Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis
Open accessThe DOI matches its Crossref record; 50 of the 50 references it cites were found in a registry or an index; 8 of the 28 sources its claims cite were checked for support.
Abstract
Objectives: Compare the safety of antiepileptic drugs (AEDs) on neurodevelopment of infants/children exposed in utero or during breast feeding. Design and setting: Systematic review and Bayesian random-effects network meta-analysis (NMA). MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials were searched until 27 April 2017. Screening, data abstraction and quality appraisal were completed in duplicate by independent reviewers. Participants: 29 cohort studies including 5100 infants/children. Interventions: Monotherapy and polytherapy AEDs including first-generation (carbamazepine, clobazam, clonazepam, ethosuximide, phenobarbital, phenytoin, primidone, valproate) and newer-generation (gabapentin, lamotrigine, levetiracetam, oxcarbazepine, topiramate, vigabatrin) AEDs. Epileptic women who did not receive AEDs during pregnancy or breast feeding served as the control group. Primary and secondary outcome measures: Cognitive developmental delay and autism/dyspraxia were primary outcomes. Attention-deficit hyperactivity disorder, language delay, neonatal seizures, psychomotor developmental delay and social impairment were secondary outcomes. Results: The NMA on cognitive developmental delay (11 cohort studies, 933 children, 18 treatments) suggested that among all AEDs only valproate was statistically significantly associated with more children experiencing cognitive developmental delay compared with control (OR=7.40, 95% credible interval (CrI) 3.00 to 18.46). The NMA on autism (5 cohort studies, 2551 children, 12 treatments) suggested that oxcarbazepine (OR 13.51, CrI 1.28 to 221.40), valproate (OR 17.29, 95% CrI 2.40 to 217.60), lamotrigine (OR 8.88, CrI 1.28 to 112.00) and lamotrigine+valproate (OR 132.70, CrI 7.41 to 3851.00) were associated with significantly greater odds of developing autism compared with control. The NMA on psychomotor developmental delay (11 cohort studies, 1145 children, 18 treatments) found that valproate (OR 4.16, CrI 2.04 to 8.75) and carbamazepine+phenobarbital+valproate (OR 19.12, CrI 1.49 to 337.50) were associated with significantly greater odds of psychomotor delay compared with control. Conclusions: Valproate alone or combined with another AED is associated with the greatest odds of adverse neurodevelopmental outcomes compared with control. Oxcarbazepine and lamotrigine were associated with increased occurrence of autism. Counselling is advised for women considering pregnancy to tailor the safest regimen. Trial registration number: PROSPERO database (CRD42014008925).
Overview
CoChat Summary
TL;DR
This systematic review and Bayesian network meta-analysis of 29 cohort studies (5100 children) found that valproate alone or combined with another AED was associated with the greatest odds of adverse neurodevelopmental outcomes. Oxcarbazepine and lamotrigine were also associated with increased autism/dyspraxia.
Key findings
- Cognitive developmental delay NMA (11 cohort studies, 933 children, 18 treatments): only valproate was significantly associated with more children experiencing cognitive developmental delay versus control (OR 7.40, 95% CrI 3.00 to 18.46).
- Autism/dyspraxia NMA (5 cohort studies, 2551 children, 12 treatments): oxcarbazepine (OR 13.51, CrI 1.28 to 221.40), valproate (OR 17.29, 95% CrI 2.40 to 217.60), lamotrigine (OR 8.88, CrI 1.28 to 112.00) and lamotrigine+valproate (OR 132.70, CrI 7.41 to 3851.00) were associated with significantly greater odds of autism/dyspraxia versus control.
- Psychomotor developmental delay NMA (11 cohort studies, 1145 children, 18 treatments): valproate (OR 4.16, CrI 2.04 to 8.75) and carbamazepine+phenobarbital+valproate (OR 19.12, CrI 1.49 to 337.50) were associated with significantly greater odds versus control.
- Language delay NMA (5 cohort studies, 509 children, 5 treatments): valproate was the only treatment significantly associated with increased odds (OR 7.95, 95% CrI 1.50 to 49.13). ADHD NMA (5 cohort studies, 816 children, 7 treatments) found no treatment comparisons reached statistical significance.
Methods
- Systematic review and Bayesian random-effects network meta-analysis; MEDLINE, EMBASE and Cochrane Central Register of Controlled Trials searched until 27 April 2017; screening, data abstraction and quality appraisal completed in duplicate; registered with PROSPERO (CRD42014008925).
- Included 29 cohort studies (5100 infants/children) of mothers consuming AEDs during pregnancy and/or breastfeeding; monotherapy and polytherapy first-generation and newer-generation AEDs were eligible; control was epileptic women who did not receive AEDs; primary outcomes were cognitive developmental delay and autism/dyspraxia, with secondary outcomes including ADHD, language delay, neonatal seizures, psychomotor developmental delay and social impairment.
- Used ORs for dichotomous outcomes, hierarchical meta-analysis and NMA models in a Bayesian framework (OpenBUGS), random-effects models, transitivity and consistency assessments, network meta-regression and sensitivity analyses, and SUCRA rankings.
Limitations
- Stated in the paper: drug dosages could not be accounted for, and different dosages of the same AED were assumed equally effective; when a study reported multiple dosages for the same treatment, data were combined.
- Stated in the paper: several polytherapies had high SUCRA estimates but very wide credible intervals due to small numbers of studies and underpowered sample sizes; ranking probabilities may be biased toward treatments with the smallest number of studies.
- Stated in the paper: no RCT evidence was available, so conclusions were based on observational studies only, with inherent confounding and shortcomings; important effect modifiers were often unreported (eg, family history of autism, ADHD, maternal IQ, severity of epilepsy), and subgroup analysis by exposure type (breastfeeding vs in utero) could not be conducted because 22 studies did not report breastfeeding exposure. Evident from the text: many NMAs included small numbers of studies/children, credible intervals were wide, and methodological quality was imbalanced across treatment comparisons.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures



Verification
Claims the paper makes
5 Valid · 0 Not valid · 13 Not checked · 67 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“AED use during pregnancy is associated with risks to the fetus as these drugs can cross the placenta or may be transferred to the infant through breast feeding and may be associated with adverse neurodevelopment outcomes.2–4”
§ Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis > Introduction
What the sources say
- Partly supportedcites [2] sourceExperimental — accuracy not yet measured
The Conclusions and Clinical context sections report that many AEDs cross the placenta or enter breast milk. But Recommendations for Future Research says studies did not systematically evaluate outcomes such as cognition and neonatal development. This source therefore supports the exposure pathways, not an association with adverse neurodevelopmental outcomes. (LLM, conf=0.86)
- Could not verifycites [3] sourceExperimental — accuracy not yet measured
The Abstract and Results report an association between prenatal AED exposure and major congenital malformations, not adverse neurodevelopmental outcomes. A search for cognitive or developmental outcomes found no assessment of those outcomes in our copy; congenital malformations alone do not establish the attributed neurodevelopmental association. (LLM, conf=0.76)
- Could not verifycites [4] sourceExperimental — accuracy not yet measured
The Introduction and Discussion address congenital malformations and other pregnancy outcomes associated with in-utero AED exposure, but do not report neurodevelopmental outcomes in children. Searches for neurodevelopment, cognition, and developmental delay found no passage in the paper's substantive text that establishes the claimed association. (LLM, conf=0.76)
hedgedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, there is a lack of sufficiently powered studies to assess the impact of AEDs on neurodevelopment in children of women exposed to these agents, especially for newer-generation drugs, thus highlighting the need for a systematic review.9 10”
§ Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis > Introduction
What the sources say
- Partly supportedcites [9] Wlodarczyk, 2012, Antiepileptic drugs and pregnancy outcomes, Am J Med Genet A sourceExperimental — accuracy not yet measured
The Introduction (page unavailable) says most studies of AED teratogenic effects are too small and underpowered, and that risks of newer drugs remain unknown because of small sample sizes and polytherapy exposure. It also discusses neurodevelopmental deficits after valproate exposure, but the cited passages do not specifically establish a lack of sufficiently powered studies of children’s neurodevelopment, particularly for newer-generation AEDs. (LLM, conf=0.84)
- Supportedcites [10] Velez-Ruiz, 2015, Neurodevelopmental effects of fetal antiepileptic drug exposure, Drug Saf source
The Human Studies limitations section says some studies had small samples and that large studies are needed to assess less frequent outcomes. The Conclusions section says the cognitive and behavioral risks of fetal exposure to other AEDs remain unknown because adequate human data are lacking, supporting the particular concern about newer drugs. The cited paper does report established findings for some AEDs, so the claim is best read as a statement about gaps in the evidence, not a complete absence of studies. (LLM, conf=0.88)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Our results are supported by findings from a cohort study, which found that children exposed to levetiracetam were not at increased risk for delayed development compared with unexposed children (see online supplementary appendix C).42”
§ Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis > Discussion
What the sources say
- Supportedcites [42] source
The Abstract (Conclusion) of our copy states directly that children exposed to LEV in utero are not at increased risk of delayed development, and the Results (p. ~7, passage on overall developmental ability) show LEV-exposed children did not differ from the control group (p = 0.62), with 8% below average vs 12% in controls. The control group is defined in Methods as children born to women without epilepsy and not exposed to medication in pregnancy (n = 97), i.e. unexposed children, and the study is described as a prospective observational cohort recruited from 2 UK cohorts, matching 'cohort study'. The only qualifier the citing sentence drops is the source's age/domain bound ('early cognitive development under the age of 24 months'), which does not alter the attributed comparison. (LLM, conf=0.92)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The NMA on cognitive developmental delay (11 cohort studies, 933 children, 18 treatments) suggested that among all AEDs only valproate was statistically significantly associated with more children experiencing cognitive developmental delay compared with control (OR=7.40, 95% credible interval (CrI) 3.00 to 18.46).”
- Also stated in: § Cognitive developmental delay · plain
“Across all AEDs, only valproate was associated with significantly increased odds of cognitive developmental delay compared with control (OR 7.40, 95% CrI 3.00 to 18.46; figure 2A; see online supplementary appendix H).”
§ Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis > Abstract > Results
plainstates an associationn = 933- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
- Also stated in: § Cognitive developmental delay · plain
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“29 cohort studies including 5100 infants/children.”
- Also stated in: § Study and patient characteristics · plain
“We included 29 cohort studies (5100 patients) published between 1989 and 2016”
- Also stated in: § Trial registration number · plain
“Twenty-nine cohort studies involving 5100 children of women who took antiepileptic drugs (AEDs) were included in this systematic review.”
§ Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis > Abstract > Participants
plainn = 5,100- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Study and patient characteristics · plain
40 more claims
Create a free account to see all 85 claims and to verify the 20 references still open.
- Probably not a substantial claim. 19
- It may state two separate claims as one. 24
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Important is also that the anti-epileptic drugs, given at young age or during pregnancy, can have major neurodevelopmental implications as well.”
From Development of prefrontal cortex (2022) · cites this paper as [393]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“reporting that only valproate was associated with more children experiencing cognitive developmental delay compared with controls (OR 7.40, 95% CI 3.00–18.46).”
From The contribution of environmental exposure to the etiology of autism spectrum disorder (2019) · cites this paper as [128]
In the citing paper
“In a subset of studies examining autism risk (5 cohort studies, 2551 children, 12 treatments), this risk was amplified (OR 17.29, 95% CI 2.40–217.60) [128].”
From The contribution of environmental exposure to the etiology of autism spectrum disorder (2019) · cites this paper as [128]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“| Cognitive developmental delay | 12 | 58.62 |”
§ Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis > Results > Study and patient characteristics
Reported 58.62 · recomputed 41.37931
Citations
Cites
Cochrane Handbook for Systematic Reviews of Interventions: Cochrane Book Series
(2008) · doi:10.1002/9780470712184 · Cited by 25,048 · cited 1×
“Strengths of our study include a comprehensive systematic review methodology that followed the Cochrane Handbook 44 and ISPOR 12 guidelines, and reported using the PRISMA extension for NMA.”
MethodDermyshi, E., Wang, Y., Yan, C., Hong, W., Qiu, G., Gong, X., & Zhang, T. (2017) · Neonatology · doi:10.1159/000454668 · Cited by 24,451
(2014) · Medical Entomology and Zoology · Cited by 17,633
Hutton, B., Salanti, G., Caldwell, D. M., Chaimani, A., Schmid, C. H., Cameron, C., Ioannidis, J. P. A., Straus, S., Thorlund, K., Jansen, J. P., et al. (2015) · Annals of Internal Medicine · doi:10.7326/m14-2385 · Cited by 8,269 · cited 3×
“Strengths of our study include a comprehensive systematic review methodology that followed the Cochrane Handbook(44) and ISPOR(12) guidelines, and reported using the PRISMA extension for NMA.(13) To the best of our knowledge, our study was the first that compared and ranked the safety of AEDs.”
MethodSelf-citationPRESS Peer Review of Electronic Search Strategies: 2015 Guideline Statement
McGowan, J., Sampson, M., Salzwedel, D. M., Cogo, E., Foerster, V., & Lefebvre, C. (2016) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2016.01.021 · Cited by 5,700 · cited 1×
“The search strategy for MEDLINE was peer-reviewed by another librarian using the Peer Review of Electronic Search Strategies (PRESS) checklist 15 and is available in the protocol.”
MethodSelf-citationSalanti, G., Ades, A. E., & Ioannidis, J. P. A. (2011) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2010.03.016 · Cited by 3,766
Publication bias in clinical research
Easterbrook, P. J., Gopalan, R., Berlin, J. A., & Matthews, D. R. (1991) · The Lancet · doi:10.1016/0140-6736(91)90201-y · Cited by 3,109 · cited 1×
“Fifth, although the tendency towards small-study effects is greater with observational studies than with randomised trials, 50 the assessment of small-study effects using adjusted funnel plots suggested no evidence for their prevalence.”
BackgroundInterpretation of random effects meta-analyses
Riley, R. D., Higgins, J. P. T., & Deeks, J. J. (2011) · BMJ · doi:10.1136/bmj.d549 · Cited by 2,884
A Re-Evaluation of Random-Effects Meta-Analysis
Higgins, J. P. T., Thompson, S. G., & Spiegelhalter, D. J. (2009) · Journal of the Royal Statistical Society Series A: Statistics in Society · doi:10.1111/j.1467-985x.2008.00552.x · Cited by 2,569 · cited 1×
“28 29 The estimated safety of the included AEDs was ranked using the surface under the cumulative ranking (SUCRA) curve.”
MethodGraphical tools for network meta-analysis in STATA
Chaimani, A., Higgins, J. P. T., Mavridis, D., Spyridonos, P., & Salanti, G. (2013) · PLoS ONE · doi:10.1371/journal.pone.0076654 · Cited by 2,473
Research works printed but not linked: 2
- Bromley R , Baxter N , Calderbank R , et al . A comprehensive review of the language abilities of children exposed to valproate or carbamazepine in utero. Texas: American Epilepsy Society, 2010.
- Gogatishvili N , Ediberidze T , Lomidze G , et al . Cognitive outcomes of children with fetal antiepileptic drug exposure at the age of 3-6 years-preliminary data. 1st 624 Congress of the European Academy of Neurology. Berlin: European Journal of Neurology, 2015:p.329.
Cited by
Barnes, T. R., Drake, R., Paton, C., Cooper, S. J., Deakin, B., Ferrier, I. N., Gregory, C. J., Haddad, P. M., Howes, O. D., Jones, I., et al. (2020) · Journal of Psychopharmacology · doi:10.1177/0269881119889296 · Cited by 606
The contribution of environmental exposure to the etiology of autism spectrum disorder
Bölte, S., Girdler, S., & Marschik, P. B. (2019) · Cellular and Molecular Life Sciences · doi:10.1007/s00018-018-2988-4 · Cited by 509
Development of prefrontal cortex
Kolk, S. M., & Rakic, P. (2022) · Neuropsychopharmacology · doi:10.1038/s41386-021-01137-9 · Cited by 378
Bjørk, M.-H., Zoega, H., Leinonen, M. K., Cohen, J. M., Dreier, J. W., Furu, K., Gilhus, N. E., Gissler, M., Hálfdánarson, Ó., Igland, J., et al. (2022) · JAMA Neurology · doi:10.1001/jamaneurol.2022.1269 · Cited by 267
30 years of second-generation antiseizure medications: impact and future perspectives
Perucca, E., Brodie, M. J., Kwan, P., & Tomson, T. (2020) · The Lancet Neurology · doi:10.1016/s1474-4422(20)30035-1 · Cited by 261
Metadata
- Authors
- Veroniki, A. A., Rios, P., Cogo, E., Straus, S. E., Finkelstein, Y., Kealey, R., Reynen, E., Soobiah, C., Thavorn, K., Hutton, B., Hemmelgarn, B. R., Yazdi, F., D'Souza, J., MacDonald, H., & Tricco, A. C.
- Affiliations
- St. Michael's Hospital (CA)
- University of Toronto (CA)
- University of Ottawa (CA)
- University of Calgary (CA)
- Corresponding author
- Tricco, Andrea C
- ORCID
- Venue
- BMJ Open
- Publisher
- BMJ
- Published
- 2017-07
- Volume
- 7
- Issue
- 7
- Pages
- e017248
- ISSN
- 2044-6055 (ISSN-L)
- PMC
- PMC5642793
- OpenAlex
- W2738349700
- Semantic Scholar
- 1927e1fe6c2f75e9ef202e803e3ecae9efc82d24
- Cited by
- 222
- References
- 50
- Citations by year
- 2026: 7 · 2025: 22 · 2024: 22 · 2023: 14 · 2022: 41 · 2021: 29 · 2020: 47 · 2019: 26 · 2018: 12 · 2017: 1
- FWCI
- 4.41
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by-nc
- Locations
- BMJ Open (publishedVersion, cc-by-nc)
- PubMed Central (publishedVersion, cc-by-nc)
- PubMed (publishedVersion)
- HighWire Press Open Archive (submittedVersion)
- Funders
- CIHR/DSEN New Investigator Award in Knowledge Synthesis
- Tier 1 Canada Research Chair in Knowledge Translation
- Tier 2 Canada Research Chair in Knowledge Synthesis
- Canadian Institutes for Health Research/Drug Safety and Effectiveness Network
- Banting Postdoctoral Fellowship Program from the CIHR
- Alberta Heritage Foundation for Medical Research
- Topics
- Health Sciences › Medicine › Pediatrics, Perinatology and Child Health › Pharmacological Effects and Toxicity Studies
- Health Sciences › Medicine › Clinical Neurology › Epilepsy research and treatment
- Health Sciences › Medicine › Pediatrics, Perinatology and Child Health › Fetal and Pediatric Neurological Disorders
- Keywords
- antiepileptic drugs, prenatal drug exposure, breastfeeding, developmental neurotoxicity, autism spectrum disorder, cognitive delay, valproic acid, lamotrigine, oxcarbazepine, attention deficit hyperactivity disorder, neonatal seizures, epilepsy in pregnancy, drug safety comparison
- MeSH terms
- Lamotrigine, Oxcarbazepine, Anticonvulsants, Autistic Disorder, Bayes Theorem, Breast Feeding, Carbamazepine, Child, Epilepsy, Female, Humans, Pregnancy, Pregnancy Complications, Prenatal Exposure Delayed Effects, Triazines, Valproic Acid, Observational Studies as Topic
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis. https://app.cochat.ai/papers/10.1136/bmjopen-2017-017248
- The paper
- Veroniki, A. A., Rios, P., Cogo, E., Straus, S. E., Finkelstein, Y., Kealey, R., Reynen, E., Soobiah, C., Thavorn, K., Hutton, B., Hemmelgarn, B. R., Yazdi, F., D'Souza, J., MacDonald, H., & Tricco, A. C. (2017). Comparative safety of antiepileptic drugs for neurological development in children exposed during pregnancy and breast feeding: a systematic review and network meta-analysis. BMJ Open. https://doi.org/10.1136/bmjopen-2017-017248
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.