NEW YORK – Fatima Marinho wants to know more about what’s killing Brazilians.
Marinho, MD, who leads a Ministry of Health department responsible for information in analysis, knows Brazil has a good system for recording births and deaths, but it needs to improve the quality of its data. 30% of the causes of death recorded in Brazil are not the true causes, Marinho said.
For example, physicians may record the blood infection septicemia as the cause of a patient’s death, but the underlying cause may actually be complications from surgery, HIV/AIDS, or cancer, said Marinho, whose team developed an app called AtestaDO that improves physicians’ ability to accurately record causes of death.
Or, “stroke” is listed as a cause of death, but was it ischemic or hemorrhagic? Public health authorities would like to know what percentage of stroke deaths are ischemic, which is linked to risk factors such as high blood pressure, heart disease, smoking, or diabetes.
And some countries are far behind Brazil in collecting critical data. Half of all deaths worldwide do not have a death certificate with a cause of death. That’s 29.4 million deaths without a recorded cause. Some countries may have legislation mandating civil registration and vital statistics, but systems for recording birth and death data aren’t actually in place.
“It’s a huge problem that results in a lot of preventable death and disease because governments just don’t have the information they need to target the causes of these things,” Michael R. Bloomberg told a gathering of experts on Tuesday.
Partners of the Data for Health initiative, supported by the Bloomberg Philanthropies and the Australian government, came to New York this week to share accomplishments and lessons learned from their efforts in the $100 million, 4-year initiative, which began in 2015 and includes 20 countries.
Bloomberg cited progress that includes 9 countries that have adopted high-quality death certificates and 12 countries collecting better data on deaths that happen outside health facilities. “In total, we’ve collected better data on more than a million deaths, which is really remarkable,” Bloomberg said.
In the absence of reliable data, countries have had to make assumptions and use statistical models to estimate how many people are born and die and what they’re dying of. The data are often way off, according to Philip Setel, a Vital Strategies vice president and director of CRVS improvement program. “We have been making due with statistical gymnastics for far too long,” Setel said. “It’s time to make people count by counting people.”
Setel cited one example of the transformation made possible by better data. When officials in a district in Tanzania learned there were many more malaria deaths of children outside of medical facilities than they had assumed, they changed the first-line antimalarial drugs used, shifted funds to address malaria and increased prevention outreach to households and communities, resulting in reductions in malaria deaths.
“The machinery of data production harnessed to data use and impact creates a virtuous cycle of data use and demand,” Setel said.
Getting the virtuous cycle up and running hasn’t been easy, said Jennifer Ellis, who directs Data for Health. “We found tons of UN calls to action for data-driven decision-making for health but not a lot of what to do,” Ellis said of her 2013 research into different countries’ data needs and capabilities. “There was a gap that needed to be filled between the rallying call for providing data and ‘here’s how you do it and how you can keep doing it.”
Governments in some developing countries were also daunted by UN recommendations—often an idealized laundry list of data collection technology and efforts that could cost hundreds of millions of dollars and take decades to implement, she said. “The barrier was people were letting the huge picture get in the way of making effective, smart, targeted resource investments that can have a big impact,” Ellis said.
Among the low-cost solutions that Ellis and her team recommend: Make better use of existing data, drawing on it to inform briefings. The goal is to “build data analysis capacity to create a culture of using data as the default approach,” she said.
To understand the causes of death outside of health facilities, they recommend use of verbal autopsies (a structured interview with family members or others close to the deceased). The algorithms for verbal autopsies have been refined in recent years and are remarkably accurate.
Ed. Note: Michael R. Bloomberg is a benefactor of the Johns Hopkins Bloomberg School of Public Health, which publishes Global Health NOW.
Join the tens of thousands of subscribers in more than 100 countries who rely on Global Health NOW summaries and exclusive articles for the latest public health news. Sign up for our free weekday enewsletter, and please share the link with friends and colleagues: http://www.globalhealthnow.org/subscribe.html