A common data model for harmonization in the Nordic Pregnancy Drug Safety Studies (NorPreSS)

Cohen, Jacqueline M.; Cesta, Carolyn E.; Kjerpeseth, Lars Jøran; Leinonen, Maarit K.; Hálfdánarson, Óskar; Karlstad, Øystein; Karlsson, Pär; Andersen, Morten; Furu, Kari; Hjellvik, Vidar

doi:10.5324/nje.v29i1-2.4053

Cited by 15 publications

(19 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We identified mother-child pairs, pregnancy characteristics, prescription fills, mother and child diagnoses, and demographic and socioeconomic information from the national health and social registers in each country. 17 We harmonized variable definitions across the 5 countries according to a common data model 18 (eTables 1 and 2 in the Supplement).…”

Section: Data Sources Design and Study Cohortmentioning

confidence: 99%

Association of Prenatal Exposure to Antiseizure Medication With Risk of Autism and Intellectual Disability

et al. 2022

Self Cite

View full text Add to dashboard Cite

IMPORTANCEWomen with epilepsy frequently need antiseizure medication (ASM) to prevent seizures in pregnancy. Risk of neurodevelopmental disorders after prenatal exposure to AMSs is uncertain.OBJECTIVE To determine whether children exposed prenatally to ASMs in monotherapy and duotherapy have increased risk of neurodevelopmental disorders. DESIGN, SETTING, AND PARTICIPANTSThe Nordic register-based study of antiepileptic drugs in pregnancy (SCAN-AED) is a population-based cohort study using health register and social register data from Denmark, Finland, Iceland, Norway, and Sweden (1996-2017; analysis performed February 2022). From 4 702 774 alive-born children with available mother-child identities and maternal prescription data, this study included 4 494 926 participants. Children from a multiple pregnancy or with chromosomal disorders or uncertain pregnancy length were excluded (n = 207 848).EXPOSURES Prenatal exposure to ASM determined from maternal prescription fills between last menstrual period and birth. MAIN OUTCOMES AND MEASURESWe estimated cumulative incidence at age 8 years in exposed and unexposed children. Cox regression adjusted for potential confounders yielded adjusted hazard ratios (aHRs) with 95% CIs for autism spectrum disorder (ASD), intellectual disability (ID), or any neurodevelopmental disorder (ASD and/or ID).RESULTS A total of 4 494 926 children were included; 2 306 993 (51.3%) were male, and the median (IQR) age at end of follow-up was 8 (4.0-12.1) years. Among 21 634 unexposed children of mothers with epilepsy, 1.5% had a diagnosis of ASD and 0.8% (numerators were not available because of personal data regulations in Denmark) of ID by age 8 years. In same-aged children of mothers with epilepsy exposed to topiramate and valproate monotherapy, 4.3% and 2.7%, respectively, had ASD, and 3.1% and 2.4% had ID. The aHRs for ASD and ID after topiramate exposure were 2.8 (95% CI, 1.4-5.7) and 3.5 (95% CI, 1.4-8.6), respectively, and after valproate exposure were 2.4 (95% CI, 1.7-3.3) and 2.5 (95% CI, 1.7-3.7). The aHRs were elevated with higher ASM doses compared with children from the general population. The duotherapies levetiracetam with carbamazepine and lamotrigine with topiramate were associated with increased risks of neurodevelopmental disorders in children of women with epilepsy: levetiracetam with carbamazepine: 8-year cumulative incidence, 5.7%; aHR, 3.5; 95% CI, 1.5-8.2; lamotrigine with topiramate: 8-year cumulative incidence, 7.5%; aHR, 2.4; 95% CI, 1.1-4.9. No increased risk was associated with levetiracetam with lamotrigine (8-year cumulative incidence, 1.6%; aHR, 0.9; 95% CI, 0.3-2.5). No consistently increased risks were observed for neurodevelopmental disorders after prenatal exposure to monotherapy with lamotrigine, levetiracetam, carbamazepin, oxcarbazepine, gapapentin, pregabalin, clonazepam, or phenobarbital. CONCLUSIONS AND RELEVANCEIn this cohort study, prenatal exposure to topiramate, valproate, and several duotherapies were associated with increased risks of neurodevel...

show abstract

Section: Data Sources Design and Study Cohortmentioning

confidence: 99%

Association of Prenatal Exposure to Antiseizure Medication With Risk of Autism and Intellectual Disability

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…One example is the OMOP 44,45 ecosystem, comprising both a common data model and tailored analytic tools that enable such federated analyses because everybody knows the format of everybody else's data without ever seeing them. The Nordic data model 46 takes a different approach: data are quality‐checked locally and remain in a reasonable local format, and the analytic code handles harmonization vis‐à‐vis mapping, for example, mapping diagnosis codes to a single concept used to operationalize the outcome of interest.…”

Section: Discussionmentioning

confidence: 99%

Exploratory pharmacovigilance with machine learning in big patient data: A focused scoping review

Kaas‐Hansen

Gentile

Caioli

et al. 2023

Basic Clin Pharma Tox

View full text Add to dashboard Cite

Background Machine learning can operationalize the rich and complex data in electronic patient records for exploratory pharmacovigilance endeavours. Objective The objective of this review is to identify applications of machine learning and big patient data in exploratory pharmacovigilance. Methods We searched PubMed and Embase and included original articles with an exploratory pharmacovigilance purpose, focusing on medicinal interventions and reporting the use of machine learning in electronic patient records with ≥1000 patients collected after market entry. Findings Of 2557 studies screened, seven were included. Those covered six countries and were published between 2015 and 2021. The most prominent machine learning methods were random forests, logistic regressions, and support vector machines. Two studies used artificial neural networks or naive Bayes classifiers. One study used formal concept analysis for association mining, and another used temporal difference learning. Five studies compared several methods against each other. The numbers of patients in most data sets were in the order of thousands; two studies used what can more reasonably be considered big data with >1 000 000 patients records. Conclusion Despite years of great aspirations for combining machine learning and clinical data for exploratory pharmacovigilance, only few studies still seem to deliver somewhat on these expectations.

show abstract

“…It is hypothesised that an optimised and standardised approach to data collection could be an important component in decreasing heterogeneity of outcome reporting, thereby improving the confidence in the synthesis of results from different studies in systematic literature reviews and meta-analysis studies. Furthermore, a standardised approach will facilitate the combination of high-quality and detailed data from multiple sources using a common data model approach [55,56], which in turn would facilitate faster generation of evidence. However, application of the CDE standards will not remove potential data-collection biases inherent to observational study designs, and as such, the associated methodological limitations will need to be continually considered during data synthesis and interpretation stages.…”

Section: Discussionmentioning

confidence: 99%

Core Data Elements for Pregnancy Pharmacovigilance Studies Using Primary Source Data Collection Methods: Recommendations from the IMI ConcePTION Project

et al. 2023

View full text Add to dashboard Cite

Introduction and ObjectiveThe risks and benefits of medication use in pregnancy are typically established through postmarketing observational studies. As there is currently no standardised or systematic approach to the post-marketing assessment of medication safety in pregnancy, data generated through pregnancy pharmacovigilance (PregPV) research can be heterogenous and difficult to interpret. The aim of this article is to describe the development of a reference framework of core data elements (CDEs) for collection in primary source PregPV studies that can be used to standardise data collection procedures and, thereby, improve data harmonisation and evidence synthesis capabilities. Methods This CDE reference framework was developed within the Innovative Medicines Initiative (IMI) ConcePTION project by experts in pharmacovigilance, pharmacoepidemiology, medical statistics, risk-benefit communication, clinical teratology, reproductive toxicology, genetics, obstetrics, paediatrics, and child psychology. The framework was produced through a scoping review of data collection systems used by established PregPV datasets, followed by extensive discussion and debate around the value, definition, and derivation of each data item identified from these systems. ResultsThe finalised listing of CDEs comprises 98 individual data elements, arranged into 14 tables of related fields. These data elements are openly available on the European Network of Teratology Information Services (ENTIS) website (http:// www. entis-org. eu/ cde). Discussion With this set of recommendations, we aim to standardise PregPV primary source data collection processes to improve the speed at which high-quality evidence-based statements can be provided about the safety of medication use in pregnancy.

show abstract

A common data model for harmonization in the Nordic Pregnancy Drug Safety Studies (NorPreSS)

Cited by 15 publications

References 12 publications

Association of Prenatal Exposure to Antiseizure Medication With Risk of Autism and Intellectual Disability

Association of Prenatal Exposure to Antiseizure Medication With Risk of Autism and Intellectual Disability

Exploratory pharmacovigilance with machine learning in big patient data: A focused scoping review

Core Data Elements for Pregnancy Pharmacovigilance Studies Using Primary Source Data Collection Methods: Recommendations from the IMI ConcePTION Project

Contact Info

Product

Resources

About