IHME bridges troubling gaps in adult mortality estimates

New methods offer greater precision in measuring death rates, including HIV mortality

Image with link to full article
Read the article

Related Content:

April 13, 2010–Researchers will be able to more accurately estimate adult mortality with new methods developed by the Institute for Health Metrics and Evaluation (IHME) at the University of Washington and collaborators at the University of Queensland.

The new analytical techniques address problems that have prevented health surveys from being used as an accurate source of adult mortality data – problems that have led to significant gaps in the mortality picture. They also give researchers a tool for directly measuring the impact of HIV instead of relying solely on theoretical models. The study, Measuring Adult Mortality Using Sibling Survival: A New Analytical Method and New Results for 44 Countries, 1974-2006, appears in the April 13, 2010, issue of the journal PLoS Medicine.

“While the research community has doubted the accuracy of measuring adult mortality using surveys, these new methods make us confident that you can,” said Dr. Christopher Murray, one of the paper’s authors and Institute Director. “We can now fill big holes in what we know about mortality around the world.”

In high-income and many middle-income countries, mortality rates are calculated from death registration systems. However, most low-income countries lack death registration systems, and researchers have to either create estimates based on child mortality rates or rely on surveys of siblings or others who have survived in families.

Surveys have yielded estimates in the past that were implausibly low, primarily for two reasons. Siblings often recall incorrectly the number of people who have died in their families and their ages at death, a problem known as recall bias. Another problem with surveys is that they often miss families that have been so devastated by mortality that there are few surviving members available to answer questions. This is known as survival bias.

IHME’s new method, called the Corrected Sibling Survival (CSS) method, compensates for both recall bias and survival bias, allowing a more precise estimate of death rates in a given country by age, sex, and time period. Based on these new methods, IHME will soon publish estimates for adult mortality worldwide. In the PLoS Medicine paper, IHME released data for a sample of 44 countries.

The data suggest that more adults are dying between the ages of 15 and 60 in developing countries than previously thought. This is especially true in countries with a high prevalence of HIV. Kenya, Zambia, Swaziland, and Tanzania all saw their death rates double between 1986 and 2006.

The new methods also show a sharp spike in mortality in Rwanda during the 1994 genocide, followed by a strong decline. The country’s death rate is now comparable to neighboring countries.

“This is a big breakthrough for public health research,” said Dr. Alan Lopez, one of the paper’s authors, an Affiliate Professor at IHME, and Head of the School of Population Health at the University of Queensland. “We can now see the effects of HIV, war, and a range of other public health threats with much greater precision than before.”

Full Citation: Obermeyer Z, Rajaratnam JK, Park CH, Gakidou E, Hogan MC, Lopez AD, Murray CJL. Measuring Adult Mortality Using Sibling Survival: A New Analytical Method and New Results for 44 Countries, 1974–2006. PLoS Medicine. 2010 Apr 13; 7(4):e1000260.

Top of Page

*Software capable of displaying a PDF is required for viewing or printing this document. You may download a free copy of Adobe Reader from the Adobe website at http://www.adobe.com/products/acrobat/readstep2.html.