A Cluster-Randomized Trial of Ultrasound Use to Improve Pregnancy Outcomes in Low Income Country Settings

Overview

In many low-income countries, the use of ultrasound by medical officers and non-physician health care staff (e.g., midwives) for antenatal identification of high risk pregnancies is a new intervention requiring authoritative investigation. The primary hypothesis to be assessed in this study is that antenatal ultrasound screenings performed by medical officers and non-physician health care staff will significantly reduce a composite outcome consisting of maternal mortality and maternal near miss, stillbirth and neonatal mortality in low-resource settings. Underpinning this hypothesis are two assumptions. The first assumption is that antenatal detection of complicated pregnancies will lead to appropriate referral at the right time for complicated pregnancies to comprehensive emergency obstetric and neonatal care (EmONC) facilities. The second assumption is that ultrasound's introduction will increase antenatal attendance leading to greater rates of institutional delivery. To assess these underlying assumptions beyond the composite end point, this study will investigate the health system impact of compact ultrasound. Secondary outcomes include antenatal attendance rates, institutional delivery rates at basic EmONC facilities, referral rates to comprehensive EmONC facilities, cesarean section rates (both planned and emergent) and an assessment of medical officers and non-physician health care provider ultrasound competence and training quality.

The use of ultrasound by physician and non-physician health care staff for antenatal identification of complicated pregnancies is a potentially effective intervention; however, authoritative investigation in many low-resource settings is needed to establish its potential impact. The investigators propose to undertake a multi-country, cluster randomized trial to assess the impact of antenatal ultrasound screening performed by community physician and non-physician health care staff on a composite outcome consisting of maternal mortality and near miss maternal mortality, stillbirth and neonatal mortality in low-resource settings. Underpinning this objective are several assumptions. The first assumption is that ultrasound's introduction will increase antenatal attendance and improved outcomes due to the antenatal care alone, and greater rates of institutional delivery. The second assumption is that ultrasound use will lead to antenatal detection of complicated pregnancies and timely and appropriate referral for complicated pregnancies to comprehensive emergency obstetric and neonatal care (EmONC) facilities. Increases in antenatal care utilization and referral will result in a decrease in a composite outcome including maternal mortality and near miss mortality, stillbirth and neonatal death. Secondary outcomes to be evaluated include antenatal attendance rates, institutional delivery rates at basic EmONC facilities, referral rates to comprehensive EmONC facilities, cesarean section rates (both planned and emergent), an assessment of community physician and non-physician health care provider ultrasound training and competence and the cost-effectiveness of ultrasound in community health facilities. The investigators will also determine causes for non-compliance with recommendations for referral. In summary, this trial will evaluate whether training antenatal health care providers to perform basic obstetric ultrasonography, and using these trainees to provide routine ultrasounds in primary care clinics and to refer appropriately will improve pregnancy outcomes in low-resource settings. To assess the impact of ultrasound, the investigators propose to utilize an existing research infrastructure, the Global Network for Women's and Children's Health Research (Global Network), which currently includes 7 sites in 6 countries, India (2), Pakistan, Kenya, Zambia, DRC and Guatemala. The investigators of the Global Network have an ongoing maternal and newborn health registry to document all pregnancies and their outcomes to 6 weeks post-delivery in more than 100 communities. Thus, population-based rates of maternal mortality and morbidity, stillbirth, and neonatal mortality and morbidity, as well as health care utilization, are being obtained. A sub-set of these clusters will be utilized for the ultrasound trial. RTI International serves as the data coordinating center for the Global Network to help facilitate the design and conduct of the trial, manage the trial related data, and provide statistical analyses of the trial results. GE Healthcare will provide the ultrasound equipment, and will also fund the University of Washington to provide training and technical support on the implementation of ultrasound in community settings. Together, the Global Network with the support of the University of Washington and GE Healthcare will maximize the resources necessary to conduct a definitive trial on the potential impact of ultrasound to reduce maternal and newborn mortality and maternal morbidity in low-resource settings.