科学与发展网络(SciDev.Net)

A mother and child in India

Flickr/Roochester

Researchers from India and the United Kingdom have developed an inexpensive surveillance system for measuring maternal mortality, which could eventually guide policymakers in protecting maternal health.

Conventional surveillance of maternal deaths is expensive and logistically challenging, write the researchers. Under-reporting is also frequent, particularly in India's remote rural areas, where maternal mortality is increased. As a result, policy decisions are made on inadequate evidence.

About two-thirds of Indian maternal deaths occur in just nine states. The study looked at two of those — Jharkhand and Orissa in eastern India — with a combined population of 228,186.

In their surveillance system, the researchers recruited key local informants —mostly traditional birth attendants — to record all births and stillbirths, and deaths of women of reproductive age (15–49 years).

By asking the causes of death from relatives of women who had died, deaths were classed as maternal, pregnancy-related or late maternal. Doctors then verified causes of death by examining the verbal autopsies.

A maternal mortality ratio of 722 per 100,000 live births was found. And the cost of the system was just US$0.02 per person per year, the researchers noted.

"The new method appears to be robust," said Prasanta Tripathy, an author of the study who is affiliated to Ekjut, an India-based nongovernmental organisation. "We always need to stress the difference between estimating — as most people do —and measuring, which is what we did," he told SciDev.Net.

"While maternal-mortality ratio in developing countries is mostly measured retrospectively, this method measured it prospectively through proper identification of live births and maternal deaths," said Sudip Saha, associate professor of obstetrics and gynaecology at the Calcutta Medical College Hospital, India. "The method is relatively easy and low-cost and helps avoid missing maternal deaths."

The authors write that the method could be used to monitor trends in maternal mortality and test the effect of interventions aiming to enhance maternal health.

The research was published in the journal BMC Pregnancy and Childbirth. 

Link to full paper in BMC Pregnancy and Childbirth 

http://www.biomedcentral.com/content/pdf/1471-2393-8-6.pdf