To maximise the potential benefits of maternity care services, pregnant women need to be able to physically get to health facilities in a timely manner. In most of sub-Saharan Africa, transport represents a major practical barrier. Here we evaluate the extent to which an innovative national ambulance service in Ethiopia, together with mobile phones, may have been successful in averting pregnancy-related deaths.
An operational assessment of pregnancy-related deaths in relation to utilisation of the new national ambulance service was undertaken in six randomly selected Districts in northern Ethiopia. All 183 286 households in the six randomly selected Districts were visited to identify live-births and deaths among women of reproductive age that occurred over a one-year period. The uptake of the new ambulance transport service for women's deliveries in the same six randomly selected Districts over the same period was determined retrospectively from ambulance log books. Pregnancy-related deaths as determined by the World Health Organization (WHO 2012) verbal autopsy tool  and the InterVA-4 model  were analysed against ambulance utilisation by District, month, local area, distance from health facility and mobile network coverage.
A total of 51 pregnancy-related deaths and 19 179 live-births were documented. Pregnancy-related mortality for Districts with above average ambulance utilisation was 149 per 100 000 live-births (95% confidence interval CI 77-260), compared with 350 per 100 000 (95% CI 249-479) for below average utilisation (P = 0.01). Distance to a health facility, mobile network availability and ambulance utilisation were all significantly associated with pregnancy-related mortality on a bivariable basis. On a multivariable basis, ambulance non-utilisation uniquely persisted as a significant determinant of mortality (mortality rate ratio 1.97, 95% CI 1.05-3.69; P = 0.03).
The uptake of freely available transport in connection with women's obstetric needs correlated with substantially reduced pregnancy-related mortality in this operational assessment, though the design did not allow cause and effect to be attributed. However, the halving of pregnancy-related mortality associated with ambulance uptake in the sampled Districts suggests that the provision of transport to delivery facilities in Africa may be a key innovation for delivering maternal health care, which requires wider consideration.
2016. Vol. 6, no 1, 010410