Comparison of vaginal birth outcomes in midwifery-led versus physician-led setting: A propensity score-matched analysis

Ingrida Poškienė; Giedrius Vanagas; Asta Kirkilytė; Rūta Jolanta Nadišauskienė

doi:10.1515/med-2021-0373

Comparison of vaginal birth outcomes in midwifery-led versus physician-led setting: A propensity score-matched analysis

Open Med (Wars). 2021 Oct 15;16(1):1537-1543. doi: 10.1515/med-2021-0373. eCollection 2021.

Authors

Ingrida Poškienė¹, Giedrius Vanagas², Asta Kirkilytė¹, Rūta Jolanta Nadišauskienė¹

Affiliations

¹ Department of Obstetrics and Gynecology, Lithuanian University of Health Sciences, Medical Academy, Eiveniu st. 2, Kaunas, Lithuania.
² Institute of Pharmacoeconomics, Kaunas, Lithuania.

Abstract

Background: Experts in many countries are recommending a scaling up midwifery-led care as a model to improve maternal and newborn outcomes, reduce rates of unnecessary interventions, realise cost savings, and facilitate normal spontaneous vaginal birth.

Objective: The aim of this study was to compare midwifery-led and obstetrician-gynaecologist-led care-related vaginal birth outcomes.

Participants: Pregnant women in Kaunas city maternity care facilities.

Methods: A propensity score-matched case-control study of midwifery-led versus physician-led low-risk birth outcomes. Patient characteristics and outcomes were compared between the groups. Continuous variables are presented as mean ± standard deviation and analysed using the Mann-Whitney U test. Categorical and binary variables are presented as frequency (percentage), and differences were analysed using the chi-square test. Analyses were conducted separately for the unmatched (before propensity score matched [PSM]) and matched (after PSM) groups.

Results: After adjusting groups for propensity score, postpartum haemorrhage differences between physician-led and midwifery-led labours were significantly different (169.5 and 152.6 mL; p = 0.026), same for hospital stay duration (3.3 and 3.1 days, p = 0.042). Also, in matched population, significant differences were seen for episiotomy rates (chi² = 4.8; p = 0.029), newborn Apgar 5 min score (9.58 and 9.76; p = 0.002), and pain relief (chi² = 14.9; p = 0.002). Significant differences were seen in unmatched but not confirmed in matched population for obstetrical procedures used during labour, breastfeeding, birth induction, newborn Apgar 1 min scores, and successful vaginal birth as an overall spontaneous vaginal birth success measure.

Conclusion: The midwifery-led care model showed significant differences from the physician-led care model in episiotomy rates, hospital stay duration and postpartum haemorrhage, and newborn Apgar 5 min scores. Midwifery-led care is as safe as physician-led care and does not influence the rate of successful spontaneous vaginal births.

Keywords: birth outcomes; midwifery-led care; physician-led care; vaginal birth.