Common data elements for maternal health research: Expert consensus emerging from a modified Delphi study

Common data elements (CDEs) are critical for providing consistency in data collection and promoting collaboration across research endeavors. This study aims to develop a minimum set of priority CDEs for maternal health research.

The Eunice Kennedy Shriver National Institute of Child Health and Human Development convened a panel of experts, including representatives from research centers, professional associations, and government agencies, to advise on CDEs for maternal health research. A landscape analysis identified constructs, such as health conditions, health behaviors, or patient attributes, commonly measured in maternal health research. Using a modified Delphi technique, panelists prioritized which constructs were important to standardize and then rated the validity and feasibility of CDE candidates for each priority construct.

Delphi panelists prioritized 36 biomedical and 22 psychosocial constructs for standardized data collection. A literature review identified 204 CDE candidates (89 biomedical and 115 psychosocial) for measuring priority constructs. Delphi panelists rated their validity and feasibility and, subsequently, endorsed top-rated CDEs. CDE recommendations include expert-endorsed, standardized measures for maternal health conditions, outcomes, medical care, health history, behavioral health, social factors, and demographics, with guidance on data sources, measurement methods, and ethical considerations.

Through an expert consensus-building process, our study produced recommendations for harmonizing data collection of a minimum set of priority constructs for maternal health research. These CDEs are expected to help improve data quality and interoperability, fostering opportunities to synthesize information and advance maternal health science.