To find out if a drug is safe to use during pregnancy, we need data – pharmacovigilance data to be exact. Pregnancy pharmacovigilance data is collected by different actors, in different ways. This has, historically, made it difficult to conduct large-scale studies with information from multiple sources. As a result of this, and the challenges of studying the effects of medicines on pregnant women and their unborn children in general, there is a lack of evidence-based medicine safety information available to this group of people. ConcePTION is working to amend this.
Part of the ConcePTION project includes efforts to harmonise data from different sources, using the Common Data Model and Core Data Elements we have developed. In a recent Drug Safety publication, ConcePTION researchers put the model and data elements to the test, using data from 10 different sources, collected in three different contexts, to see if data harmonisation is even possible. And the results are promising!
In their test run, the authors included data from four pregnancy registries, two special programmes for tracking medicine safety, and four teratology information services. They checked the data from these sources against the 51-item Core Data Elements, covering medical history, deliveries, the health of both mother and child, and more.
97 % of the data was a match including 85% with a directly taken variable and 12% with a combination of other variables. Most of the data from these data access providers matched up with the guide we developed. The results were obtained either by direct translation from the source to the Core Data Elements or by combining different pieces of data to fit the guide. It was only for 1 % of the data that definitions did not align, and 2 % was not available.
“This means that getting everyone on the same page in terms of definition and data analysis is an actual possibility, and could become a reality. This would provide better information for women and their healthcare professionals to make informed decisions about medicine use during pregnancy. This is an important step towards realising the ConcePTION end-goal of producing and making available sound pharmacovigilance information available to pregnant and breastfeeding women,” says Guillaume Favre, Work package 2.5 ConcePTION collaborator working as an obstetrician or the Swiss Teratology Information Service (STIS) and Obstetrics unit at Lausanne University hospital and one of the authors of the paper.
By Anna Holm Bodin
Favre, G., Richardson, J.L., Moore, A. et al. Improving Data Collection in Pregnancy Safety Studies: Towards Standardisation of Data Elements in Pregnancy Reports from Public and Private Partners, A Contribution from the ConcePTION Project. Drug Saf (2023). DOI: 10.1007/s40264-023-01384-3
Common Data Model & Core Data Elements
The ConcePTION Common Data Model (CDM) is a structured system designed to efficiently gather information about the safety of medications during pregnancy and breastfeeding. It combines data from different sources, including both existing data and newly collected data. Data Access Providers follow a specific process to put the data into the CDM when a new study is approved, but they don’t permanently store it there. Instead, they take what they need for the study and put it back afterwards. This CDM makes data analysis clear and repeatable while ensuring that local data characteristics are maintained. In essence, it’s a way to ensure high-quality information while respecting the diverse sources of data within the ConcePTION project.
The ConcePTION Core Data Elements (CDE) is a reference framework with 98 key data points, organised into 14 categories. Its purpose is straightforward: to standardize data collection in pregnancy studies, making it faster and easier to provide clear, evidence-based advice about medication safety to expectant mothers. This will benefit both healthcare professionals and pregnant women.