 month after delivery. Children were monitored from their date of birth until their continuous enrollment ended, they developed the specific neurodevelopmental disorder (NDD) of interest, the study period ended (December 2018 for Medicaid and December 2020 for MarketScan), or they died, whichever came first.

This study was approved by the institutional review board of Brigham and Women’s Hospital, which waived the need for informed consent because data were deidentified. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline was followed.

Stimulant Exposure

We identified pregnant individuals filling prescriptions for amphetamine/dextroamphetamine and methylphenidate during pregnancy. The primary exposure window was defined as the time period from 127 days after the last menstrual period (week 19 of gestation) to delivery, the approximate period of peak synaptogenesis.²² Because the precise etiologically relevant exposure window for neurodevelopmental impairment is unknown, we defined a secondary exposure window during the first half of pregnancy (last menstrual period through 126 days after last menstrual period). For each exposure group of interest (amphetamine/dextroamphetamine and methylphenidate), pregnancies with exposure to other stimulant medications from 90 days prior to the last menstrual period through delivery were excluded. The unexposed group was defined as having no stimulants dispensed from 90 days prior to the last menstrual period through delivery.

Neurodevelopmental Outcomes

Any NDD was defined as a composite of ASD, ADHD, specific learning disorders, developmental speech or language disorder, developmental coordination disorder, intellectual disability, and behavioral disorder. Validated claims-based algorithms with high positive predictive values were used to define each outcome (eTable 1 in Supplement 1).²³ This composite outcome was considered given the frequent co-occurrence of NDDs and the potential for shared mechanisms across NDDs.²⁴ In addition to this composite outcome, we considered ASD, one of the more severe NDDs for which there has been a steady increase in diagnoses over the study period,²⁵ and ADHD, the most common NDD, as individual outcomes.²⁴

Covariates

Potential confounders included demographic characteristics (maternal age; race and ethnicity [Medicaid only], reported as recorded in Medicaid application and included in this study because race and ethnicity are associated with stimulant medication use and NDD diagnosis^7,24,26,27; state; and delivery year), ADHD, other maternal mental health diagnoses (depression, anxiety, and bipolar disorder), proxies for severity of mental health conditions (numbers of psychiatric visits, inpatient and emergency visits for mental health diagnoses, mental health diagnoses, and dispensations of other psychotropic medications), lifestyle factors (smoking, alcohol, and substance misuse), other medications, maternal comorbidities, adequacy of prenatal care, and socioeconomic level indicators (Medicaid only, through linkage of maternal zip codes with county-level SES measures provided by the US Department of Agriculture²⁸). Covariates were assessed from 90 days prior to the last menstrual period through the first half of pregnancy or through the day prior to delivery (depending on the covariate; refer to eTable 2 in Supplement 1 for the full list of covariates and their respective assessment periods).

Main Analyses

To estimate the cumulative incidence of the outcomes, children were followed up from birth until the outcome or censoring, marked by the end of enrollment, the end of the study period, or death. Differences in covariates across exposure groups were compared using standardized differences).²⁹ To control for measured confounders, weighting based on fine stratification of the propensity score was used to weight the referent population to the target population of the treated pregnancies.³⁰ Propensity scores were estimated for each exposure group using logistic regression including all covariates listed above without variable selection. Fifty strata were created after trimming the nonoverlapping regions of the propensity score distributions.³⁰ Cox proportional hazards regression was used to calculate crude and weighted hazard ratios (HRs) with robust estimates of standard errors to account for multiple pregnancies in a single individual. HRs were estimated in the Medicaid and MarketScan cohorts separately and were pooled using fixed-effect meta-analysis (results for the Medicaid and MarketScan cohorts separately are presented in eTables 3-12 in Supplement 1).

Sensitivity Analyses

Two additional analyses were conducted to further reduce the potential for residual confounding by severity of the indication. First, we restricted the cohort to pregnant individuals with a diagnosis of ADHD (rather than adjust for the indication). ADHD diagnoses were identified at any time during pregnancy or any time in their enrollment history prior to pregnancy (minimum of 90 days prior to pregnancy based on inclusion criteria). Second, we compared pregnant individuals with stimulant exposure in pregnancy to individuals who were unexposed during pregnancy but filled a prescription for the medication of interest prior to pregnancy. The rationale for this analysis is that those who discontinued prescription use might be more comparable to those who continued prescription use than individuals who were never treated with stimulants. Individuals taking amphetamine/dextroamphetamine were compared to those who discontinued amphetamine/dextroamphetamine, and those taking methylphenidate were compared to those who discontinued methylphenidate. Discontinuation was defined as having a dispensing for the medication in the window from 90 to 31 days prior to the last menstrual period but not during the window of 30 days prior to the last menstrual period through delivery. Characteristics of the ADHD population and discontinuer populations are presented in eTables 13-20 in Supplement 1. To address exposure misclassification due to not consuming the dispensed medication, we required a minimum of 2 dispensations during the exposure window to meet the exposure definition.

In the main analyses, follow-up started at birth. However, most NDDs cannot be diagnosed before 2 years of age and some children will be censored during the first 2 years. If censoring is differential by exposure this may lead to selection bias. To explore this, we restricted the cohort to those mother-infant dyads for whom the children were still enrolled at age 2 years and started follow-up at 2 years. When interpreting the results, we focused on the consistency of estimates across the main and sensitivity analyses, the strength of the adjusted HR, and the precision of the estimate as reflected in the width of the 95% CI, instead of dichotomizing the results as statistically significant (P values less than .05) or not.³¹

Post Hoc Analyses

We implemented propensity score overlap weighting to assess whether results were sensitive to the weighting method used. Small outcome counts can be vulnerable to outsized impacts when using weighting for confounding control. Overlap weights are bound between 0 and 1, making extreme weights impossible.^32,33 The target population of this analysis was pregnancies with overlapping propensity score distributions, which can be interpreted as the population with similar characteristics and probability of being in either the treated or untreated group.³² All analyses were conducted using SAS version 9.4 (SAS Institute). No adjustments were made for multiple comparisons.

Results

The Medicaid cohort included 2 522 490 pregnancies (mean, [SD] age of 25.2 [6.0] years), among which 2 496 771 were unexposed to stimulants during pregnancy. During the second half of pregnancy, 4693 were exposed to amphetamine/dextroamphetamine only and 786 were exposed to methylphenidate only. The corresponding numbers during the first half of pregnancy were 11 886 and 2934 (eFigure 1A in Supplement 1). The MarketScan cohort included 1 795 012 pregnancies mean (SD) age of 31.6 (4.6) years, among which 1 773 501 were unexposed to stimulants throughout pregnancy. During the second half of pregnancy, 2372 were exposed to amphetamine/dextroamphetamine and 337 to methylphenidate. The corresponding numbers during the first half of pregnancy were 9687 and 1443 (eFigure 1B in Supplement 1). Among pregnant individuals enrolled in Medicaid, those taking amphetamine/dextroamphetamine tended to be older and those taking methylphenidate were younger compared to the unexposed group (Table; eTables 21-24 in Supplement 1). Similar age trends were not observed in the MarketScan cohort. Publicly insured amphetamine/dextroamphetamine and methylphenidate users were predominately White, unlike the unexposed group (race and ethnicity information available in Medicaid only; see Table). Mental health conditions, alcohol, tobacco, and substance use; other prescription medication use; and psychiatric visits were all more common among those taking amphetamine/dextroamphetamine and methylphenidate than the unexposed in both cohorts. After fine stratification weighting, all covariates were balanced based on absolute standardized differences of less than 10% (eTable 21-24 in Supplement 1).

Table. Select Characteristics of the Stimulant-Exposed (in Late Pregnancy) and Stimulant-Unexposed Pregnant Cohorts.

	No. (%)
	Medicaid					MarketScan Commercial Claims Database
	Stimulant unexposed (n = 2 496 771)	AMPH/DEX exposed (n = 4693)	Absolute std diff (%), AMPH/DEX vs unexposed	MPH exposed (n = 786)	Absolute std diff (%), MPH vs unexposed	Stimulant unexposed (n = 1 773 501)	AMPH/DEX exposed (n = 2372)	Absolute std diff (%), AMPH/DEX vs unexposed	MPH exposed (n = 337)	Absolute std diff (%), MPH vs unexposed
Maternal age, y
≤19	467 197 (18.7)	603 (12.8)	16.1	220 (28.0)	22.1	4769 (0.3)	3 (0.1)	3.2	6 (1.8)	15.1
20-24	795 903 (31.9)	976 (20.8)	25.4	159 (20.2)	26.8	87 995 (5.0)	73 (3.1)	9.6	9 (2.7)	12.0
25-29	663 907 (26.6)	1427 (30.4)	8.5	205 (26.1)	1.2	494 094 (27.9)	615 (25.9)	4.4	73 (21.7)	14.4
30-34	367 327 (14.7)	1149 (24.5)	24.8	134 (17.0)	6.4	721 102 (40.7)	1077 (45.4)	9.6	136 (40.4)	0.6
35-39	161 489 (6.5)	461 (9.8)	12.3	57 (7.3)	3.1	381 361 (21.5)	509 (21.5)	0.1	93 (27.6)	14.2
≥40	40 948 (1.6)	77 (1.6)	0	11 (1.4)	2.0	84 180 (4.7)	95 (4.0)	3.6	20 (5.9)	5.3
Race and ethnicity^a
Asian or Other Pacific Islander	87 023 (3.5)	15 (0.3)	23.3	3 (0.4)	22.7	NA	NA	NA	NA	NA
Black or African American	756 165 (30.3)	341 (7.3)	61.7	110 (14.0)	40.0	NA	NA	NA	NA	NA
Hispanic or Latino^b	515 307 (20.6)	197 (4.2)	51.5	28 (3.6)	54.3	NA	NA	NA	NA	NA
White	998 837 (40.0)	3837 (81.8)	94.7	597 (76.0)	78.2	NA	NA	NA	NA	NA
Unknown or other^c	139 439 (5.6)	303 (6.5)	3.7	48 (6.1)	2.2	NA	NA	NA	NA	NA
Indications and mental health conditions
ADHD	11 938 (0.5)	3116 (66.4)	195.3	460 (58.5)	165	2855 (0.2)	1442 (60.8)	175	156 (46.3)	130.4
Depression	227 074 (9.1)	1620 (34.5)	64.7	285 (36.3)	68.6	91 647 (5.2)	480 (20.2)	46.5	86 (25.5)	58.9
Anxiety	168 174 (6.7)	1724 (36.7)	78.1	241 (30.7)	64.5	89 899 (5.1)	580 (24.5)	56.8	79 (23.4)	54.5
Bipolar disorder	61 691 (2.5)	879 (18.7)	54.8	140 (17.8)	52.5	8477 (0.5)	99 (4.2)	24.7	13 (3.9)	23.4
Other mental health disorders	51 105 (2.0)	300 (6.4)	21.7	58 (7.4)	25.4	13 981 (0.8)	59 (2.5)	13.4	16 (4.7)	24.3
Sleep disorders	28 832 (1.2)	272 (5.8)	25.6	45 (5.7)	25.3	21 406 (1.2)	160 (6.7)	28.6	53 (15.7)	54.0
Psychiatric visits
0 Visits	2 444 347 (97.9)	4072 (86.8)	42.8	682 (86.8)	42.8	1 736 585 (97.9)	2074 (87.4)	41.1	298 (88.4)	38.3
1 Visit	18 390 (0.7)	154 (3.3)	18.2	36 (4.6)	24.1	7361 (0.4)	77 (3.2)	21.2	9 (2.7)	18.4
2-3 Visits	12 634 (0.5)	137 (2.9)	18.7	28 (3.6)	21.8	8461 (0.5)	96 (4.0)	24.2	7 (2.1)	14.3
≥4 Visits	21 400 (0.9)	330 (7.0)	32.1	40 (5.1)	25.1	21 094 (1.2)	125 (5.3)	23.2	23 (6.8)	29.0
Mental health diagnoses
0 Diagnoses	2 062 821 (82.6)	754 (16.1)	178.4	160 (20.4)	159.3	1 617 138 (91.2)	719 (30.3)	159.4	119 (35.3)	142.2
1 Diagnosis	207 951 (8.3)	595 (12.7)	14.2	94 (12.0)	12.1	78 283 (4.4)	490 (20.7)	50.6	67 (19.9)	48.7
2-3 Diagnoses	147 679 (5.9)	1472 (31.4)	69.2	250 (31.8)	70.1	59 174 (3.3)	757 (31.9)	80.9	97 (28.8)	73.9
≥4 Diagnoses	78 320 (3.1)	1872 (39.9)	100.0	282 (35.9)	90.7	18 906 (1.1)	406 (17.1)	58.1	54 (16.0)	55.5
Any mental health ED visit	99 322 (4.0)	647 (13.8)	35.0	97 (12.3)	30.9	11 827 (0.7)	82 (3.5)	19.7	11 (3.3)	18.8
Any mental health hospitalization	27 655 (1.1)	177 (3.8)	17.3	29 (3.7)	16.9	2438 (0.1)	18 (0.8)	9.3	4 (1.2)	13.0
Other nonstimulant psychiatric medications
0 Drugs	2 205 165 (88.3)	2119 (45.2)	103.1	405 (51.5)	87.6	1 632 286 (92.0)	1563 (65.9)	67.7	217 (64.4)	71.1
1 Drug	215 632 (8.6)	1318 (28.1)	51.9	186 (23.7)	41.7	117 561 (6.6)	518 (21.8)	44.6	85 (25.2)	52.5
2-3 Drugs	66 395 (2.7)	996 (21.2)	59.8	150 (19.1)	54.7	21 885 (1.2)	249 (10.5)	40.2	28 (8.3)	33.7
≥4 Drugs	9579 (0.4)	260 (5.5)	30.8	45 (5.7)	31.4	1769 (0.1)	42 (1.8)	17.4	7 (2.1)	19.1
Substance use
Alcohol use disorder	26 173 (1.0)	161 (3.4)	16.2	27 (3.4)	16.2	2088 (0.1)	17 (0.7)	9.3	3 (0.9)	10.9
Tobacco use	244 190 (9.8)	1389 (29.6)	51.5	184 (23.4)	37.3	24 677 (1.4)	168 (7.1)	28.5	11 (3.3)	12.4
Substance use disorders	126 352 (5.1)	1078 (23.0)	53.4	164 (20.9)	48.4	6703 (0.4)	107 (4.5)	27.0	9 (2.7)	18.8
Obstetric comorbidity index score
0	1 260 395 (50.5)	1395 (29.7)	43.3	290 (36.9)	27.7	767 763 (43.3)	802 (33.8)	19.6	114 (33.8)	19.5
1	581 685 (23.3)	1081 (23.0)	0.6	158 (20.1)	7.8	472 274 (26.6)	636 (26.8)	0.4	85 (25.2)	3.2
2	327 312 (13.1)	932 (19.9)	18.3	139 (17.7)	12.7	271 704 (15.3)	401 (16.9)	4.3	61 (18.1)	7.5
≥3	327 379 (13.1)	1285 (27.4)	36.1	199 (25.3)	31.4	261 760 (14.8)	533 (22.5)	19.9	77 (22.8)	20.8
Other prescription medications
Antidepressants	271 149 (10.9)	2336 (49.8)	93.5	383 (48.7)	91.0	151 027 (8.5)	761 (32.1)	61.3	155 (46.0)	92.8
Anticonvulsants	66 041 (2.6)	973 (20.7)	58.7	151 (19.2)	55.1	21 139 (1.2)	155 (6.5)	28.0	29 (8.6)	34.9
Antipsychotics	65 812 (2.6)	703 (15.0)	44.6	148 (18.8)	54.2	10 925 (0.6)	99 (4.2)	23.4	24 (7.1)	34.2
Anxiolytics/sedatives/hypnotics	215 986 (8.7)	1085 (23.1)	40.4	178 (22.6)	39.3	70 198 (4.0)	378 (15.9)	40.9	61 (18.1)	46.3
Benzodiazepines	89 630 (3.6)	1695 (36.1)	89.3	197 (25.1)	64.4	70 743 (4.0)	552 (23.3)	58.6	64 (19.0)	48.4
Opioids	671 179 (26.9)	2346 (50.0)	48.9	385 (49.0)	46.8	239 097 (13.5)	632 (26.6)	33.3	87 (25.8)	31.4
Corticosteroids	478 077 (19.1)	1387 (29.6)	24.4	247 (31.4)	28.5	337 872 (19.1)	615 (25.9)	16.5	102 (30.3)	26.2
Antihypertensives	142 498 (5.7)	653 (13.9)	27.9	135 (17.2)	36.6	100 529 (5.7)	261 (11.0)	19.4	44 (13.1)	25.6

Open in a new tab

Abbreviations: ADHD, attention-deficit/hyperactivity disorder; AMPH/DEX, amphetamine/dextroamphetamine; ED, emergency department; MPH, methylphenidate; NA, not applicable; std diff, standardized difference.

^{^a}

Race and ethnicity were determined on the basis of information submitted to the Centers for Medicare & Medicaid Services by individual states, which was based on information that had been collected and coded from Medicaid applications. Race and ethnicity were reported in this study because of their known association with stimulant medication use and neurodevelopmental disorder diagnosis. Race and ethnicity data were not available in the MarketScan Commercial Claims Database.

^{^b}

The category Hispanic or Latino included those with missing race information, whereas data under other or unknown included Hispanic or Latino with 1 or more races.

^{^c}

The category other or unknown included individuals who were American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, Hispanic or Latino (≥1 race), more than 1 race, and unknown.

The cumulative incidence of any neurodevelopmental outcome was greater among amphetamine/dextroamphetamine– and methylphenidate-exposed children than unexposed children in both cohorts (Figure 1). By age 12 years, the cumulative incidence of any NDD was 52.0% (95% CI, 42.5-62.1) in the Medicaid cohort and 42.4% (95% CI, 32.6-57.3) in the MarketScan cohort with late pregnancy amphetamine/dextroamphetamine exposure, compared with 33.4% (95% CI, 33.2-33.6) and 22.6% (95% CI, 22.4-22.8), respectively, among unexposed individuals. The cumulative incidence by age 12 years of any NDD with late pregnancy methylphenidate exposure was 68.2% (95% CI, 54.9-80.8) in the Medicaid cohort, and 42.7% (95% CI, 25.2-65.8) in the MarketScan cohort. Both ASD and ADHD were more frequent among amphetamine/dextroamphetamine– and methylphenidate-exposed than among unexposed children (eFigures 1-2 in Supplement 1).

Figure 1. — Crude cumulative incidence curves are presented for the composite of any neurodevelopmental disorder, stratified by stimulant exposure status (defined as having a stimulant dispensed between 127 days after last menstrual period to the day prior to delivery) and data source. The cumulative incidence at age 12 years in each stratum is noted in the figure.

In unadjusted analyses pooled across cohorts, amphetamine/dextroamphetamine exposure in both late and early pregnancy was associated with approximately a doubling of risk of ASD, ADHD, and NDD overall (Figure 2A). After adjustment, amphetamine/dextroamphetamine exposure was not associated with any of the three neurodevelopmental outcomes considered (eg, late pregnancy exposure: ASD HR, 0.80; 95% CI, 0.56-1.14; ADHD HR, 1.07; 95% CI, 0.89-1.28; any NDD HR, 0.91; 95% CI, 0.81-1.28). When restricting to a population with a maternal ADHD diagnosis, no increase in risk was observed for any outcome in either unadjusted or adjusted analyses (Figure 2B). When comparing to individuals who discontinued amphetamine/dextroamphetamine prior to pregnancy, no association was observed for ADHD and NDD overall (Figure 2C), and the adjusted HR for ASD was 1.81 (95% CI, 1.04-3.12) for late exposure and 1.35 (0.84-2.15) for early exposure. The association for ASD was driven primarily by results in the MarketScan cohort (eg, Medicaid HR, 1.26; 95% CI, 0.58-2.73 and MarketScan HR, 2.59; 95% CI, 1.20-5.60 for late pregnancy exposure) (eTable 7 in Supplement 1). Results for ADHD and NDD overall were consistent with no association in both cohorts.

Figure 2. — The exposed cohort is defined by amphetamine/dextroamphetamine dispensing in late (127 days after the last menstrual period to the day prior to delivery) or early (last menstrual period to 126 days after last menstrual period) pregnancy. The reference group consists of all stimulant-unexposed pregnancies (A), all stimulant-unexposed pregnancies with a diagnosis of ADHD (B), or pregnancies with a dispensing of amphetamine/dextroamphetamine between 90 and 31 days prior to the last menstrual period but not from 30 days prior to the last menstrual period through delivery (C). ADHD indicates attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; NDD, neurodevelopmental disorder.

Unadjusted estimates for methylphenidate exposure suggested a 2- to 3-fold increase in risk of ASD, ADHD, and NDD overall (Figure 3A). Adjustment for measured confounders shifted estimates substantially toward the null. The adjusted HRs for late pregnancy exposure were 1.06 (95% CI, 0.62-1.81) for ASD, 1.43 (95% CI, 1.12-1.82) for ADHD, and 1.15 (95% CI, 0.97-1.36) for NDD overall. Associations were further attenuated in analyses restricted to the population with maternal ADHD (Figure 3B) and when comparing to individuals who discontinued methylphenidate prior to pregnancy (Figure 3C).

Results from sensitivity analyses defining exposure as having at least 2 dispensations of the medication and restricting to the population with at least 2 years of follow-up after birth were consistent with those from the main analyses (Figure 4). In post hoc analysis using propensity score overlap weights, results were similar to the main analyses when comparing to stimulant unexposed and within the ADHD population (eFigures 5A-B, 6A-B in Supplement 1). When comparing to those who discontinued medication, there was no association for any outcome with amphetamine/dextroamphetamine exposure, including for ASD (late exposure HR, 1.24; 95% CI, 0.73-2.11; early exposure HR, 1.17; 95% CI, 0.75-1.10) (eFigure 5C in Supplement 1).

Figure 4. — The exposed cohort is defined by stimulant dispensing (A, amphetamine/dextroamphetamine and B, methylphenidate) in late (127 days after the last menstrual period to the day prior to delivery) or early (last menstrual period to 126 days after last menstrual period) pregnancy. The reference group consists of all stimulant-unexposed pregnancies. Sensitivity analysis results presented include requiring 2 or more instances of dispensing the exposure medication during the exposure window to be included in the exposed group (≥2 fills) and restricting the cohort to children with at least 2 years of continuous enrollment after delivery (>2 years of follow-up). ADHD indicates attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; NDD, neurodevelopmental disorder.

Discussion

In this cohort of more than 4 million insured pregnancies in the US with up to 18 years of follow-up, considering results of both main and sensitivity analyses, amphetamine/dextroamphetamine (7065 late and 21 573 early exposure pregnancies) and methylphenidate (1123 late and 4377 early exposed pregnancies) were not meaningfully associated with an increased risk of NDDs overall. While unadjusted associations indicated a 2- to 3-fold increase in risk of ASD, ADHD, and NDD overall following in utero exposure to stimulants, these risks were no longer observed after implementing various approaches that improved our ability to adjust for confounding by ADHD and other factors.

To our knowledge, only 2 studies have explored the potential association between ADHD medications used in pregnancy and neurodevelopmental outcomes in children; most patients were treated with methylphenidate in both populations. A population-based study in Denmark including 898 pregnancies exposed to ADHD medications reported no increase in risk of several neurodevelopmental and growth outcomes in exposed offspring, but an association with childhood ADHD could not be ruled out, although the authors speculated this may be driven by severity of maternal ADHD.²⁰ A study in Quebec, Canada, reported an increased risk of ADHD (approximately 2-fold) among the offspring exposed to ADHD medications compared to unexposed after adjustment for measured confounders. This association was attenuated after implementing additional analyses to control for confounding by indication, including a comparison of exposed and unexposed siblings. Estimates from the sibling analysis were close to null but were imprecise as they were based on 118 sibling pairs only.¹⁹ We similarly observed an association with ADHD when comparing methylphenidate-exposed pregnancies to unexposed (late pregnancy exposure: HR, 1.43; 95% CI, 1.12-1.82). However, the HR shifted toward the null when restricting to a maternal ADHD population (HR, 1.20; 95% CI, 0.89-1.60) and when using methylphenidate discontinuers as a reference group (HR, 0.91; 95% CI, 0.59-1.42). These results corroborate the role of confounding by indication in results from analyses that compare to unexposed pregnancies.

The results for amphetamine/dextroamphetamine consistently suggested no increase in risk for any NDD in children after in utero exposure across the main and sensitivity analyses, with 1 exception. The discontinuer referent analysis resulted in an increased risk of ASD with amphetamine/dextroamphetamine exposure in late pregnancy, which was inconsistent with results observed in the unexposed referent and ADHD-restricted analyses (no association). The association between amphetamine/dextroamphetamine use compared to treatment discontinuation and ASD was attenuated when using an alternative propensity score weighting method. Taken together, these results do not suggest that amphetamine/dextroamphetamine exposure in pregnancy is associated with an increase in risk of ASD.

While we did not observe an increased risk of NDDs with stimulant exposure, some adjusted HRs were below the null, regardless of the propensity score weighting method used, particularly for amphetamine/dextroamphetamine, suggesting a lower risk of NDDs with stimulant exposure. The observed protective association may be due to chance or to improved lifestyle, health, and prenatal care behaviors among stimulant-treated pregnant individuals.

Strengths and Limitations

Our study has several strengths, including the use of mother-child linked birth cohorts of publicly and commercially insured individuals nested in nationally representative data sources²⁴; the large cohort size that allowed for the assessment of different exposure windows; careful attention to a broad range of potential confounders; and sensitivity analyses to evaluate the robustness of the findings to residual confounding, exposure misclassification, and selection bias.

The results of this study should be interpreted in light of limitations inherent in its design. Stimulants for ADHD may be used as needed; therefore, evidence of a dispensing may not indicate consumption. However, results were consistent when requiring 2 dispensations of the medication to qualify as exposed, which suggests exposure misclassification is unlikely to explain our null findings. Outcome misclassification is possible due to the use of algorithms based on recorded diagnoses and treatments to identify NDD in children. However, the relative risks are expected to be unbiased when using highly specific algorithms (which is demonstrated for our outcomes by high positive predictive values), even if sensitivity is low.³⁴ Moreover, event rates for NDD based on these definitions are consistent with US statistics.²⁴ There was substantial loss to follow-up throughout the maximum study period of 18 years due to insurance disenrollment. Restricting the cohort to children with at least 2 years of enrollment resulted in similar findings as for the main analysis. In earlier studies, we have demonstrated that selection bias is unlikely due to disenrollment throughout childhood.^24,35,36 Characteristics of pregnant individuals and children did not differ by length of enrollment after birth.²⁴ Application of censoring weights in an analysis of antipsychotics in pregnancy and NDDs in these cohorts also did not have a meaningful impact on results, suggesting little to no bias from informative censoring.³⁵

Conclusions

The results of this cohort study of pregnant individuals and their children in 2 large health care utilization databases in the US suggest that in utero exposure to amphetamine/dextroamphetamine and methylphenidate are not associated with an increase in NDDs in children. Given the recent rise in use of stimulant medications for ADHD in adults and during pregnancy, these results are reassuring for patients who depend on these medications throughout pregnancy for control of debilitating ADHD symptoms that interfere with daily functioning.

Supplement 1.

eTable 1. Outcome definitions for neurodevelopmental disorders

eTable 2. Full list of covariates included in propensity score models and assessment periods

Results by data source

eTable 3. Neurodevelopmental outcomes among amphetamine/dextroamphetamine exposed compared to stimulant unexposed in pregnancy, by data source

eTable 4. Neurodevelopmental outcomes among methylphenidate exposed compared to stimulant unexposed in pregnancy, by data source

eTable 5. Neurodevelopmental outcomes among amphetamine/dextroamphetamine exposed compared to stimulant unexposed in pregnancy, ADHD population, by data source

eTable 6. Neurodevelopmental outcomes among methylphenidate exposed compared to stimulant unexposed in pregnancy, ADHD population, by data source

eTable 7. Neurodevelopmental outcomes among amphetamine/dextroamphetamine exposed in pregnancy compared to amphetamine/dextroamphetamine discontinuers prior to pregnancy, by data source

eTable 8. Neurodevelopmental outcomes among methylphenidate exposed in pregnancy compared to methylphenidate discontinuers prior to pregnancy, by data source

eTable 9. Neurodevelopmental outcomes among amphetamine/dextroamphetamine exposed compared to stimulant unexposed in pregnancy, requiring at least 2 fills of amphetamine/dextroamphetamine to be counted as exposed, by data source

eTable 10. Neurodevelopmental outcomes among methylphenidate exposed compared to stimulant unexposed in pregnancy, requiring at least 2 fills of methylphenidate to be counted as exposed, by data source

eTable 11. Neurodevelopmental outcomes among amphetamine/dextroamphetamine exposed compared to stimulant unexposed in pregnancy, restricting to children with at least 2 years of follow-up, by data source

eTable 12. Neurodevelopmental outcomes among methylphenidate exposed compared to stimulant unexposed in pregnancy, restricting to children with at least 2 years of follow-up, by data source

eTable 13. Full list of characteristics of the late pregnancy stimulant exposed and unexposed pregnant cohorts, restricted to pregnancies with ADHD diagnosis, Medicaid cohort

eTable 14. Full list of characteristics of the late pregnancy stimulant exposed and unexposed pregnant cohorts, restricted to pregnancies with ADHD diagnosis, MarketScan cohort

eTable 15. Full list of characteristics of the early pregnancy stimulant exposed and unexposed pregnant cohorts, restricted to pregnancies with ADHD diagnosis, Medicaid cohort

eTable 16. Full list of characteristics of the early pregnancy stimulant exposed and unexposed pregnant cohorts, restricted to pregnancies with ADHD diagnosis, MarketScan cohort

Cohorts with stimulant discontinuer reference group

eTable 17. Full list of characteristics of the amphetamine/dextroamphetamine exposed and amphetamine/dextroamphetamine discontinuer pregnant cohorts, Medicaid cohort

eTable 18. Full list of characteristics of the amphetamine/dextroamphetamine exposed and amphetamine/dextroamphetamine discontinuer pregnant cohorts, MarketScan cohort

eTable 19. Full list of characteristics of the methylphenidate exposed and methylphenidate discontinuer pregnant cohorts, Medicaid cohort

eTable 20. Full list of characteristics of the methylphenidate exposed and methylphenidate discontinuer pregnant cohorts, MarketScan cohort

Cohorts with unexposed reference group

eTable 21. Full list of characteristics of the late pregnancy stimulant exposed and unexposed pregnant cohorts, Medicaid cohort

eTable 22. Full list of characteristics of the late pregnancy stimulant exposed and unexposed pregnant cohorts, MarketScan cohort

eTable 23. Full list of characteristics of the early pregnancy stimulant exposed and unexposed pregnant cohorts, Medicaid cohort

eTable 24. Full list of characteristics of the early pregnancy stimulant exposed and unexposed pregnant cohorts, MarketScan cohort

eFigure 1. Flowchart of inclusion in the cohort

eFigure 2. Cumulative incidence of ASD by late pregnancy stimulant exposure and data source

eFigure 3. Cumulative incidence of ADHD by late pregnancy stimulant exposure and data source

eFigure 4. Hazard ratio and 95% CI results comparing stimulant exposed to unexposed children, requiring 2 fills of the medication of interest to be classified as exposed

eFigure 5. Hazard ratio and 95% CI results comparing stimulant exposed to unexposed children, restricting the population to children with at least 2 years of enrollment after birth

eFigure 6. Hazard ratios and 95% CI intervals for neurodevelopmental disorders among children exposed to amphetamine/dextroamphetamine in pregnancy, confounding adjustment using overlap weights

eFigure 7. Hazard ratios and 95% CI intervals for neurodevelopmental disorders among children exposed to methylphenidate in pregnancy, confounding adjustment using overlap weights

jamapsychiatry-e235073-s001.pdf^{(1.3MB, pdf)}

Supplement 2.

Data sharing statement

jamapsychiatry-e235073-s002.pdf^{(6.5KB, pdf)}

References

1.Fayyad J, Sampson NA, Hwang I, et al. ; WHO World Mental Health Survey Collaborators . The descriptive epidemiology of DSM-IV adult ADHD in the World Health Organization World Mental Health Surveys. Atten Defic Hyperact Disord. 2017;9(1):47-65. doi: 10.1007/s12402-016-0208-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Song P, Zha M, Yang Q, Zhang Y, Li X, Rudan I. The prevalence of adult attention-deficit hyperactivity disorder: a global systematic review and meta-analysis. J Glob Health. 2021;11:04009. doi: 10.7189/jogh.11.04009 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Zhu Y, Liu W, Li Y, Wang X, Winterstein AG. Prevalence of ADHD in publicly insured adults. J Atten Disord. 2018;22(2):182-190. doi: 10.1177/1087054717698815 [DOI] [PubMed] [Google Scholar]
4.Knight TK, Kawatkar A, Hodgkins P, et al. Prevalence and incidence of adult attention deficit/hyperactivity disorder in a large managed care population. Curr Med Res Opin. 2014;30(7):1291-1299. doi: 10.1185/03007995.2014.901940 [DOI] [PubMed] [Google Scholar]
5.Montejano L, Sasané R, Hodgkins P, Russo L, Huse D.. Adult ADHD: prevalence of diagnosis in a US population with employer health insurance. Curr Med Res Opin. 2011;27(sup2):5-11. doi: 10.1185/03007995.2011.603302 [DOI] [PubMed] [Google Scholar]
6.Chung W, Jiang SF, Paksarian D, et al. Trends in the prevalence and incidence of attention-deficit/hyperactivity disorder among adults and children of different racial and ethnic groups. JAMA Netw Open. 2019;2(11):e1914344. doi: 10.1001/jamanetworkopen.2019.14344 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Fairman KA, Peckham AM, Sclar DA. Diagnosis and treatment of ADHD in the United States: update by gender and race. J Atten Disord. 2020;24(1):10-19. doi: 10.1177/1087054716688534 [DOI] [PubMed] [Google Scholar]
8.Raman SR, Man KKC, Bahmanyar S, et al. Trends in attention-deficit hyperactivity disorder medication use: a retrospective observational study using population-based databases. Lancet Psychiatry. 2018;5(10):824-835. doi: 10.1016/S2215-0366(18)30293-1 [DOI] [PubMed] [Google Scholar]
9.Anderson KN, Ailes EC, Danielson M, et al. Attention-deficit/hyperactivity disorder medication prescription claims among privately insured women aged 15-44 years - United States, 2003-2015. MMWR Morb Mortal Wkly Rep. 2018;67(2):66-70. doi: 10.15585/mmwr.mm6702a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Louik C, Kerr S, Kelley KE, Mitchell AA. Increasing use of ADHD medications in pregnancy. Pharmacoepidemiol Drug Saf. 2015;24(2):218-220. doi: 10.1002/pds.3742 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Schein J, Childress A, Adams J, et al. Treatment patterns among adults with attention-deficit/hyperactivity disorder in the United States: a retrospective claims study. Curr Med Res Opin. 2021;37(11):2007-2014. doi: 10.1080/03007995.2021.1968814 [DOI] [PubMed] [Google Scholar]
12.Smith LM, Diaz S, LaGasse LL, et al. Developmental and behavioral consequences of prenatal methamphetamine exposure: a review of the Infant Development, Environment, and Lifestyle (IDEAL) study. Neurotoxicol Teratol. 2015;51:35-44. doi: 10.1016/j.ntt.2015.07.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Kalaitzopoulos DR, Chatzistergiou K, Amylidi AL, Kokkinidis DG, Goulis DG. Effect of methamphetamine hydrochloride on pregnancy outcome: a systematic review and meta-analysis. J Addict Med. 2018;12(3):220-226. doi: 10.1097/ADM.0000000000000391 [DOI] [PubMed] [Google Scholar]
14.Li L, Sujan AC, Butwicka A, et al. Associations of prescribed ADHD medication in pregnancy with pregnancy-related and offspring outcomes: a systematic review. CNS Drugs. 2020;34(7):731-747. doi: 10.1007/s40263-020-00728-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Shah R, Diaz SD, Arria A, et al. Prenatal methamphetamine exposure and short-term maternal and infant medical outcomes. Am J Perinatol. 2012;29(5):391-400. doi: 10.1055/s-0032-1304818 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Smith LM, LaGasse LL, Derauf C, et al. Motor and cognitive outcomes through three years of age in children exposed to prenatal methamphetamine. Neurotoxicol Teratol. 2011;33(1):176-184. doi: 10.1016/j.ntt.2010.10.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.LaGasse LL, Derauf C, Smith LM, et al. Prenatal methamphetamine exposure and childhood behavior problems at 3 and 5 years of age. Pediatrics. 2012;129(4):681-688. doi: 10.1542/peds.2011-2209 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Eze N, Smith LM, LaGasse LL, et al. School-aged outcomes following prenatal methamphetamine exposure: 7.5-year follow-up from the infant development, environment, and lifestyle study. J Pediatr. 2016;170:34-8.e1. doi: 10.1016/j.jpeds.2015.11.070 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Lemelin M, Sheehy O, Zhao JP, Bérard A. Maternal ADHD medication use during pregnancy and the risk of ADHD in children: importance of genetic predispositions and impact of using a sibling analysis. Eur Neuropsychopharmacol. 2021;44:66-78. doi: 10.1016/j.euroneuro.2021.01.003 [DOI] [PubMed] [Google Scholar]
20.Bang Madsen K, Robakis TK, Liu X, et al. In utero exposure to ADHD medication and long-term offspring outcomes. Mol Psychiatry. 2023;28(4):1739-1746. doi: 10.1038/s41380-023-01992-6 [DOI] [PubMed] [Google Scholar]
21.Palmsten K, Huybrechts KF, Mogun H, et al. Harnessing the Medicaid Analytic eXtract (MAX) to evaluate medications in pregnancy: design considerations. PLoS One. 2013;8(6):e67405-e67405. doi: 10.1371/journal.pone.0067405 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Tau GZ, Peterson BS. Normal development of brain circuits. Neuropsychopharmacology. 2010;35(1):147-168. doi: 10.1038/npp.2009.115 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Straub L, Bateman BT, Hernandez-Diaz S, et al. Validity of claims-based algorithms to identify neurodevelopmental disorders in children. Pharmacoepidemiol Drug Saf. 2021;30(12):1635-1642. doi: 10.1002/pds.5369 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Straub L, Bateman BT, Hernandez-Diaz S, et al. Neurodevelopmental disorders among publicly or privately insured children in the United States. JAMA Psychiatry. 2022;79(3):232-242. doi: 10.1001/jamapsychiatry.2021.3815 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Harris E. Autism prevalence has been on the rise in the US for decades—and that’s progress. JAMA. 2023;329(20):1724-1726. doi: 10.1001/jama.2023.6078 [DOI] [PubMed] [Google Scholar]
26.Yuan J, Li M, Lu ZK. Racial/ethnic disparities in the prevalence and trends of autism spectrum disorder in US children and adolescents. JAMA Netw Open. 2021;4(3):e210771. doi: 10.1001/jamanetworkopen.2021.0771 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Shi Y, Hunter Guevara LR, Dykhoff HJ, et al. Racial disparities in diagnosis of attention-deficit/hyperactivity disorder in a US national birth cohort. JAMA Netw Open. 2021;4(3):e210321. doi: 10.1001/jamanetworkopen.2021.0321 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.US Department of Agriculture’s Economic Research Service . County-level data sets. Accessed March 18, 2020. Available from: https://www.ers.usda.gov/data-products/county-level-data-sets/
29.Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med. 2009;28(25):3083-3107. doi: 10.1002/sim.3697 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Desai RJ, Rothman KJ, Bateman BT, Hernandez-Diaz S, Huybrechts KF. A propensity-score-based fine stratification approach for confounding adjustment when exposure is infrequent. Epidemiology. 2017;28(2):249-257. doi: 10.1097/EDE.0000000000000595 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Wasserstein RL, Lazar NA. The ASA statement on P values: context, process, and purpose. Am Stat. 2016;70(2):129-133. doi: 10.1080/00031305.2016.1154108 [DOI] [Google Scholar]
32.Li F, Thomas LE, Li F. Addressing extreme propensity scores via the overlap weights. Am J Epidemiol. 2019;188(1):250-257. [DOI] [PubMed] [Google Scholar]
33.Desai RJ, Franklin JM. Alternative approaches for confounding adjustment in observational studies using weighting based on the propensity score: a primer for practitioners. BMJ. 2019:367:l5657. doi: 10.1136/bmj.l5657 [DOI] [PubMed] [Google Scholar]
34.Chubak J, Pocobelli G, Weiss NS. Tradeoffs between accuracy measures for electronic health care data algorithms. J Clin Epidemiol. 2012;65(3):343-349.e2. doi: 10.1016/j.jclinepi.2011.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Straub L, Hernández-Díaz S, Bateman BT, et al. Association of antipsychotic drug exposure in pregnancy with risk of neurodevelopmental disorders: a national birth cohort study. JAMA Intern Med. 2022;182(5):522-533. doi: 10.1001/jamainternmed.2022.0375 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Suarez EA, Bateman BT, Hernández-Díaz S, et al. Association of antidepressant use during pregnancy with risk of neurodevelopmental disorders in children. JAMA Intern Med. 2022;182(11):1149-1160. doi: 10.1001/jamainternmed.2022.4268 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials