- Medication and Pregnancy KB

Study	Type of data	Exposure measurement	Outcome assessment	Adjustment
Agosti, 2015	retrospective cohort	Not specified.	An ad-hoc created questionnaire was submitted to women who were followed during pregnancy by a multidisciplinary team (rheumatologist, gynaecologist, neonatologist). Data on birth, lactation, weaning, growth parameters, developmental milestones, vaccinations and disease conditions were collected	None
Broms (Controls exposed to other treatments), 2020	population based cohort retrospective	The national prescribed drug registers. Information was also identified from the National Patient Register (Denmark), from the register for biologic treatment (Finland) and from the ARTIS and PsoReg registers (Sweden).	Medical birth registers and patient registers. The registers prospectively record information on pregnancy, delivery and the neonatal period.	Adjustment 1: for potential confounding by country, maternal age, parity, smoking, BMI, and maternal disease (CD, UC, RA, AS, psoriatic arthritis, and psoriasis). Adjustment 2: for Denmark and Sweden, adjusted for previous surgeries, eg, bowel surgery for IBD and prosthetic hip or knee surgery for rheumatologic diseases, and disease-related hospitaliza- tions during pregnancy.
Broms (Controls unexposed, disease free), 2020	population based cohort retrospective	The national prescribed drug registers. Information was also identified from the National Patient Register (Denmark), from the register for biologic treatment (Finland) and from the ARTIS and PsoReg registers (Sweden).	Medical birth registers and patient registers. The registers prospectively record information on pregnancy, delivery and the neonatal period.	No adjustment for this group of comparison.
Bröms a (Controls unexposed, disease free), 2016	population based cohort retrospective	The national registers on prescribed drugs. Anti-TNF treatment also identified from visits recorded in the patient register covering all Danish hospitals using a specific treatment code (Denmark) or from the ARTIS and PsoReg registers (Sweden).	The national medical birth registers and patient registers.	None
Bröms a (Controls unexposed, sick), 2016	population based cohort retrospective	The national registers on prescribed drugs. Anti-TNF treatment also identified from visits recorded in the patient register covering all Danish hospitals using a specific treatment code (Denmark) or from the ARTIS and PsoReg registers (Sweden).	The national medical birth registers and patient registers.	Adjusted for maternal age, parity, smoking, body mass index, multiple gestation, country, and chronic inflammatory diagnosis.
Bröms b, 2016	case control	Information on drug treatment was obtained from the Prescribed Drug Register, the Medical Birth Register and from medical charts. Exposure: 90 days before pregnancy or in first trimester.	Review of medical charts of cases and controls from departments of obstetrics, gastroenterology and internal medicine nationwide for clinical data for the period starting from three months before pregnancy until delivery.	Cases were matched 1:1 by maternal age, parity and IBD diagnosis (CD or UC); and preterm OR adjusted for previous preterm birth, any comorbidity (for anti-TNF association).
Carman - Etanercept (Controls unexposed, disease free), 2017	retrospective cohort (claims database)	The ETN‐exposure was defined as having at least one administration or pharmacy dispensing during pregnancy. Provider‐administered drug exposures were identified using procedure codes for injections or infusions pharmacy dispensing claims.	Infant outcomes were identified using claims‐based procedure or ICD‐9 diagnosis codes for: Low Birth Weight; prematurity and malformations. Medical records of all Optum‐affiliated infants with MCM claims, and their mothers, were subsequently sought for expert adjudication of the claims.	Matched by maternal age and year of pregnancy end. No adjustment for this group of exposure.
Carman - Etanercept (Controls unexposed, sick), 2017	retrospective cohort (claims database)	The ETN‐exposure was defined as having at least one administration or pharmacy dispensing during pregnancy. Provider‐administered drug exposures were identified using procedure codes for injections or infusions pharmacy dispensing claims.	Infant outcomes were identified using claims‐based procedure or ICD‐9 diagnosis codes for: Low Birth Weight; prematurity and malformations. Medical records of all Optum‐affiliated infants with MCM claims, and their mothers, were subsequently sought for expert adjudication of the claims.	No adjustment for this group of exposure.
Casanova, 2013	retrospective cohort	Data were obtained from the review of medical records and an interview with the patient when additional information was necessary was conducted. Collected data including data on drugs received during 3 months before conception and during pregnancy.	Data were obtained from the review of medical records and an interview with the patient when additional information was necessary was conducted. Collected data including data on delivery, pregnancy complications and neonatal complications.	In the multivariate analysis, the dependent variable was the GPO and the independent variables were: the type of IBD, IBD surgeries before pregnancy, maternal age at conception, consumption of toxic substances during pregnancy, IBD activity at the moment of conception, activity of disease during pregnancy, exposure to thiopurines, and exposure to anti-TNF-α drugs.
Chambers - Adalimumab (Controls unexposed, disease free), 2017	prospective cohort	Not specified. Women in the USA or Canada were enrolled in pregnancy prior to 20 weeks’ gestation.	Not specified	Not specified. Logistic regression or Cox Proportional Hazards with propensity score adjustment for confounding were used to estimate adjusted Odds Ratios (aOR) or Hazard Ratios (aHR).
Chambers - Adalimumab (Controls unexposed, sick, ADA only), 2017	prospective cohort	Not specified.	Not specified	Not specified. Logistic regression or Cox Proportional Hazards with propensity score adjustment for confounding were used to estimate adjusted Odds Ratios (aOR) or Hazard Ratios (aHR).
Chambers - Etanercept, 2015	prospective cohort	Women were followed up with multiple telephone interviews and medical record review.	Women were followed up with multiple telephone interviews and medical record review. Development was evaluated with the Ages and Stages Questionnaire (ASQ)	Outcomes were compared using multivariable regression and survival methods.
Chaparro, 2018	retrospective cohort	Practitioners specialized in IBD identified women who received or did not receive anti-TNFα drugs for IBD during their pregnancy. Data (including medical treatment during conception, pregnancy...) were obtained from the medical records and the mothers were asked in case of missing data.	Data (including complications during pregnancy and delivery, newborn complications, infections...) were obtained from the medical records at the participating center and medical reports on the delivery. In addition, the mothers provided information of the children’s admissions to hospital.	Categorical variables: none. Severe infection: Cox regression analysis was performed to identify independent predictors variables considered clinically relevant (e.g., type of IBD, maternal age at conception, consumption of toxics, IBD activity, low birth weight, prematurity, and exposure to anti- TNFα drugs) and those which reached statistical significance in the univariate analysis.
Cooper, 2014	retrospective cohort (claims database)	The Tennessee Medicaid and Kaiser Permanente Northern California and Southern California claims databases.	Outcomes were identified from vital records and medical records: birth certificate check-boxes, infant claims data, fetal death or death certificates, records of maternal hospitalization. Medical records and autopsy reports were reviewed to confirm all adverse fetal outcomes.	Propensity score including maternal age, race, ethnicity, years of education, parity, chronic health diagnoses (smoking, diabetes, epilepsy, asthma, renal disease, cancer, heart disease, hypertension, HIV, substance use, obesity, headaches), medications for chronic diseases, chronic immune mediated diseases... Singletons only. Exclusion of births with exposure to teratogenic medications.
De Lima, 2018	prospective cohort	Data on medication were collected before pregnancy and during each visit at our IBD outpatient clinic, bi-monthly during pregnancy by an experienced IBD physician.	Birth outcomes (birth weight, gestational age and the presence of congenital abnormalities were obtained from mothers during the first visit after delivery. One year after birth, health outcomes were obtained from the general practitioner.	None
De Lima (Controls unexposed, disease free), 2016	prospective cohort	At this specialised outpatient clinic, an experienced gastroenterologist counsels and treats patients with IBD before pregnancy and bimonthly during pregnancy. At every visit, data on medication use was recorded.	One-year follow-up of the child was obtained through telephonic questionnaires with the mothers and by obtaining the medical information of the children from the general practitioner.	Controls matched by maternal age and ethnicity.
De Lima (Treatment stopped during pregnancy, sick), 2016	prospective cohort	At this specialised outpatient clinic, an experienced gastroenterologist counsels and treats patients with IBD before pregnancy and bimonthly during pregnancy. At every visit, data on medication use was recorded.	One-year follow-up of the child was obtained through telephonic questionnaires with the mothers and by obtaining the medical information of the children from the general practitioner.	Controls matched by maternal age and ethnicity
De Lorenzo (Controls unexposed, disease free), 2020	retrospective cohort	Data were collected retrospectively during paediatric consultations.	Data were collected retrospectively during paediatric consultations.	None
De Lorenzo (Controls unexposed, sick), 2020	retrospective cohort	Data were collected retrospectively during paediatric consultations.	Data were collected retrospectively during paediatric consultations.	None
Desai, 2017	retrospective cohort (claims database)	Pharmacy dispensing databases of health insurances.	The outcome was identified using discharge diagnosis codes from hospital admission records. To identify these infections we used a previously validated algorithm, which is reported to have a high positive predictive value (80%) in administrative claims.	Propensity scores. Potential confounding factors patient demographic variables, insurance program, diagnosis of the specific autoimmune inflammatory conditions, other medical conditions and medication use, healthcare utilization, medical conditions, drug misuse or dependence, recorded obesity or smoking, medicines use, diagnosis of serious bacterial or opportunistic infections.
Diav-Citrin (Controls exposed to other treatments), 2014	prospective cohort	Details of exposure were collected during pregnancy, at the initial contact with the TIS and before pregnancy outcome was known, using a structured questionnaire.	Follow-up was conducted by a telephone interview with the woman to obtain details on the pregnancy outcome, gestational age at delivery, birth weight, congenital anomalies and neonatal complications. In cases of anomalies an attempt to obtain medical records was made.	None
Diav-Citrin (Controls unexposed, disease free), 2014	prospective cohort	Details of exposure were collected during pregnancy, at the initial contact with the TIS and before pregnancy outcome was known, using a structured questionnaire.	Follow-up was conducted by a telephone interview with the woman to obtain details on the pregnancy outcome, gestational age at delivery, birth weight, congenital anomalies and neonatal complications. In cases of anomalies an attempt to obtain medical records was made.	None
Drechsel, 2020	prospective cohort	Patients were asked to participate in a structured telephone interview about their pregnancies. Patients were contacted at the first notification of pregnancy.	By patient interview, data on pregnancy and outcomes were collected. Whenever complications, hospitalizations, child medical findings were reported, confirmation was obtained from the attending physicians or with the child a medical report.	None
Duricova, 2019	retrospective cohort	Information on IBD women, including type of diagnosis, details of anti-TNF-alpha treatment, and concomitant medication, was obtained from the patients’ medical files.	A structured questionnaire was used for data collection and was filled out by the treating pediatrician and the mother of each child in both the exposed and unexposed control groups. Data collected: perinatal period, psychomotor development, vaccination, infections, antibiotics, allergy, ...	The control group was intended to be matched 1:1 with exposed children by calendar year of birth. 2 outcomes adjusted (allergy and infections) for for preterm birth (<37 gestational weeks), low birth weight (<2500 g), breastfeeding, length of follow-up, and family history of allergy in children’s parents.
Fu - Etanercept, 2019	randomized controlled trial	The patients were randomized assigned to the two arms of the study, by means of a computer-generated randomization number sequence.	All the patients were followed up during the pregnancy, and every 2 weeks, they underwent transvaginal ultrasound scans from the 4th through 12th GW to observe embryo or to diagnose miscarriage. All the babies born underwent a paediatric examination to exclude congenital malformations.	No adjustment. Randomisation.
Hoxha, 2017	prospective cohort	A form including information on biological agents exposure, the concomitant use of disease modifying antirheumatic drugs was filled out by the treating rheumatologist.	A form including information about the pregnancy and foetal outcomes was filled out by the treating rheumatologist collecting pregnancy complications, congenital malformation and vaccine adverse events. Children’s follow-up was performed through maternal reports during their outpatient follow-up.	None
Komoto, 2016	retrospective cohort	From the medical records, therapy was collected. Each patient was given a questionnaire to answer.	Each patient completed a questionnaire, notably regarding pregnancy outcomes: live birth, spontaneous abortion, birth weight, and presence of congenital abnormalities. To identify the course and outcome of their pregnancies, patients referred to the notebooks, filled out by physicians or nurses.	None.
Langen, 2014	retrospective cohort	Data were collected from review of medical records and included medication use.	Data were collected from review of medical records and included pregnancy outcomes.	None
Lichtenstein - infliximab, 2018	prospective cohort	Patient data were collected at registry enrollment by gastroenterologists and then semi-annually (January, July), including medication use, concomitant CD medication use; and the number of infliximab infusions before, during, and after pregnancy.	Patient data were collected at registry enrollment by gastroenterologists and then semi-annually (January, July), including pregnancy outcomes.	None (No a priori sample size estimates were performed and no adjustment was made for multiple comparisons)
Luu, 2018	retrospective cohort (claims database)	Data were extracted from the SNIIRAM database that includes the database containing all prescriptions for expensive drugs fully reimbursed by the national health insurance and the reimbursement data for out-of-hospital drug purchases with the Presentation Identifer Code codes.	Data for all diagnoses and complications were extracted from the SNIIRAM database that includes data from the hospital discharge abstract database and the procedures performed during hospital stays using the common classification system for medical procedures.	Adjusted for exposure to antiTNF, age at pregnancy, IBD type, IBD disease duration, disease severity, and exposure to thiopurines (and term of delivery, disease severity for infant outcomes).
Moens (Controls exposed to Vedolizumab), 2020	retrospective cohort	Information from Vedolizumab pregnancies was collected through a structured electronic case report form (CRF) that had to be filled out by the treating gastroenterologist, including dose and duration of VDZ therapy, other IBD medications during pregnancy.	Information from Vedolizumab pregnancies was collected through a structured electronic case report form (CRF) that had to be filled out by the treating gastroenterologist (that were requested to contact the patients, general practitioners, paediatricians or other colleagues).	None
Moens (Controls unexposed, sick), 2020	retrospective cohort	Information from Vedolizumab pregnancies was collected through a structured electronic case report form (CRF) that had to be filled out by the treating gastroenterologist, including dose and duration of VDZ therapy, other IBD medications during pregnancy.	Information from Vedolizumab pregnancies was collected through a structured electronic case report form (CRF) that had to be filled out by the treating gastroenterologist (that were requested to contact the patients, general practitioners, paediatricians or other colleagues).	None
Schnitzler (Controls unexposed, disease free), 2011	prospective cohort	Patients were treated with IFX or with ADA in our IBD unit at the University Hospital in Leuven and were prospectively followed.	Data on pregnancy outcome were obtained from hospital files, from family doctors, and from the patients themselves. A detailed questionnaire was also sent out to all patients and/or the patients were intensively interviewed during the contacts in the outpatient clinics.	Matched control group for age at onset of pregnancy and number of pregnancies.
Schnitzler (Unexposed control, sick), 2011	prospective cohort	Patients were treated with IFX or with ADA in our IBD unit at the University Hospital in Leuven and were prospectively followed.	Data on pregnancy outcome were obtained from hospital files, from family doctors, and from the patients themselves. A detailed questionnaire was also sent out to all patients and/or the patients were intensively interviewed during the contacts in the outpatient clinics.	None
Seirafi, 2014	retrospective cohort	Inclusion: patient records from hospital IBD units and patients were contacted by phone when data were missing. During pregnancy, information collected included dose and duration of anti-TNFs and concomitant treatments.	Inclusion: patient records from hospital IBD units and patients were contacted by phone when data were missing. During pregnancy, information collected included pregnancy outcomes. Term and mode of delivery, and newborn characteristics were also recorded.	None
Verstappen, 2011	prospective cohort	Data available at recruitment. Follow-up information (include any changes to treatment) is collected from medical records. All reports of pregnancies are followed up with an additional questionnaire (patient reports) which includes data on exposure to biological agents at the time of conception.	Follow-up information (including pregnancies) is collected from medical records. All reports of pregnancies are followed up with an additional questionnaire (patient reports) which includes details of pregnancy outcome including live births, spontaneous abortions and terminations.	None
Viktil (Controls exposed to other treatments), 2012	population based cohort propective	Prescriptions were recorded from 3 months prior to conception until labour from the Norwegian Prescription Database (NorPD).	The Medical Birth Registry of Norway (MBRN), a population based register that contains information about all births, including late abortion, from 12 weeks of gestation onwards.	None for this group of exposure. Singletons only.
Viktil (Controls unexposed, NOS), 2012	population based cohort propective	Prescriptions were recorded from 3 months prior to conception until labour from the Norwegian Prescription Database (NorPD).	The Medical Birth Registry of Norway (MBRN), a population based register that contains information about all births, including late abortion, from 12 weeks of gestation onwards.	None for this group of exposure. Singletons only.
Vinet, 2013	case control	The RAMQ (Régie de l’assurance maladie du Québec) prescription database.	Induced abortion based either on 1 procedure code and/or diagnostic code for induced abortion present in the MED-E´CHO hospitalization and/or RAMQ billing databases.	Each case was matched to >1 controls for age, calendar time, and cohort entry. Rate ratio adjusted for disease duration.
Vinet (Controls unexposed, disease free), 2018	retrospective cohort	The MarketScan commercial database is a large prospective US database of >230 million subjects containing drug prescription claims.	The MarketScan commercial database, a large prospective US database of >230 million subjects, containing data on hospitalizations and outpatient visits. Serious infections in the offspring was based on ≥1 hospitalization with infection within the first 12 months of life.	A group of women without RA was randomly selected and matched ≥4:1 for age, year of delivery, and geographic location (i.e., state, division, and region) of the primary beneficiary’s residence. We performed multivariable analyses, adjusting for maternal demographics, comorbidities, pregnancy complications, and drugs.
Vinet (Controls unexposed, sick), 2018	retrospective cohort	The MarketScan commercial database is a large prospective US database of >230 million subjects containing drug prescription claims.	The MarketScan commercial database, a large prospective US database of >230 million subjects, containing data on hospitalizations and outpatient visits. Serious infections in the offspring was based on ≥1 hospitalization with infection within the first 12 months of life.	Multivariable analyses, adjusting for maternal demographics, comorbidities, pregnancy complications, and drugs.
Weber-Schoendorfer, 2015	prospective cohort	Data were collected on exposure (including both prescription and over-thecounter) from structured telephone or face-to-face interviews and/or mailed questionnaires obtained from both the mother and/or her physician(s) after oral informed consent.	Data were collected on outcome from structured telephone or face-to-face interviews and/or mailed questionnaires obtained from both the mother and/or her physician(s) after oral informed consent. Blinded classification according EUROCAT.	Propensity score estimation used boosted regression trees, including maternal age, alcohol consumption, smoking status, and numbers of previous pregnancies, miscarriages and previous infants
Zbinden (rheumatoid arthritis), 2018	prospective cohort	All patients were seen at each trimester during pregnancy, with record of the current medication. The data from healthy pregnant women were recorded once: at the last trimester.	At the postpartum visit, information was collected on delivery mode, birth, neonatal health and complications such as preeclampsia, gestational diabetes, infection and preterm premature rupture of membranes.	Age-matched control group. Adjustments made for age and gravidity.
Zbinden b (axial spondyloarthritis), 2018	prospective cohort	All patients were seen at each trimester during pregnancy, with record of the current medication. The data from healthy pregnant women were recorded once: at the last trimester.	At the postpartum visit, information was collected on delivery mode, birth, neonatal health and complications such as preeclampsia, gestational diabetes, infection and preterm premature rupture of membranes.	Age-matched control group. Adjustments made for age and gravidity.

master protocol