Search

Background: In 2015, over 800 women died in pregnancy and childbirth every day. There are effective treatments for the leading causes of maternal death, but they require early detection by measurement of vital signs, and timely administration to save lives. The CRADLE-Vital Sign Alert accurately measures blood pressure and pulse and calculates shock index. Results are displayed on a traffic light early warning system. This aim of this thesis was to evaluate the impact of this intervention on maternal mortality and morbidity in low and middle-income countries. Methods: A pragmatic, stepped-wedge randomised-controlled trial with a nested mixed-methods process evaluation was undertaken. This was preceded by a mixed-method feasibility study. The intervention was introduced into every level of routine maternity care across 10 clusters in Africa, India and Haiti. The primary composite outcome was at least one of eclampsia, emergency hysterectomy and maternal death per 10,000 deliveries. Delivery of the intervention, its uptake and potential mechanism of action were measured and integrated with qualitative findings and measures of resources and staffing in each cluster. Findings: Between April 1st 2016 and November 30th 2017, among 536,223 deliveries, the primary outcome occurred in 4067 women. There was an 8% decrease in the primary outcome from 79.4/10,000 deliveries pre-intervention to 72.8/10,000 post-intervention. After planned adjustments for variation in event rates between and within clusters over time, the unexpected degree of variability meant we were unable to judge the benefit or harms of the intervention (OR 1.22, 95% CI 0.73-2.06a p=0.45). Overall, the intervention was delivered with reasonable fidelity and improved the availability of vital signs equipment and number of women with BP measurements (79.2%-s. 97.6%a OR 1.30, 95% CI 1.29 - 1.31). Therewere significant differences in the effect of the intervention between 8 individual clusters which could not be explained by the measures of implementation or local context. Interpretation: The acceptability and feasibility of the intervention has been demonstrated. Despite the rigorous trial design, effectiveness of the intervention at reducing mortality and morbidity has not been shown. Measuring implementation alongside effectiveness was feasible and beneficial in describing differences between clusters in maternal health. In this case, these differences could not explain the difference in the effect of the intervention between individual clusters. Stepped-wedge trials across multiple countries ha -e considerable methodological challenges and should be powered to demonstrate an effect of the intervention in each country. Further research in the selection and integration of process measures in low and middle -income countries is required.