In order to solve a health problem, it may sound a bit obvious to say you first need to know what it is. A diagnosis.
What may not be so obvious is that one of the biggest, and perhaps less appreciated challenges, in global health today is the lack of complete, accurate and comparable health data.
A series of reports on child mortality published this week in The Lancet, entitled Every Newborn, indirectly illustrated this challenge. The Lancet report said, in a nutshell, that while we’ve made progress reducing child mortality overall, there’s been little progress made when it comes to reducing death rates in newborns worldwide.
As many news stories reported, and which Melinda Gates emphasized in a speech at the World Health Assembly, nearly 3 million babies die within 28 days of birth due to lack of attention – and resources – devoted to this high-risk time of life.
What was not emphasized is the difficulty in even getting an accurate count of these deaths. The study notes that about a third of infants lack birth certificates even at one year of age. The authors of The Lancet study also noted that countries with higher death rates are the least able to track those deaths.
Without good data, how can we be sure we understand the biggest challenges we face in improving health worldwide? How can we determine if our efforts are making a difference?
At the Institute for Health Metrics and Evaluation (IHME), we develop and employ cutting-edge tools to overcome the problem of missing and poor-quality data. But sophisticated scientific tools, while helpful for shedding light on some of the most perplexing global health questions, are no substitute for high-quality, comprehensive data.
“In countries where births and deaths are undercounted or not registered at all, we don’t have to wait for good data—we can compensate by using a variety of methods,” said Dr. Haidong Wang, an Assistant Professor at the Institute for Health Metrics and Evaluation and the lead author of a recent study of child mortality published in The Lancet. “The downside of these data problems is that we end up with a huge range of possible estimates for the number of child deaths in a given year.”
For a country like Nigeria, which lacks complete civil registration (a system that collects data via birth and death certificates), IHME’s estimates of under-5 child death rates ranged from 796,100 to 991,600 in 2013.
Compare that to Mexico, which has a higher-quality civil registration system that helps strengthen our estimates of child deaths. IHME’s estimates of under-5 deaths in Mexico had a much narrower range: 34,700 to 42,000.
The figures below show the different data points IHME researchers used to estimate deaths among children under 5 in Nigeria and Mexico. The data points used in the Nigeria analysis often contradict each other, while the Mexican data points form a relatively smooth line.
Data sources used to estimate under-5 deaths in Nigeria, 1990-2013
Data sources used to estimate under-5 deaths in Mexico, 1990-2013
- Lancet
Note: These graphs were derived from the webappendix of a paper entitled Global, regional, and national levels of neonatal, infant, and under-5 mortality during 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013.
These examples show how better data would boost the quality of the research findings produced by IHME and other global health research organizations.
“What data improvements would make the biggest difference in your work?” I asked Dr. Wang.
“Better civil registration in most sub-Saharan African countries as well as in India, Pakistan, and Bangladesh,” he answered.
Dr. Wang noted that India has made progress toward better counting births and deaths with its sample registration system. Sample registration is a representative survey of the population used to understand death rates. It is a good intermediary solution to track death trends as a country is scaling up its civil registration system, but it is not as accurate as a system that registers all births and deaths.
“To improve estimates of child mortality, countries need to invest in civil registration systems,” said Dr. Wang.
Some countries are making huge strides when it comes to counting deaths. According to Dr. Wang, China, for example, started a “Direct Reporting System” for deaths in 2004 after the SARS epidemic. The system scaled up quickly and was tracking an estimated 60% of deaths in the country by 2013.Through the system, community health workers, hospital staff, and local government officials report deaths electronically. As it is in the process of expanding its Direct Reporting System, China is also bolstering its sample registration to fill in the information gaps. In 2012, the sample registration system gathered data in 162 sites, but the number of sites reached 600 in 2013.
Tracking deaths through studies such as the Global Burden of Disease and other efforts around the world provides critical information for decision-making, but investing in better data collection is essential for ensuring that the global health community’s efforts are well targeted and making a difference.
Katie Leach-Kemon, a weekly contributor of global health visual information posts for Humanosphere, is a policy translation specialist from the University of Washington’s Institute for Health Metrics and Evaluation.
