Early detection and a treatment bundle strategy for postpartum haemorrhage: a mixed-methods process evaluation.

Abstract

Background: Postpartum haemorrhage is a leading cause of maternal mortality. A multicountry, cluster-randomised trial (E-MOTIVE) demonstrated a 60% reduction in adverse postpartum haemorrhage outcomes. The E-MOTIVE intervention included early postpartum haemorrhage detection using calibrated blood-collection drapes, followed by a postpartum haemorrhage treatment bundle (ie, uterine massage, oxytocics, tranexamic acid, intravenous fluids, examination and escalation [MOTIVE]), supported by implementation strategies. We report a mixed-methods process evaluation assessing the implementation of the E-MOTIVE intervention in Kenya, Nigeria, South Africa, and Tanzania.

Methods: In this mixed-methods process evaluation, data sources were observations of health workers providing clinical care to pregnant women and pregnant people during vaginal birth and postpartum haemorrhage at intervention sites, and surveys and qualitative interviews with health workers at intervention and control sites. Intervention sites received the calibrated drapes, MOTIVE bundle, and implementation strategies and control sites used uncalibrated drapes. Primary implementation outcomes included fidelity, adoption, adaptation, acceptability, feasibility, and contamination to the calibrated drape, MOTIVE bundle, and implementation strategies.

Findings: Between June 1, 2022, and Jan 31, 2023, 2578 births were observed, 295 pregnant women and people had postpartum haemorrhage, 47 qualitative interviews were done, and 889 surveys were completed. Fidelity to calibrated drape use was high (birth observations 2578 [100%] of 2578; survey 451 [98·3%] of 459). Among health workers, calibrated drape acceptability was high; however, they reported barriers to pregnant women's and people's acceptability. Fidelity to postpartum haemorrhage treatment bundle delivery was high (birth observations 286 [96·9%] of 295), with moderate to high fidelity in median time from postpartum haemorrhage diagnosis to final treatment initiation (≤15 min initiation time in 191 [66·8%] of 295 birth observations, 16-20 min in 42 [14·7%] birth observations), and high acceptability and feasibility. Research midwives participated in clinical assessments after birth and bundle delivery in some sites (mixed fidelity).

Interpretation: This process evaluation shows generally high levels of fidelity, feasibility, and acceptability of the calibrated drape and treatment bundle across evaluation methods and countries. The E-MOTIVE intervention should be included in national policies, with consideration for health workforce, supplies, and medication issues, which might need addressing for successful implementation.

Funding: The Bill and Melinda Gates Foundation and the UNDP-UNFPA-UNICEF-WHO-World Bank Special Programme of Research, Development and Research Training in Human Reproduction, a co-sponsored programme of WHO.

Translation: For the Swahili translation of the abstract see Supplementary Materials section.