A recent study from the Lancet says that malaria caused 1.24 million deaths in 2010 (compared to previous estimates of 655,000), suggesting that the number of deaths from malaria globally may be twice as large as previously believed. Needless to say, this new finding is making headlines and bringing renewed attention to malaria as a central focus of global health efforts.
Estimating disease burden is important for budget and resource allocation—and in the case of malaria it has been done many times before. The Lancet study cites 14 different estimates for malaria burden, noting the “highly variable results” of this research. So perhaps the best question to ask is how these results could be so different? How could over half a million more people be dying from malaria annually without us knowing about it?
For the most part, the development community is largely in the dark when it comes to knowing who lives and dies, where, and why. The methods of this new study (as with many studies that estimate disease burden) relied on data based from vital statistics registries and verbal autopsy results. But in many malaria endemic countries, these systems offer low coverage and poor quality—leaving the vast majority of deaths poorly-coded or never documented in the first place.
For instance, from 1995-2004 the proportion of the population living in countries with complete death registrations in Africa, South-East Asia, and the Western Pacific was 7%, 1%, and 13% respectively (see here) and only half of the countries in Africa and South Asia even record cause of death data. In Africa, only the islands of Mauritius and the Seychelles currently have complete registration of births, deaths, and cause of death.
In the absence of complete information, researchers must model their estimates based on what data they have access to, even if poorly coded cause-of-death data and low sample sizes result in high levels of uncertainty. Accurate information on cause of death—and other vital events—is essential for understanding disease burden, influencing policy, and determining resource allocation. In the case of malaria, knowing how people are dying can save lives by allowing funders and governments to understand the true burden of disease and enabling them to realign their priorities and scale up programs. Still, an empirical evidence base for global health resource allocation will be absent as long as the vast majority of causes of deaths go unrecorded or miscoded.