Evaluating the real-world scale-up of the CRADLE Vital Signs Alert intervention into routine maternity care in Sierra Leone (CRADLE-5): a stepped-wedge, type 2 hybrid implementation–effectiveness, cluster-randomised controlled trial

Elsevier, The Lancet Obstetrics, Gynaecology, & Women's Health, Volume 1, Issue 4, December 2025, Pages e270-e280
Authors: 
Alexandra E Ridout MD , Francis L Moses MPhil , Simren Herm-Singh MSc , Candace Beoku-Betts MSc , Edward Businge MPH , Cristina Fernandez Turienzo PhD , Venetia Goodhart MBBS , Alice Hurrell PhD , Daniel Kamara MSc , Jennifer Kamara MSc , Kellie Koroma MBA , Katy Kuhrt MBBS , Anastasia Martin MBBS , Mariama Momoh MSc , Paul T Seed CSTAT , Betty Sam SRN , Sorie Samura MSc , Prince Tommy Williams MA , Prof Lucy C Chappell PhD , Francis Smart MPH , Senesie Sheriff

Background

Sierra Leone has one of the highest maternal mortality ratios globally. Most preventable maternal deaths result from haemorrhage, hypertension, or sepsis, yet many facilities do not have the basic tools for vital sign monitoring. This study evaluated whether national scale-up of the CRADLE intervention (a vital signs device with a traffic-light early warning system and training) could reduce adverse maternal and perinatal outcomes and be sustainably integrated into routine care.

Methods

We conducted a pragmatic, hybrid type 2 implementation–effectiveness, stepped-wedge, cluster-randomised controlled trial across eight districts in Sierra Leone. Each district (cluster) transitioned from routine care to the CRADLE intervention in a randomised sequence every 6 weeks. The intervention included device distribution, cascade training by local CRADLE champions, and integration into clinical pathways. The primary clinical outcome was a composite of eclampsia, emergency hysterectomy, maternal death, or stillbirth (without double counting). Implementation and scale-up evaluation included outcomes related to fidelity, adoption, and system integration. All analyses followed an intention-to-treat principle. This trial is registered with ISRCTN, 94429427, and is completed.

Findings

Between May 23, 2022, and June 30, 2023, 2171 CRADLE devices were distributed to 643 government health facilities, and 2135 frontline workers were trained. Among 93 811 deliveries (40 339 pre-intervention and 53 472 post-intervention, 1568 (1·7%) women (699 [1·7%] in the pre-intervention group and 869 [1·6%] in the post-intervention group) experienced at least one primary clinical outcome event (134 vs 205 eclamptic fits, 32 vs 39 emergency hysterectomies, 60 vs 66 maternal deaths, and 551 vs 650 stillbirths), with no significant reduction in the composite primary outcome for deliveries taking place pre-intervention and post-intervention (incidence risk ratio [IRR] 1·01, 95% CI 0·87–1·16). With respect to implementation outcomes, vital signs measurement frequency increased with implementation of the intervention (IRR 1·38, 95% CI 1·07–1·77), with improved accuracy (terminal digit preference, ie, recording values ending in 0 or 5, declined from 32·8% to 8·7%; IRR 0·44, 95% CI 0·25–0·79). Sustainable scale-up was strengthened by national leadership, integration into training curricula, and community engagement. Persistent system-level barriers—including insufficient availability of essential drugs, workforce shortages, inadequate infrastructure such as electricity supply, and low clinical impact—restricted effectiveness.

Interpretation

National scale-up of the CRADLE intervention improved detection of maternal complications and referral practices but did not reduce overall adverse outcomes. The trial showed that large-scale adoption of CRADLE is feasible and identified barriers to facilitate an impact on maternal and perinatal mortality. Strengthening of the broader health system in Sierra Leone is required.

Funding