Effect of a scaled-up neonatal resuscitation quality improvement package on intrapartum-related mortality in Nepal: A stepped-wedge cluster randomized controlled trial
Ashish KC, Uwe Ewald, Omkar Basnet, Abhishek Gurung, Sushil Nath Pyakuryal, Bijay Kumar Jha, Anna Bergström, Leif Eriksson, Prajwal Paudel, Sushil Karki, Sunil Gajurel, Olivia Brunell, Johan Wrammert, Helena Litorp, Mats Målqvist
PLOS Medicine · 2019 · DOI: 10.1371/journal.pmed.1002900
Countries: Nepal
What Was Studied
Researchers ran a stepped-wedge cluster randomized controlled trial across 12 public hospitals in Nepal (four high-volume, four medium-volume, and four low-volume, each handling more than 1,000 deliveries a year) to test the Nepal Perinatal Quality Improvement Package (NePeriQIP), a Ministry of Health and Population initiative combining hospital leadership training, Plan-Do-Study-Act quality-improvement cycles, in-hospital facilitators, and Helping Babies Breathe-based neonatal resuscitation and Essential Newborn Care training. Over 18 months (14 April 2017 to 17 October 2018), hospitals were randomly allocated in blocks of three to cross from a control period to an intervention period at three-month intervals across four 'wedges'. The trial enrolled 92,322 eligible women admitted in labour with a detectable fetal heartbeat, yielding 88,524 recorded deliveries and 89,014 live-born and stillborn infants analysed on an intention-to-treat basis using generalized linear mixed models to account for hospital-level clustering.
What They Found
Intrapartum-related mortality (intrapartum stillbirth plus death in the first 24 hours) fell from 10.7 per 1,000 births in the control period to 7.8 per 1,000 births in the intervention period (adjusted odds ratio [aOR] 0.79, 95% CI 0.69–0.92, p=0.002). This was driven mainly by fewer intrapartum stillbirths (6.8 to 4.7 per 1,000 births; aOR 0.73, 95% CI 0.61–0.88, p<0.001); first-day mortality (3.9 to 3.1 per 1,000 live births; aOR 0.92, p=0.49) and seven-day early neonatal mortality (12.7 to 10.1 per 1,000 live births; aOR 0.89, p=0.09) did not fall significantly, which the authors attribute to insufficient statistical power given only 12 hospital clusters. Bag-and-mask ventilation use in babies with a low 1-minute Apgar score (<7) rose from 3.2% to 4.0% (aOR 1.52, 95% CI 1.32–1.77, p=0.003), consistent with improved resuscitation practice. The effect was uneven across hospital types: small-volume hospitals had the largest relative reduction (15.2 to 6.7 per 1,000 births, aOR 0.46, 95% CI 0.27–0.76), while medium-volume hospitals showed no significant change (8.4 to 8.5 per 1,000 births, aOR 1.06, 95% CI 0.83–1.36, p=0.796) — a caution against assuming uniform scale-up effects.
What This Means for Nepal
Because this is a Ministry of Health and Population-led trial of MoHP's own NePeriQIP package, it speaks directly to a live scale-up decision for DoHS and the Family Welfare Division: leadership-and-facilitation-based quality improvement can meaningfully cut intrapartum deaths in Nepal's own hospitals, but the benefit is concentrated in stillbirth prevention rather than post-birth neonatal survival, and it did not work in the trial's medium-volume hospitals. Two actions follow. First, before further national scale-up, MoHP/DoHS should diagnose why medium-volume hospitals underperformed (leadership engagement, facilitator capacity, or case-mix) rather than assume automatic replication — especially since only 37.9% of sanctioned physician/GP posts are filled nationally (0% in Karnali), and the trial's meso-level 'hospital leadership' lever presupposes staffing levels its own study hospitals mostly had. Second, since institutional delivery is already 79.4% (NDHS 2022) yet neonatal mortality has stagnated at 21 per 1,000, DoHS should pair any resuscitation-training scale-up with FCHV-linked referral and postnatal home-visit follow-up, so that facility-side gains in stillbirth prevention are matched by community-side monitoring of infants who survive labour but remain at risk in the first week of life.
Contextualisation
Nepal's institutional delivery rate has risen from 9% in 2001 to 79.4% (NDHS 2022), so most Nepali mothers now give birth in a facility — yet neonatal mortality has stagnated at 21 per 1,000 live births since 2016, meaning the bottleneck has shifted from getting women to facilities toward the quality of care they receive once there. This trial evaluates the Ministry of Health and Population's own Nepal Perinatal Quality Improvement Package (NePeriQIP), scaled up to 12 government hospitals, making it a direct, government-owned test of whether hospital leadership and resuscitation-training reforms can convert Nepal's high institutional-delivery rate into fewer intrapartum deaths. Because the trial (2017–2018) pre-dates full devolution of facility governance under Nepal's federal structure — where local governments now own "basic health and sanitation" and provinces own "health services" — its centrally-led model needs re-testing under the current three-tier system.