On the re-opening of SLPP Office … “We would never encourage violence” John Benjamin...

Sierra Leone News: Dueling Data

There’s a common saying that goes something like, “you can’t fix what you don’t measure.” Keeping detailed records on an issue like maternal mortality or procurement helps find the problem areas and measure how things have improved or worsened. It’s not a glamorous job, but it is a necessary one. Without data guiding the way, we might as well be children playing hide and seek in the dark.
But what happens when the numbers don’t add up; when multiple sources of data disagree with each other? Let’s take maternal mortality for example. There are two main estimates for maternal mortality rates (MMR) in this country: The World Health Organization’s from 2015 and the Demographic Health Survey in 2013. Each estimate measures MMR as the number of deaths per 100,000 live births. The WHO says the MMR rate is 1,360 while the DHS estimate is slightly lower 1,165 deaths per 100,000 live births. For years, these have been held up as the public health community’s best guess at the severity of the issue, but new data is showing they might not be so airtight.
The government’s Maternal Death Surveillance & Response (MDSR) report is an annual survey of all the recorded maternal deaths over the previous year. The 2016 report is the third annual iteration and shows the highest death toll yet. This isn’t because the MMR is getting worse, it’s because a higher percentage of the deaths are being officially reported and recorded. This increased documentation is actually a really great development because it gives the Ministry of Health and NGOs concrete data on how many mothers died and for what reasons.
The MDSR report recorded about 218,000 live births in 2016. Using the more conservative DHS mortality estimate, they expected about 3,000 maternal deaths, but that’s not what the records show. Only 706 maternal deaths were recorded in MDSR report, a mere 24% of what was expected. The authors said this huge disparity signals, “a need for improving the reporting of maternal deaths, or revisiting the current maternal mortality estimates for the country.” In other words, either thousands of mothers are dying undetected or the old estimates were inflated. Most likely it’s a bit of both.
The report does concede that, “it is important to note that the report does not provide a complete picture of maternal deaths in the country but it does provide useful information for strengthening major areas of intervention for MDSR.” There is no doubt many maternal deaths go unreported, but it doesn’t rule out the idea that prevailing estimates about maternal mortality are too high.
My favorite and most trusted source for statistics on global health is the Institute of Health Metrics and Evaluation (IHME), an independent organization funded by the Bill & Melinda Gates Foundation. Luckily for me, they are partnered with the university I go to so I’m constantly checking for new studies they publish. According to the IHME’s groundbreaking 2013 report published in the prestigious journal The Lancet, Sierra Leone’s MMR is actually 623 deaths per 100,000 live births. This is less than half of the WHO’s best estimate and far below the DHS findings. So the question is, which estimate is the right one?
This is an almost impossible question to answer right now. The only way to know for sure is to improve the MDSR reporting system so it records every single maternal death. Looking at the current MDSR report, it seems very unlikely that the high MMR estimates put forth by the DHS and WHO accurately reflect what’s happening in Sierra Leone. I find it hard to believe that 75% of maternal deaths go completely unreported when an increasingly high majority of women are delivering in health facilities. Using the IHME estimates, only about half of deaths would be unaccounted for. Only time will tell which one is right.
These statistics might seem boring – thank you for even reading this far through the column – but millions of dollars in aid money are given based on global health statistics like this. Inflated estimates might help bring in more international aid but they funnel money away from other worthwhile causes, often causing more harm than good.
Timothy’s Take
Monday August 14, 2017.

Comments are closed.