# Optimising gravimetric method interpretation using population-specific blood densities among Indonesian pregnant women: A cross-sectional study

## TL;DR

This study measured blood density in Indonesian pregnant women to improve the accuracy of gravimetric methods for estimating postpartum blood loss.

## Contribution

The study provides the first population-specific blood density data for Indonesian pregnant women to optimize gravimetric blood loss estimation.

## Key findings

- The mean blood density among Indonesian pregnant women was 1.060 ± 0.080 g/ml.
- Blood density showed no significant correlation with maternal age, pre-pregnancy BMI, or gestational weight gain.
- The observed variability in blood density suggests the need for further investigation into non-demographic factors affecting it.

## Abstract

The diagnosis of postpartum haemorrhage (PPH) depends heavily on accurate and timely blood loss assessment, yet no gold standard assessment method is available. Gravimetric methods are recommended for low-resource settings, but they require an accurate blood density value for interpretation. Currently, there are no data on the blood density of pregnant or labour women.

This study measured the blood density of Indonesian pregnant women to improve the reliability of the gravimetric results in estimating postpartum blood loss.

This cross-sectional study was conducted at a Private Midwifery Service Clinic in Bogor, Indonesia, involving 44 pregnant women who were at 37 weeks of gestation or later. During antenatal visits, 5 ml venous blood samples were collected, with women experiencing high-risk pregnancies being excluded. Descriptive statistics summarised the characteristics of the participants. Spearman’s rank-order correlation was used to assess associations between study variables, and the Kruskal–Wallis test evaluated group differences.

The mean blood density was 1.060 ± 0.080 g/ml (range: 0.860–1.194 g/ml), with stable indoor and outdoor temperatures of 27 °C (SD = 1.13 °C and 1.22 °C, respectively). No significant correlations were found between blood density and maternal age, pre-pregnancy BMI, or gestational weight gain (p > 0.05). Kruskal–Wallis tests also revealed no significant differences across these variables (p > 0.05).

Blood density among Indonesian pregnant women corresponds with established human reference values, supporting the use of population-specific blood density values to improve gravimetric blood loss estimation accuracy, thus facilitating early detection and prompt intervention of PPH. The observed variability suggests that factors other than demographics, maternal anthropometry, and the environment may affect blood density, necessitating further investigation in larger multisite studies.

This study was reported in accordance with the STROBE checklist; the completed checklist is available as Supporting Information (S1 File).

This study employed a cross-sectional design to measure blood densities of pregnant women at a single point in time. Participants were recruited during antenatal visits at Dyan’s Private Midwifery-Service Clinic, Gunung Putri, Bogor, West Java, Indonesia. The health facility is a primary-level facility led by midwives and serves as the community’s first point of contact for maternal and child health services. The health facility offers routine maternal health services, including antenatal care, childbirth assistance, postpartum follow-up, and family planning. As a primary-level facility, this health facility only provided care for women with low-risk pregnancies following the midwife’s authority, while women with high-risk pregnancies or obstetrical/pathological conditions were referred to referral or obstetrician-led care facilities. The facility, situated in Gunung Putri, an industrial area adjacent to the capital city (Jakarta), serves a stable yet demographically diverse population drawn from various provinces and socioeconomic backgrounds, offering heterogeneous and broadly representative of Indonesia’s diverse demographics. Although not fully representative of rural or coastal populations, the site offers a scientifically appropriate environment for studying blood density among pregnant Indonesian women. Prior to the commencement of the study, the participating facility provided permission for data collection. Data collection began on 8th January 2025 and ended on 21st April 2025.

This study included pregnant women 37 weeks pregnant or older (based on the first day of the last menstrual cycle). To measure blood density, 5 ml venous blood samples were collected from all eligible participants. Pregnant women at high risk of pregnancy complications were not eligible to participate in the study. To ensure appropriate representation of the target population in this study, the sample size was determined using the Slovin formula by considering the total number of antenatal visits at Dyan’s Private Midwifery Service Clinic between November 2023 and October 2024, which was 2,576. With an error tolerance of 0.15, a sample of 44 pregnant women was deemed sufficient to balance statistical precision with study feasibility. Given the exploratory, cross-sectional nature of the work, which focused on estimating blood density rather than group comparisons or hypothesis testing, this margin of error provided an acceptable level of precision. Accordingly, a sample size of 44 was adequate to support the optimisation of the gravimetric method for estimating postpartum blood loss. A consecutive sampling method was used to recruit participants. All pregnant women at the study site who met the predefined inclusion criteria were invited to participate in the study. Recruitment continued until the target sample size was achieved. As a preliminary cross-sectional study to estimate blood density and not population-level prevalence, consecutive sampling was considered suitable. All eligible participants were enrolled during the study period to minimise selection bias.