Methods to establish a Pregnancy Register in the QResearch Database.

Area of Science:

• Epidemiology
• Health Informatics
• Reproductive Health

Background:

• Electronic health records (EHRs) are crucial for pregnancy research due to underrepresentation of pregnant individuals.
• Combining primary and secondary care data enhances pregnancy research capabilities.
• This study details an algorithm for a unified pregnancy cohort within the QResearch database during the COVID-19 pandemic's emergency phase.

Purpose of the Study:

• To develop and describe an algorithm for creating a unified pregnancy cohort in the QResearch database.
• To facilitate pregnancy-related research by consolidating data from primary and secondary care.
• To establish a comprehensive resource for studying pregnancy outcomes and events during a specific period.

Main Methods:

• Linked national primary care records (QResearch®) with Hospital Episode Statistics (HES) data.
• Included females aged 15-50 with recorded pregnancy outcomes between December 2020 and September 2022.
• Employed a three-stage algorithm to identify pregnancy episodes and derive start dates using combined or individual data sources.

Main Results:

• The register includes 266,758 women and 279,027 pregnancies.
• 83.4% of pregnancies resulted in delivery (99.5% livebirths), and 16.6% were pregnancy losses.
• Pregnancy loss rates varied by ethnicity, highest in Caribbean (23.1%) and lowest in Pakistani (13.9%) groups.
• Data sources included HES maternity records (82.4%), primary care (10.6%), HES Admissions (3.4%), and HES Procedures (3.6%).

Conclusions:

• The QResearch® pregnancy register is a valuable resource for future research.
• The developed methodology is adaptable for constructing cohorts across different time periods.
• This approach provides a comprehensive dataset on pregnancy outcomes and events.