Atypical antipsychotics (Second generation)

Study	Type of data	Exposure measurement	Outcome assessment	Adjustment
Anderson, 2020	case control	Mothers were invited to participate in an English or Spanish computer-assisted telephone interview 6 weeks to 24 months after delivery to report information about over-the-counter and prescription medication use (start/stop dates, duration, and frequency) before and during pregnancy.	Clinical data of cases were abstracted from medical records to confirm that birth defect(s) met eligibility criteria, and clinical geneticists and other expert clinicians then classified cases into homogeneous defect categories. Data of controls abstracted from hospital birth logs or vital records.	No adjustment. Exclusion of pre-pregnancy diabetes and anticonvulsivant exposure in early pregnancy.
Bellet - Aripiprazole, 2015	prospective cohort	Interview of mother/clinician using structured questionnaires at initial telephone contact and after birth.	Data ascertainment was performed using standardised questions at initial telephone contact and structured questionnaires after birth.	Unexposed patients were matched for age (± 2 years) and gestational age at first call (± 2 weeks). According to the available data, none of them had pre-existing diabetes. Exclusion of co-exposure to teratogens during embryogenesis. Exclusion of multiple births for growth restriction.
Boden - Olanzapine and/or clozapine, 2012	population based cohort retrospective	The Swedish Prescribed Drug Register contains information on all prescriptions filled in Sweden, including the dispensed sub- stances’ Anatomical Therapeutic Chemical code and the amount formulation, and dates the substance was prescribed and dispensed.	Diagnosis recorded in the Medical Birth Register. The information is obtained by midwives and attending physicians in connection with visits and hospitalizations from the antenatal visit through the neonatal period.	Singleton only. As potential confounders, are included maternal country of origin, smoking, height, and cohabitation status at the first antenatal visit; maternal age when giving birth; and birth order of the infant.
Bruno - Quetiapine, 2024	population based cohort retrospective	Data on filled or reimbursed prescriptions were obtained from the nation prescription registries.	Information on child neurodevelopmental disorders was ascertained from ICD-10 codes recorded in specialist care registers. Poor academic performance in the first national standardized school test administered was assessed using national education registers.	Singletons only. Adjusted for child’s country of birth, year of birth and sex, maternal country of birth, education, age, parity, cohabitation, smoking status, and body mass index (BMI) in early pregnancy, other medication use, known or suspected teratogen use, maternal comorbidity during pregnancy, and psychiatric diagnosis.
Chan (Controls exposed to FGA), 2024	retrospective cohort (claims database)	The exposure data were obtained from the Clinical Data Analysis and Reporting System (CDARS), including prescribing and dispensing records.	The outcomes data were obtained from the Clinical Data Analysis and Reporting System (CDARS), including clinical information: diagnoses, attendances to outpatient clinics and emergency departments, hospital admissions (ICD-9-classification).	Singleton. Exclusion of fetal alcohol syndrome, abnormalities due to maternal infection or exposure to known teratogens. No adjustment for this group of comparison.
Chan (Controls unexposed, general pop), 2024	retrospective cohort (claims database)	The exposure data were obtained from the Clinical Data Analysis and Reporting System (CDARS), including prescribing and dispensing records.	The outcomes data were obtained from the Clinical Data Analysis and Reporting System (CDARS), including clinical information: diagnoses, attendances to outpatient clinics and emergency departments, hospital admissions (ICD-9-classification).	Singleton. Exclusion of fetal alcohol syndrome, abnormalities due to maternal infection or exposure to known teratogens. PS-weighted logistic regression models with age, parity, maternal pre-existing diseases including diabetes, hypertension, epilepsy, physical comorbidity burden, psychiatric disorders, alcohol use disorders, history of postpartum depression / psychosis, other medications...
Chan (Controls unexposed, sick), 2024	retrospective cohort (claims database)	The exposure data were obtained from the Clinical Data Analysis and Reporting System (CDARS), including prescribing and dispensing records.	The outcomes data were obtained from the Clinical Data Analysis and Reporting System (CDARS), including clinical information: diagnoses, attendances to outpatient clinics and emergency departments, hospital admissions (ICD-9-classification).	Singleton. Exclusion of fetal alcohol syndrome, abnormalities due to maternal infection or exposure to known teratogens. PS-weighted logistic regression models with age, parity, maternal pre-existing diseases including diabetes, hypertension, epilepsy, physical comorbidity burden, psychiatric disorders, alcohol use disorders, history of postpartum depression / psychosis, other medications...
Cohen, 2022	prospective cohort	Participants are prospectively interviewed across pregnancy and the early postpartum period by telephone, at enrollment, at 7 months’ gestation, and 12 weeks after delivery.	Data collected from interviews and medical records (from pediatrics and relevant specialists). If a major malformation is suspected, the records are redacted and sent to a dysmorphologist for final blind adjudication. Medical records were obtained and reviewed for 81.3% of the participants.	No adjustment.
Ellfolk (Control exposed to FGA), 2019	population based cohort retrospective	The Finnish Prescription Registry (Kela) contains data on reimbursed prescription drug purchases.	The Medical Birth Register, the Abortion Register and the National Register of Congenital Malformations.	Adjustment for year of delivery and covariates associated with exposure and outcome at p < 0.1.
Ellfolk (Control unexposed), 2019	population based cohort retrospective	The Finnish Prescription Registry (Kela) contains data on reimbursed prescription drug purchases.	The Medical Birth Register, the Abortion Register and the National Register of Congenital Malformations.	Controls in this group were matched for year of birth of child (± 1 month). Adjusted for year of delivery and covariates associated with exposure and outcome at p < 0.1 (among maternal age, parity, body mass index, marital status, smoking, other psychiatric drugs, teratogens, psychosis, pre-gestational diabetes, other chronic disease). Singletons only.
Ellfolk (Controls exposed to FGA), 2021	population based cohort retrospective	Data from the Drugs and Pregnancy database which includes data from the Prescription Register.	Data from the Medical Birth Register, the Abortion Register, the Register of Congenital Malformations (ICD-9 diagnoses).	Matched for year of birth. Adjusted for year of delivery, maternal age at delivery, parity, prepregnancy BMI, cohabitation, smoking, SES, other psychiatric drugs, psychotic and other severe mental disorders, diabetes (pre- and/or gestational). Alcohol use is not routinely collected in the MBR and could therefore not be included in analyses. Exclusion of pregnancies exposed to known teratogens.
Ellfolk (Controls unexposed, NOS), 2021	population based cohort retrospective	Data from the Drugs and Pregnancy database which includes data from the Prescription Register.	Data from the Medical Birth Register, the Abortion Register, the Register of Congenital Malformations (ICD-9 diagnoses).	Matched for year of birth. Adjusted for year of delivery, maternal age at delivery, parity, prepregnancy BMI, cohabitation, smoking, SES, other psychiatric drugs, psychotic and other severe mental disorders, diabetes (pre- and/or gestational). Alcohol use is not routinely collected in the MBR and could therefore not be included in analyses. Exclusion of pregnancies exposed to known teratogens.
Frayne (Control exposed to antidepressants), 2018	retrospective cohort	Medication use was reported at each clinic attendance and recorded throughout the pregnancy. Women were grouped according to the medication prescribed for their mental health requirements in pregnancy, based on shared decision making with the perinatal psychiatrist, not by randomization.	Maternal and neonatal variables collected at hospital (NOS).	Placental weight–to–birth weight ratios adjusted for baby's sex and gestational age. No adjustment for other outcomes. Exclusion of multiple pregnancy, fetal death in utero, or termination of pregnancy and pregnancies with multiple mix of medications.
Frayne (Control unexposed, sick), 2018	retrospective cohort	Medication use was reported at each clinic attendance and recorded throughout the pregnancy. Women were grouped according to the medication prescribed for their mental health requirements in pregnancy, based on shared decision making with the perinatal psychiatrist, not by randomization.	Maternal and neonatal variables collected at hospital.	Placental weight–to–birth weight ratios adjusted for baby's sex and gestational age. No adjustment for other outcomes. Exclusion of multiple pregnancy, fetal death in utero, or termination of pregnancy and pregnancies with multiple mix of medications.
Habermann (Control exposed to FGA), 2013	prospective cohort	Data ascertainment was performed using 2 structured questionnaires at (1) the first contact and (2) 8 weeks after the estimated date of birth. A detailed history of drug use is recorded at the first contact.	Follow up is especially focused on congenital anomalies and postnatal disorders. For this purpose, during interview, the hospital discharge summaries are asked for.	Major malformations: maternal age, alcohol consumption, smoking habits, number of previous spontaneous abortions, number of previous malformed children, and gestational week at delivery were considered as covariates and potential confounders. Exclusion of women exposed to teratogenic or fetotoxic agents.
Habermann (Control unexposed, disease free), 2013	prospective cohort	Data ascertainment was performed using 2 structured questionnaires at (1) the first contact and (2) 8 weeks after the estimated date of birth. A detailed history of drug use is recorded at the first contact.	Follow up is especially focused on congenital anomalies and postnatal disorders. For this purpose, during interview, the hospital discharge summaries are asked for.	Major malformations/post natal disorders: maternal age, alcohol consumption, smoking habits, number of previous spontaneous abortions, number of previous malformed children, and gestational week at delivery were considered as covariates and potential confounders. Exclusion of women exposed to teratogenic or fetotoxic agents.
Halfdanarson (Controls exposed to FGA), 2022	population based cohort retrospective	Medication exposure determined through the nationwide prescription registers and identified antipsychotics according to the Anatomical Therapeutic Chemical (ATC)18 classification group N05A.	All (non-mutually exclusive) diagnoses of ADHD and ASD identified through records in the Nordic health registers, which cover inpatient hospital care, outpatient specialist visits and prescription fills.	No adjustment for this group of comparison.
Halfdanarson (Controls unexposed, NOS), 2022	population based cohort retrospective	Medication exposure determined through the nationwide prescription registers and identified antipsychotics according to the Anatomical Therapeutic Chemical (ATC)18 classification group N05A.	All (non-mutually exclusive) diagnoses of ADHD and ASD identified through records in the Nordic health registers, which cover inpatient hospital care, outpatient specialist visits and prescription fills.	Adjusted for source country, birth year, sex of child, maternal age, parity, maternal education, cohabitation, mother born within source country, BMI, smoking, maternal comorbidity, use of known teratogens, use of suspected teratogens, other medications during pregnancy.
Heinonen a (Controls unexposed, NOS), 2022	population based cohort retrospective	Exposure data were acquired from both the Medical Birth Register (MBR) and the Prescribed Drug Register. The Prescribed Drug Register stores data on drugs prescribed in ambulatory care and dispensed at pharmacies.	The Medical Birth Register (MBR) was used to collect information about maternal and foetal background characteristics as well as the main outcome gestational diabetes (GDM).	Adjusted for age, parity, smoking and Body mass index (BMI). Exclusion of pregnancies with pre-pregnancy diabetes and women using valproic acid during pregnancy. Singletons only.
Heinonen a (Controls unexposed, sick), 2022	population based cohort retrospective	Exposure data were acquired from both the Medical Birth Register (MBR) and the Prescribed Drug Register. The Prescribed Drug Register stores data on drugs prescribed in ambulatory care and dispensed at pharmacies.	The Medical Birth Register (MBR) was used to collect information about maternal and foetal background characteristics as well as the main outcome gestational diabetes (GDM).	Adjusted for age, parity, smoking and Body mass index (BMI). Exclusion of pregnancies with pre-pregnancy diabetes and women using valproic acid during pregnancy.
Heinonen b (Controls exposed to FGA), 2022	population based cohort retrospective	Information on drug exposure was collected from the Swedish Medical Birth Register (MBR), registered in the MBR at the first visit of the complimentary antenatal care offered to all pregnant.	Data on neonatal outcomes were extracted from the Swedish Neonatal Quality Register (SNQ) and the Perinatal Revision South Register (PRS), registered according to the International Classification of Diseases (ICD-10), or as checkboxes in the infant’s medical record.	No adjustment for this group of comparison. Singleton births only. Women with a diagnosis of pre-pregnancy diabetes or with valproate treatment (N03AG01) were excluded from the analysis.
Heinonen b (Controls unexposed, NOS), 2022	population based cohort retrospective	Information on drug exposure was collected from the Swedish Medical Birth Register (MBR), registered in the MBR at the first visit of the complimentary antenatal care offered to all pregnant.	Data on neonatal outcomes were extracted from the Swedish Neonatal Quality Register (SNQ) and the Perinatal Revision South Register (PRS), registered according to the International Classification of Diseases (ICD-10), or as checkboxes in the infant’s medical record.	Adjusted for: primipara, age, body mass index, smoking, caesarean section and concurrent neurotropic drugs. Singleton births only. Women with a diagnosis of pre-pregnancy diabetes or with valproate treatment (N03AG01) were excluded from the analysis.
Heinonen b (Controls unexposed, sick), 2022	population based cohort retrospective	Information on drug exposure was collected from the Swedish Medical Birth Register (MBR), registered in the MBR at the first visit of the complimentary antenatal care offered to all pregnant.	Data on neonatal outcomes were extracted from the Swedish Neonatal Quality Register (SNQ) and the Perinatal Revision South Register (PRS), registered according to the International Classification of Diseases (ICD-10), or as checkboxes in the infant’s medical record.	Adjusted for: primipara, age, body mass index, smoking, caesarean section and concurrent neurotropic drugs. Singleton births only. Women with a diagnosis of pre-pregnancy diabetes or with valproate treatment (N03AG01) were excluded from the analysis.
Hironaka (Control exposed to BZD), 2011	retrospective cohort	The subjects’ medical records were investigated concerning medication history.	Medical records.	None
Hironaka (Control unexposed, disease free), 2011	retrospective cohort	The subjects’ medical records were investigated concerning medication history.	Medical records.	None.
Hironaka (Control unexposed, sick), 2011	retrospective cohort	The subjects’ medical records were investigated concerning medication history.	Medical records.	None.
Huybrechts (Controls unexposed, NOS), 2023	population based cohort retrospective	Exposure to atypical and typical antipsychotics was defined based on filling 1 or more prescriptions of the respective drug class during the first trimester, the period for organogenesis. => Prescription databases.	Nordic Countries: outcomes are defined based on data from the Medical Birth, Malformation and/or Patient Registers from the date of birth to one year after birth. USA: claims in infant record between birth and birth more 90 days and/or in the maternal record, using both in- and out-patient data.	Propensity score approach to control for potential confounders: demographic factors (maternal age..), treatment indications/mental disorders, maternal/obstetrical conditions (hypertension, diabetes, ...), lifestyle behaviors (tobacco, alcohol, ...), other medications, and health care utilization metrics. Exclusion of pregnancies exposed to a known teratogenic medication. Singleton births only.
Huybrechts (Controls unexposed, sick), 2023	population based cohort retrospective	Exposure to atypical and typical antipsychotics was defined based on filling 1 or more prescriptions of the respective drug class during the first trimester, the period for organogenesis. => Prescription databases.	Nordic Countries: outcomes are defined based on data from the Medical Birth, Malformation and/or Patient Registers from the date of birth to one year after birth. USA: claims in infant record between birth and birth more 90 days and/or in the maternal record, using both in- and out-patient data.	Propensity score approach to control for potential confounders: demographic factors (maternal age..), treatment indications/mental disorders, maternal/obstetrical conditions (hypertension, diabetes, ...), lifestyle behaviors (tobacco, alcohol, ...), other medications, and health care utilization metrics. Exclusion of pregnancies exposed to a known teratogenic medication. Singleton births only.
Ishikawa, 2024	nested case control	Exposure were identified according to the WHO-ATC or alternate entries in the administrative claims database from JMDC Inc. (Tokyo, Japan), which contains pharmacy claims received from the insurers.	Birth outcomes was unavailable in the Japanese claims database. Pregnancies and birth outcomes were estimated using two methods: the first one was for those whose pregnancy-and birth outcome-related entries were available, and their pregnancies and birth outcomes were estimated using these entries.	Exclusion of women with recurrent pregnancy loss, antiphospholipid syndrome, medications at risk of miscarriage. Adjusted for indications (schizophrenia, manic episodes, bipolar, depressive or anxiety disorder), uterine diseases (endometriosis, uterus and cervix malformations, ...), maternal comorbidities (polycystic ovarian, diabetes, obesity, or thyroid disorders), alcohol/tobacco dependence,...
Källen, 2013	population based cohort retrospective	The exposure data are either based on midwife interviews from the first antenatal visit (usually during weeks 10–12) or on linkage with a prescribed drug register.	Medical Birth Register (MBR) supplemented with data from the Register of Birth Defects (RCM, previously Register of Congenital Malformations) and from a Hospital Discharge Register (HDR).	Adjustment was made for year of birth, maternal age (5-year class), parity (1–4), smoking in early pregnancy (unknown, none, <10 cigarettes/day, ≥10 cigarettes per day), and BMI (unknown, <18.5, 18.5–24.9, 25–29, 9. 30–34.9, ≥35).
Kananen - Quetiapine, 2023	retrospective cohort	The women self-reported their use of medications at any point during pregnancy via an online questionnaire and midwifes checked the data collected and further reported medications administrated at the Kuopio University Hospital (KUH) during pregnancy and childbirth.	The ICD-10 diagnoses were set by physician and other obstetric and neonatal outcomes were recorded by midwife.	Adjusted for maternal age, pre-pregnancy body mass index, smoking, maternal use of alcohol or illicit drugs, and other psychotropic medication. Multi-fetal pregnancies and women who had hyper- emesis or migraine as an indication of antipsychotics were excluded from the regression models.
Kernizan, 2019	retrospective cohort	Not reported.	Not reported.	None.
Kernizan (Controls exposed to LTG), 2024	retrospective cohort	Analyzed clinical data extracted from the electronic medical records. Maternal information collected included bipolar treatment regimens at the first visit, throughout pregnancy, and at delivery and illicit drug use by the mother during pregnancy.	Pertinent neonatal information was collected from a review of neonatal electronic medical records and included gender, delivery type, birth weight, length, head circumference, Apgar Score at 1 and 5 min, presence and type of anatomical malformation, and neonatal intensive care unit (NICU) admission.	None. No difference was observed between groups (no other details).
Kernizan (Controls unexposed, sick), 2024	retrospective cohort	Analyzed clinical data extracted from the electronic medical records. Maternal information collected included bipolar treatment regimens at the first visit, throughout pregnancy, and at delivery and illicit drug use by the mother during pregnancy.	Pertinent neonatal information was collected from a review of neonatal electronic medical records and included gender, delivery type, birth weight, length, head circumference, Apgar Score at 1 and 5 min, presence and type of anatomical malformation, and neonatal intensive care unit (NICU) admission.	None. No difference was observed between groups (no other details).
Kulkarni - Quetiapine, 2024	prospective cohort	Pregnant women were interviewed and monitored by the research team. With consent, information was also sought from treating clinicians and medical records.	Pregnant women were interviewed and monitored by the research team. With consent, information was also sought from treating clinicians and medical records.	No adjustment. The confounding variables were analyzed using univariate models with just 1 explanatory variable at a time.
Lin (Control exposed to FGA), 2010	population based cohort retrospective	Medical claims data registered in National Health Insurance Research Dataset (NHIRD).	The National birth certificate registry.	No adjustment for this group of comparison. Singletons only.
Lin (Control unexposed, disease free), 2010	population based cohort retrospective	Medical claims data registered in National Health Insurance Research Dataset (NHIRD).	The National birth certificate registry.	No adjustment for this group of comparison. Singletons only.
Lin (Control unexposed, sick), 2010	population based cohort retrospective	Medical claims data registered in National Health Insurance Research Dataset (NHIRD).	The National birth certificate registry.	Adjusted for maternal age, educational level, marital status, hypertension, and diabetes, together with infant's gender and parity, family monthly income, paternal educational level, and parental age difference. Singletons only.
McKenna, 2005	prospective cohort	Exposure declared by women during exposure and data were achieved by sending each general practitioner a detailed questionnaire with questions regarding drug history.	This was achieved by sending each general practitioner a detailed questionnaire with questions regarding pregnancy outcome. Once the questionnaire was completed, permission was requested to receive a report from the physician primarily caring for the infant.	Control group matched for maternal age (plus or minus 2 years) and gestational age at time of call (plus or minus 2 weeks). No statistical difference in terms of reported alcohol use and pre-gestational diabetes.
Newham (Control exposed to FGA), 2008	prospective cohort	When an enquiry is made to the NTIS (by the healthcare professionals ), the drug(s) the mother has been exposed to are recorded at the time of enquiry.	Cases are subsequently followed-up after the estimated delivery date to collect information from clinicians on the outcomes of pregnancies	No adjustment
Newham (Control unexposed, disease free), 2008	prospective cohort	When an enquiry is made to the NTIS (by the healthcare professionals), the drug(s) the mother has been exposed to are recorded at the time of enquiry.	Cases are subsequently followed-up after the estimated delivery date to collect information from clinicians on the outcomes of pregnancies.	No adjustment. Exclusion criteria were if the infant displayed congenital malformations or if maternal diabetes was recorded.
Newport, 2007	prospective cohort	Maternal interview at monthly intervals during pregnancy (at which time maternal use of prescription and nonprescription medications was documented). Laboratory confirmation of maternal compliance with the antipsychotic medication (i.e., detectable maternal plasma concentrations).	Obstetrical outcome data were ascertained from maternal reports and reviews of medical records.	No adjustment for these exposed and unexposed groups.
Ozturk, 2016	retrospective cohort	At the first contact, a detailed patient history form was used to record the following information: maternal demographic data and obstetric history, consanguineous marriage, smoking and alcohol consumption, X-ray and all drug exposures (dose, duration and timing in pregnancy).	After counseling about the risk of drug exposure, the women and their babies were followed up during a 2-year period. Each newborn baby was checked at birth for signs of problems or complications.	None
Panchaud, 2017	prospective cohort	Participants are prospectively interviewed across pregnancy and the early postpartum period by telephone, at enrollment, at 7 months’ gestation, and 12 weeks after delivery.	GDM was defined as a maternal report of a diabetes diagnosis at the baseline pregnancy interview (excluding pre-pregnancy existing DM), 7 months gestation, final postpartum interviews, or extracted from the medical record if the diagnosis occurred after 12 weeks of gestation.	Multivariate analysis: maternal age, marital status, race, employment status, level of education, smoking, primary psychiatric diagnosis. Exclusion of pregestational diabetes. The effect of obesity on GDM was assessed by stratification on body mass index (BMI) categories among the exposed and reference groups.
Park, 2018	retrospective cohort (claims database)	The Medicaid Analytic eXtract, nationwide claims database that contains information on pharmacy dispensing records.	The Medicaid Analytic eXtract, nationwide claims database that contains information on hospitalizations, and outpatient visits.	Adjusted for age, race, and Medicaid eligibility type, psychiatric diagnoses (anxiety, ADHD, bipolar disorder, depression, schizophrenia, ...), comorbidity (hypertension, obesity, and dyslipidemia, ...), other medication use, history of gestational diabetes, and the duration of antipsychotic treatment. Exclusion of women with preexisting diabetes.
Paulus, 2005	prospective cohort	Not specified	Not specified	None.
Paulus - Aripiprazole only, 2013	prospective cohort	Teratology Information Service (TIS) was contacted by physicians and patients after exposure to aripiprazole in the first trimester.	Not specified	None
Peng, 2013	prospective cohort	A detailed questionnaire completed by all enrolled pregnant women prior to delivery. Once the questionnaire was completed, permission was requested to receive a report from the physician and obstetrician.	The Bayley-III was administered by a psychologist who specialized in this scale and assessed neurodevelopment of children for the hospital. The psychologist was not a part of research team and was blinded during all neurobehavioral development assessments.	Each participant with schizophrenia was matched with a control for maternal age (plus or minus 1 year) and equivalent education level (determined by graduation diploma). Singleton pregnancy at more than 38 weeks. All expectant mothers were excluded from the study if they had evidence of liver or renal dysfunction, diabetes mellitus, or cardiovascular diseases during their pregnancy.
Petersen (Control discontinued treatment before pregnancy), 2016	retrospective cohort (claims database)	Prescriptions are issued electronically and directly recorded on the general practice computer systems.	Antenatal and postnatal care records.	Atypical antipsychotics (subgroup analysis): no adjustment.
Petersen (Control exposed to FGA), 2016	retrospective cohort (claims database)	Prescriptions are issued electronically and directly recorded on the general practice computer systems.	Antenatal and postnatal care records.	Atypical antipsychotics (subgroup analysis): no adjustment.
Petersen (Control unexposed, disease free), 2016	retrospective cohort (claims database)	Prescriptions are issued electronically and directly recorded on the general practice computer systems.	Antenatal and postnatal care records.	Atypical antipsychotics (subgroup analysis): no adjustment.
Raguideau, 2017	retrospective cohort (claims database)	The French national health insurance database (DCIR) containing all individualized and anonymous health care claims reimbursed by French National Health Insurance.	The French hospital discharge database (PMSI) covering the entire French population.	No adjustment for the subgroup of atypical antipsychotics.
Reis (Control exposed to FGA), 2008	population based cohort retrospective	Maternal drug use in early pregnancy is recorded from interviews performed by the midwife at the first antenatal care visit, usually before the end of the first trimester.	3 national health registers: the Medical Birth Register, the Swedish Register of Congenital Malformations, and the Hospital Discharge Register.	No adjustment for this subgroup of exposure.
Reis (Control unexposed, NOS), 2008	population based cohort retrospective	Maternal drug use in early pregnancy is recorded from interviews performed by the midwife at the first antenatal care visit, usually before the end of the first trimester.	3 national health registers: the Medical Birth Register, the Swedish Register of Congenital Malformations, and the Hospital Discharge Register.	No adjustment for this subgroup of exposure.
Sadowski, 2013	prospective cohort	Telephone interviews of mothers.	Information provided by the mothers and data obtained from physicians (request of a report from the child’s physician, which included hospital birth records, postnatal assessments and information on congenital malformations and the child’s health).	Matched for age at conception (±3 years) and pregnancy duration at the initial time of contact (±2 weeks). Exclusion of fertility-assisted pregnancies, twin/triplet pregnancies, pregnancies exposed to teratogenic medications unrelated to their psychiatric disorder treatment, such as acutane, or who abused substances (eg, alcohol, marijuana, cocaine, heroin, etc).
Schaffer (Controls unexposed, NOS), 2019	retrospective cohort (claims database)	The Pharmaceutical Benefits Scheme (PBS) data.	The records from the New South Wales (NSW) Perinatal Data Collection, the NSW Admitted Patient Data Collection, and the NSW Registry of Births, Deaths and Marriages.	None.
Schaffer (Controls unexposed, sick), 2019	retrospective cohort (claims database)	The Pharmaceutical Benefits Scheme (PBS) data.	The records from the New South Wales (NSW) Perinatal Data Collection, the NSW Admitted Patient Data Collection, and the NSW Registry of Births, Deaths and Marriages.	None. Exclusion of pregnancies among women who had a previous birth in the pre-conception period, as a pregnancy or birth during this time period may affect antipsychotic use.
Sorensen - Olanzapine, 2015	population based cohort retrospective	Any prescription recorded in the Danish National Prescription Register.	Any spontaneous abortion (< 22 weeks) or stillbirth recorded in the Danish National Hospital Register and the Danish Medical Birth Register respectively.	Adjusted for maternal age divided into three groups and history of drug abuse.
Straub, 2022	retrospective cohort (claims database)	The nationwide Medicaid Analytic eXtract (MAX) and the IBM Health MarketScan Research Database (MarketScan). Both sources included outpatient medication dispensings.	The nationwide Medicaid Analytic eXtract (MAX) and the IBM Health MarketScan Research Database (MarketScan). Both sources included diagnosis and procedure claims during hospitalizations, outpatient and emergency department visits.	Adjusted for maternal age, race and ethnicity, treatment indications, smoking, alcohol dependance, Substance use disorder, Proxy for severity of Mental Health-related illness, Other prescription medication exposure, pregestational diabetes and hypertension, hyperemesis, Obstetric Comorbidity Index, County-level socioeconomic status measures.
Sutter-Dallay, 2015	retrospective cohort	Information on psychotropic drug use during pregnancy was collected from maternal recall and medical records.	Information about the course of the pregnancy and infant postnatal health collected from the mother and from the child's health record.	All the models were adjusted a priori for maternal age, education level (< 12 years vs at least 12 years), parity (primiparity vs multiparity), presence of partner, maternal diagnosis and type of unit.
Swetlik, 2024	-
Vial (Control exposed to FGA), 2009	prospective cohort	Not specified.	Not specified.	Patients from the control group were matched to those of the atypical antipsychotic group based on maternal age (2 years) and gestational age at the time of the first request (2 weeks).
Vial (Control unexposed, NOS), 2009	prospective cohort	Not specified	Not specified	Patients from the control group were matched to those of the atypical antipsychotic group based on maternal age (2 years) and gestational age at the time of the first request (2 weeks).
Vigod, 2015	retrospective cohort (claims database)	Ontario Drug Benefit (ODB) database.	Maternal medical outcomes identified in Canadian Institutes of Health Information Discharge Abstract Database (CIHI-DAD). Main perinatal outcomes identified using ICES’ MOMBABY datafile.	Control matched by means of a high dimensional propensity score (HDPs) algorithm (notably on maternal age, pre-pregnancy diabetes, chronic hypertension, psychiatric disorders, Alcohol or substance disorder (including smoking), ...) and adjusted for a prescribed selective serotonin reuptake inhibitor (SSRI), non-SSRI, mood stabiliser, or benzodiazepine during the index pregnancy. Singletons only.
Viguera b, 2023	prospective cohort	Participants are prospectively interviewed twice during pregnancy (at enrollment and at 7 months’ gestation) and once postpartum (at 3 months following delivery). Information collected includes all medication use (including prescriptions, over- the-counter medications, and supplements) and dosages.	Labor and delivery, neonatal, and pediatric medical records were collected. Medical records are reviewed and abstracted using a standardized form developed by the investigators. The neonatal outcomes form also includes a verbatim checklist of withdrawal and extrapyramidal symptoms (EPS) symptoms.	No adjustment. The two groups did not differ significantly in mean maternal age, preterm delivery rates or birth weight.
Wang a (Controls exposed to FGA), 2021	retrospective cohort (claims database)	Prescriptions of any antipsychotic were extracted from the prescribing and dispensing records.	Study outcomes were extracted from the Clinical Data Analysis and Reporting System (CDARS), which contains the electronic health records.	No adjustment for this subgroup of exposure.
Wang a (Controls unexposed, NOS), 2021	retrospective cohort (claims database)	Prescriptions of any antipsychotic were extracted from the prescribing and dispensing records.	Study outcomes were extracted from the Clinical Data Analysis and Reporting System (CDARS), which contains the electronic health records.	No adjustment for this subgroup of exposure.
Wang b - HK cohort, 2021	retrospective cohort (claims database)	Prescriptions of any antipsychotic listed in Chapter 4.2.1 of the British National Formulary (BNF) were extracted from the prescribing and dispensing records.	The Hong Kong Clinical Data Analysis and Reporting System (CDARS) which consists of anonymized electronic health records from HK hospitals and contains 43 hospitals and institutions, 49 specialist outpatient clinics, and 73 general outpatient clinics which is accessible to all HK residents.	Maternal age, calendar year, maternal underlying medical conditions (epilepsy, hypertension and SMI [e.g. schizophrenia, bipolar disorder and depression]), duration of antipsychotics, other psychotropics use, household income status. Primiparous pregnancies only. Exclusion of women with a diagnosis of pre-pregnancy diabetes (Type 1 and Type 2).
Wang b - UK cohort, 2021	retrospective cohort (claims database)	Prescriptions of any antipsychotic listed in Chapter 4.2.1 of the British National Formulary (BNF) were extracted from the prescribing and dispensing records.	The UK The Health Improvement Network (THIN) which consists of anonymized electronic health records from UK primary care. THIN covers medical records of patients registered at 744 participating practices.	Maternal age, calendar year, maternal underlying medical conditions (epilepsy, hypertension and SMI [e.g. schizophrenia, bipolar disorder and depression]), duration of antipsychotics, other psychotropics use, Body Mass Index, smoking and alcohol status and family history of diabetes. Primiparous pregnancies only. Exclusion of women with a diagnosis of pre-pregnancy diabetes (Type 1 and Type 2).
Wurtz, 2017	population based cohort retrospective	Danish Register of Medicinal Product Statistics.	Use of primary health care system in childhood: contacts to the general practitioner (GP) were used as a proxy for the overall health of the children.	Sex and birth date of the child, maternal age, parity, cohabitation status, income, education, smoking status, diagnosis of substance abuse, severe psychiatric disorder, depression and epilepsy as well as the use of antiepileptic drugs, antidepressants, benzodiazepines and insulin.
Yakuwa, 2019	prospective cohort	Information on the characteristics of pregnant women and their medications were collected using interview sheets by questionnaires completed by the women at the time they applied for counseling.	Pregnancy outcomes were collected by questionnaires were sent to pregnant women in one month following the expected date of delivery (EDD).	None.
Yaris, 2005	prospective cohort	Data surveyed by the interviews.	Examinations were made by obstetric ultrasound for the mothers and birth weight and height, gestational age, APGAR scores, congenital malformation and developmental problems for the babies. Periodic checks of all the babies in the neonatal period, were made the first year, and following years.	None for this group of exposure.
Yeh, 2021	retrospective cohort (claims database)	The Taiwan National Health Insurance Research Database (THIRD) provides prescriptions information about insured individuals.	The THIRD provides diagnosis information about insured individuals. Psychiatric disorders were diagnosed based on clinical judgement and interviews by psychiatrists. In this study, the diagnosis of ADHD and ASD required at least twice outpatient visits within each year for these ICD-9 codes.	Adjusted for demographic characteristics (the level of urbanization, monthly income, and other demographics).

master protocol