Reliability of student midwives' visual estimate of blood loss in the immediate postpartum period: A cross-sectional study

Background: In France, postpartum hemorrhage (blood loss >= 500 mL in the first 24 h postpartum) is the leading direct obstetric cause of maternal mortality. In French practice, PPH is mainly diagnosed by a quantitative assessment of blood loss, performed by subjective methods such as visual estimates. Various studies have concluded that visual estimates are imprecise, tend to underestimate blood loss, and thus to delay diagnosis of PPH. Objectives: The principal objective of this study was to assess the accuracy of visual estimates of blood loss by student midwives. The secondary objectives were to study intraobserver agreement of these assessments, to assess the accuracy of visual estimates for threshold values, and to look for a region effect. Design: A cross-sectional multicentre study. Setting: All French midwifery schools (n = 35). Participants: Volunteer French student midwives at their fifth (final) year (n = 463). Methods: The online questionnaire contained 16 photographs (8 different, each presented twice) of simulated volumes of blood loss (100, 150, 200, 300, 500, 850, 1000, and 1500 mL). A 50-mL reference standard for calibration accompanied each photograph. Only one answer could be selected among the 7 choices offered for each photograph. Comparisons used chi(2) and Kappa tests. Results: The participation rate was 48.43% (463/956), and 7.408 visual estimates were collected. Estimates were accurate for 35.34% of the responses. The reproducibility rate for the visual estimates (0.17 <= kappa <= 0.48) and for the accurate visual estimates (0.11 <= kappa <= 0.55) were moderate for 4 of the 8 volumes (100, 300, 1000, and 1500 mL). The percentage of accurate responses was significantly higher for volumes <= 300 mL than for those >= 500 mL (52.94% vs. 17.17%, p < 0.0001) and those >= 1000 mL (52.94% vs. 18.30%, p < 0.0001). The percentage of accurate responses varied between the regions (p = 0.042). Conclusion: Despite the help of a visual aid, both the accuracy and reproducibility of the visual estimates were low. (C) 2015 Elsevier Ltd. All rights reserved.