Investigation and Application of a Pregnancy Register based on Electronic Primary Care Data

J Campbell ; (2023) Investigation and Application of a Pregnancy Register based on Electronic Primary Care Data. PhD thesis, London School of Hygiene & Tropical Medicine. DOI: 10.17037/PUBS.04670993
Copy

Background: Electronic Health Records (EHR) are useful for studying pregnancy exposures and the outcomes for mother and child. However, as these data are not collected primarily for research accurately identifying the timing of pregnancies is challenging. Pregnancy episodes can be uncertain either because they have no recorded outcome or they overlap with another pregnancy episode. The aims of this work were to better understand how to handle uncertain pregnancy episodes in EHR data and to apply this knowledge to the question: does having COVID-19 during pregnancy increase the risk of pregnancy loss? Methods: Scenarios were identified potentially explaining why uncertain pregnancy episodes occur. Criteria were established and systematically applied to determine whether episodes had evidence of each scenario. Recommendations on how to handle these episodes were generated. A matched cohort study was conducted using EHR data to examine whether COVID-19 infection during pregnancy is associated with pregnancy loss and to test the implementation of developed recommendations. Results: Evidence found suggests that most uncertain pregnancy episodes are true and current pregnancies for which the data contain valuable information. Utilising EHR data found evidence that women who had COVID-19 during pregnancy had an 18% higher risk of pregnancy loss compared to contemporary controls and a 39% higher risk compared to pre-pandemic controls. Adjustments to the study population to include uncertain pregnancy episodes highlighted the potential risk of exposure misclassification associated with including all uncertain episodes. Blanket decisions to include or exclude uncertain episodes may lead to under ascertainment of pregnancies, biased study populations and errors in analysis such as exposure misclassification. Conclusions: Researchers should consider a tailored approach to utilising uncertain pregnancy episodes dependent on the design and purpose of the study. COVID-19 during pregnancy may increase the risk of pregnancy loss supporting the use of vaccination campaigns to protect pregnant women.


picture_as_pdf
2023_EPH_PhD_Campbell_J.pdf
subject
Accepted Version
Available under Creative Commons: Attribution-NonCommercial-No Derivative Works 4.0

View Download
picture_as_pdf

Accepted Version


Atom BibTeX OpenURL ContextObject in Span Multiline CSV OpenURL ContextObject Dublin Core Dublin Core MPEG-21 DIDL Data Cite XML EndNote HTML Citation JSON MARC (ASCII) MARC (ISO 2709) METS MODS RDF+N3 RDF+N-Triples RDF+XML RIOXX2 XML Reference Manager Refer Simple Metadata ASCII Citation EP3 XML
Export

Downloads