Act · For Researchers
Where the claims come from, and where the evidence is
still incomplete.
This page is the rigorous front door: derivations, limits and gaps. Every number on this site resolves here, with its study, its denominator and its caveat.
Reading the numbers
How to read the headline numbers
52% missed vs 7% missed: visual-estimation sensitivity 48% (95% CI 44–53) vs calibrated measurement 93% (92–93), pooled in a Cochrane diagnostic test accuracy review; four studies, 196,305 women, and two studies, 53,762 women, against objective reference standards [1]. Caveat: strongest for hospital-based vaginal birth; does not transfer to caesarean or community settings.
RR 0.40 (95% CI 0.32–0.50) for severe postpartum haemorrhage, laparotomy or death: E-MOTIVE, cluster-randomised, 80 hospitals, 210,132 women [2]. Caveat: a system-level effect; components are not isolated, and absolute effect sizes may differ where baseline care differs.
Adjusted OR 2.36 (95% CI 1.60–3.48) for death with severe anaemia: analysis of over 310,000 women, 359 facilities, 29 countries [3]. Caveat: observational; confounding by indication and setting cannot be excluded.
The 300 mL + one vital sign trigger: sensitivity 87–88%, specificity 67–76%: WHO individual participant data meta-analysis, 312,151 women, 23 countries [4]. Caveat: criteria deliberately prioritise sensitivity; expect more false positives by design.
One chain
One evidence chain
Definition, measurement, bundle, and anaemia are linked steps that form a single pathway. The trigger is based on objective measurement, the bundle depends on that trigger, and anaemia influences the blood loss a woman can withstand. They are mentioned together because this pathway serves as the primary unit of evidence. [1] [2] [4].
The open questions
The open questions
1
Implementation of the updated definition across contexts and case mixes.
2
Objective measurement methods for caesarean, community and low-resource births (E-MASTER addresses the first) [5].
4
Which E-MOTIVE components are essential, and which are context-dependent [2].
5
Fidelity and time-based indicators as the next generation of quality metrics [6].
6
Long-term recovery: anaemia, psychological morbidity, fertility, subsequent pregnancy outcomes [6].
Reporting standards
Reporting standards
Comparability remains the weakest aspect of the field. Each postpartum haemorrhage study should clearly specify, at a minimum, the case definition, measurement approach, timing window, mode of birth, and vital sign criteria. Without these details, estimates cannot be reliably pooled, compared, or cited. Additionally, reviews should acknowledge when such information is missing. [1] [6].
References
[1] Yunas I, Sindhu KN, Devall AJ, et al. Diagnostic accuracy of subjective visual estimation and calibrated blood-collection drapes for postpartum haemorrhage. Cochrane diagnostic test accuracy review. Paper
[2] Gallos I, Devall A, Martin J, et al. Randomized Trial of Early Detection and Treatment of Postpartum Hemorrhage (E-MOTIVE). New England Journal of Medicine, 2023;389:11–21. Paper · DOI
[3] Gallos ID, Sindhu KN, Yunas I, et al. Prevention of postpartum haemorrhage: from evidence to implementation at scale. The Lancet, 2026. Paper · DOI
[4] Gallos I, et al. Prognostic accuracy of clinical markers of postpartum bleeding in predicting maternal mortality or severe morbidity: a WHO individual participant data meta-analysis. The Lancet, 2025. Paper · DOI
[5] University of Oxford, Nuffield Department of Women’s and Reproductive Health. E-MASTER Phase I study. Study page
