Though the World Health Organization released a report last spring suggesting that maternal mortality had not improved in the last 30 years, with roughly half a million mothers dying during pregnancy and childbirth each year, a new study published online earlier this week in the journal The Lancet found that the number of maternal deaths had actually dropped by some 35% during that time period—from 500,000 in 1980, to 343,000 in 2008. So, how can the figures be so different? To learn more, TIME spoke with Dr. Christopher Murray, an author of the new study and director of the Institute for Health Metrics and Evaluation at the University of Washington:
TIME: For this new study, you reportedly gathered three times as much data as previous investigations of global maternal mortality. How were you able to get so much more information?
Dr. Murray: We spent two years on this study, so we were able to invest a lot of time in both finding national statistical data from systems that hadn’t been used before—including studies that had been done not necessarily by governments but researchers in different countries—and found ways to get more measurements out of household surveys.
TIME: How you were able to get more information from these surveys?
Dr. Murray: One of the main ways that we learn about child health and maternal health in developing countries is through these household surveys, and a mainstay for measuring maternal mortality rates, is through a sibling history. You go to households and ask women about their sisters—did they die during pregnancy or within 6 weeks of childbirth? But, you can only ask living women about their sisters, which creates survivor bias. Their sisters tend to live at a higher rate.
Statisticians figured out a way of solving the problem [of survivor bias] and we figured out a way to apply that to these surveys, which also enables us to get measurements going back in time. That’s jut one example of how we got the largest and highest quality database.
TIME: What’s another example of how you were able to gather more data for this study?
Dr. Murray: We did attempt to identify where maternal deaths might be hidden in the cause of death data. One of the things that we looked at in the U.S. [and other countries] is that there’s quite a lot of misclassification of maternal death. Quite often they end up being assigned other causes like septicemia and pulmonary embolism, and don’t get on the death certificate as a maternal death. That was a big part of the study to try to deal with that misclassification problem.
TIME: While your research indicates that worldwide maternal mortality has declined, you also found that, in the U.S., it has risen, and more so than previous research suggested. Were you surprised by these findings?
Dr. Murray: I’ll have to confess that I wasn’t surprised. When we look in other research at pretty much any of the main health outcomes—whether child mortality or maternal mortality or life expectancy—the US is ranked somewhere between 30th and 45th. In that way, this is a confirmation that in this largely preventable cause of death, the US is not doing as well as many countries in Europe, contrast the US with Italy for example, and that’s a pretty big difference. [The maternal mortality ratio—a measurement of maternal deaths per 100,000 live births—dropped from 7 in 1990 to 4 in 2008 in Italy, while it rose from 12 to 17 in the U.S. during the same time period. Italy had the lowest maternal mortality ratio of 181 countries included in the study, while the U.S. was ranked 40th for this measure.] It’s alarming, but not surprising.
TIME: What do you think are the most important findings in your new research?
Dr. Murray: There are a couple of really important points. First, although there’s been progress globally, if you look by country (XLS) it’s very mixed. You have places like Egypt or Bolivia or China [where the maternal mortality rate has gone down significantly] and then there are other countries where it’s actually going up. The global picture is moving in the right direction, but it’s a combination of really stellar performance and things getting worse in some places. Understanding that is really important because it means we can learn lessons from where things are improving at a fast pace. What did Egypt do that’s right? What did Bolivia do? [In Egypt, the maternal mortality ratio dropped from 195 per 100,000 live births in 1990 to 43 in 2008; in Bolivia, it fell from 439 to 180 during the same period.]
Second, the rise in East and Southern Africa is related to the HIV epidemic. For the first time we’re making a very strong connection between maternal mortality and maternal HIV. [The study found that 64,100 maternal deaths were attributable to HIV in 2008.] It makes the connection between the HIV service agenda and maternal health much stronger than it has been in the past.
TIME: In an editorial accompanying your research in The Lancet, the journal’s editor, Dr. Richard Horton, says that advocacy groups tried to delay the publication, believing that this new data indicating progress in decreasing maternal mortality could undermine education and outreach efforts. What was your reaction when you learned of such requests?
Dr. Murray: As a scientist that’s clearly disappointing. We [should] just have an open debate about what the evidence is on such an important topic as maternal mortality. But, in the end, but what happened is the right thing: we did a study, it went through the process of peer review, and was published in a journal.
TIME: Are you concerned by any of the criticism or inquiry surrounding the fact that your data yields such different figures for maternal mortality than previous research? And how do you think this new assessment can be reconciled with previous investigations?
Dr. Murray: This is as the process of science should work. What we should do through this is stimulate a very vigorous public, open, and transparent debate on the data, country by country, and on the methods, and how we can improve them. Hopefully the broad discussion of will unlock more data and will yield improvements. We’ve improved on what was available before, and we’re sure that somebody will come up with a better way to do this in the future.
This interview was edited for clarity and continuity.
Read an earlier post about the new maternal mortality data here.