Effect of The CRADLE Vital Signs Alert Device Intervention on Referrals for Obstetric Haemorrhage in Low-Middle Income Countries: A Secondary Analysis of A Stepped-Wedge Cluster-Randomised Control Trial

Background: Obstetric haemorrhage is the leading cause of maternal death worldwide, 99% of which occur in low and middle income countries. The majority of deaths and adverse events are associated with delays in identifying compromise and escalating care. Management of severely compromised pregnant women may require transfer to tertiary centres for specialised treatment, therefore early recognition is vital for efficient management. The CRADLE vital signs alert device accurately measures blood pressure and heart rate, calculates the shock index (heart rate divided by systolic blood pressure) and alerts the user to compromise through a traffic light system reflecting previously validated shock index thresholds. Methods: This is a planned secondary analysis of data from the CRADLE-3 trial from ten clusters across Africa, India and Haiti where the device and training package were randomly introduced. Referral data were prospectively collected for a 4-week period before, and a 4-week period 3 months after implementation. Referrals from primary or secondary care facilities to higher level care for any cause were recorded. The denominator was the number of women seen for maternity care in these facilities. Results: Between April 1 2016 and Nov 30th, 2017 536 223 women attended maternity care facilities. We reviewed a total of 144 cases that were referred, 119 cases 3.84% pre-implementation and 4.48% post-implementation were referred to higher-level facilities. Overall referrals did not significantly reduce (OR 0.89 (0.39- 2.05)) but there was a significant reduction in referrals for bleeding (OR 0.5 (0.39-0.65) following introduction of the device, with homogeneity (I-squared 10.4) between sites. There was no increase in any bleeding-related morbidity. Conclusions: Referrals for obstetric haemorrhage reduced following implementation of the CRADLE Vital Signs Alert Device, occurring without an increase in maternal death or emergency hysterectomy. This demonstrates the potential benefit of Shock Index in management pathways for obstetric haemorrhage and targeting limited resources in low- middle- income settings. Trial Registration This study is registered with the ISRCTN registry, number ISRCTN41244132 (02/02/2016)