Implementation research priorities for addressing the maternal health crisis in the USA: results from a modified Delphi study among researchers

Abstract

Background: Maternal health outcomes in the USA are far worse than in peer nations. Increasing implementation research in maternity care is critical to addressing quality gaps and unwarranted variations in care. Implementation research priorities have not yet been defined or well represented in the plans for maternal health research investments in the USA.

Methods: This descriptive study used a modified Delphi method to solicit and rank research priorities at the intersection of implementation science and maternal health through two sequential web-based surveys. A purposeful, yet broad sample of researchers with relevant subject matter knowledge was identified through searches of published articles and grant databases. The surveys addressed five implementation research areas in maternal health: (1) practices to prioritize for broader implementation, (2) practices to prioritize for de-implementation, (3) research questions about implementation determinants, (4) research questions about implementation strategies, and (5) research questions about methods/measures.

Results: Of 160 eligible researchers, 82 (51.2%) agreed to participate. Participants were predominantly female (90%) and White (75%). Sixty completed at least one of two surveys. The practices that participants prioritized for broader implementation were improved postpartum care, perinatal and postpartum mood disorder screening and management, and standardized management of hypertensive disorders of pregnancy. For de-implementation, practices believed to be most impactful if removed from or reduced in maternity care were cesarean delivery for low-risk patients and routine discontinuation of all psychiatric medications during pregnancy. The top methodological priorities of participants were improving the extent to which implementation science frameworks and measures address equity and developing approaches for involving patients in implementation research.

Conclusions: Through a web-based Delphi exercise, we identified implementation research priorities that researchers consider to have the greatest potential to improve the quality of maternity care in the USA. This study also demonstrates the feasibility of using modified Delphi approaches to engage researchers in setting implementation research priorities within a clinical area.

Keywords: Delphi method; Implementation science; Maternal health; Priority setting; Research priorities.

Fig. 1

Participant flow chart

Fig. 2

a Practices most recommended for implementation, as represented in a bubble chart. Numbered practices in Table 3, section A, correspond to the bubble labels. X-axis = feasibility of routinely implementing this practice in US maternity care settings (scale of 1–3). Y-axis = likelihood that wide implementation of this practice will improve outcomes (scale of 1–3). Bubble size indicates how many survey #2 participants selected the practice in their top 3 practices for implementation. Bubble color indicates the quartile of the likelihood that wide implementation of this practice will reduce disparities in maternity outcomes rating. b Practices most recommended for de-implementation, as represented in a bubble chart. Numbered practices in Table 3, section B, correspond to the bubble labels. X-axis = feasibility of de-implementing this practice in US maternity care settings (scale of 1–3). Y-axis = likelihood that de-implementation of this practice will improve outcomes (scale of 1–3). Bubble size indicates how many survey #2 participants selected the practice in their top 3 practices for de-implementation. Bubble color indicates the quartile of the likelihood that wide implementation of this practice will reduce disparities in maternity outcomes rating