Cause-of-death data for many developing countries are not available. Information on deaths in hospital by cause is available in many low- and middle-income countries but is not a representative sample of deaths in the population. We propose a method to estimate population cause-specific mortality fractions (CSMFs) using data already collected in many middle-income and some low-income developing nations, yet rarely used: in-hospital death records.
Methods and Findings
For a given cause of death, a community's hospital deaths are equal to total community deaths multiplied by the proportion of deaths occurring in hospital. If we can estimate the proportion dying in hospital, we can estimate the proportion dying in the population using deaths in hospital. We propose to estimate the proportion of deaths for an age, sex, and cause group that die in hospital from the subset of the population where vital registration systems function or from another population. We evaluated our method using nearly complete vital registration (VR) data from Mexico 1998–2005, which records whether a death occurred in a hospital. In this validation test, we used 45 disease categories. We validated our method in two ways: nationally and between communities. First, we investigated how the method's accuracy changes as we decrease the amount of Mexican VR used to estimate the proportion of each age, sex, and cause group dying in hospital. Decreasing VR data used for this first step from 100% to 9% produces only a 12% maximum relative error between estimated and true CSMFs. Even if Mexico collected full VR information only in its capital city with 9% of its population, our estimation method would produce an average relative error in CSMFs across the 45 causes of just over 10%. Second, we used VR data for the capital zone (Distrito Federal and Estado de Mexico) and estimated CSMFs for the three lowest-development states. Our estimation method gave an average relative error of 20%, 23%, and 31% for Guerrero, Chiapas, and Oaxaca, respectively.
Where accurate International Classification of Diseases (ICD)-coded cause-of-death data are available for deaths in hospital and for VR covering a subset of the population, we demonstrated that population CSMFs can be estimated with low average error. In addition, we showed in the case of Mexico that this method can substantially reduce error from biased hospital data, even when applied to areas with widely different levels of development. For countries with ICD-coded deaths in hospital, this method potentially allows the use of existing data to inform health policy.
Working in Mexico and using vital registration data, Chris Murray and colleagues achieved encouraging results with a new method to estimate population cause-specific mortality fractions.
Governments and international health agencies need accurate information on the leading causes of death in different populations to help them develop and monitor effective health policies and programs. It is pointless investing money in screening programs for a type of cancer in a country where that cancer is very rare, for example, or setting up treatment centers for an infectious disease in a region where the disease no longer occurs. In developed countries, most deaths are recorded in vital registration (VR) systems. These databases record the specific cause of death, which is assigned by doctors using the International Classification of Diseases (ICD), an internationally agreed-upon list of codes for hundreds of diseases. Across the developing world, however, only one death in four is recorded by VR systems; in some very poor countries, only one death in 20 is recorded accurately. With this paucity of cause-of-death data, developing countries cannot make good decisions about how to spend their limited resources.
Why Was This Study Done?
The establishment of full VR systems in all developing countries will take time and may not always be possible, but many of these nations already collect ICD-coded data on in-hospital deaths. Unfortunately, this information does not accurately reflect the causes of death across whole populations. For example, the diseases that affect rich people differ from those that affect poor people, and rich people are more likely to die in hospital than poor people. Thus, although for each cause of death, the number of deaths in hospital equals the total number of deaths in the community multiplied by the proportion of deaths occurring in hospital, this proportion is different for each cause. If these proportions could be estimated, then in-hospital death records could be used to determine the fraction of the population that dies from each cause—the population's “cause-specific mortality fractions” (CSMFs). In this study, the researchers have devised a method that allows them to do this, and have used near-complete VR data collected between 1998 and 2005 in Mexico to test their method.
What Did the Researchers Do and Find?
The researchers developed a mathematical method that estimates the proportion of deaths occurring in hospitals for people grouped together by their age, sex, and cause of death (an “age–sex–cause group”) using VR data from a subset of the whole population. They tested their method for 45 nonoverlapping but all-encompassing diseases using the Mexican VR data (which records when a person has died in the hospital). They found that if they decreased the amount of VR data used to estimate the proportion of each age, sex, cause group dying in hospital from 100% to 9%, the maximum relative error between the true and estimated CSMFs was only 12%. When they just used the VR information from the capital city (9% of the population), the average relative error in CSMFs (a measure of how much the estimated and true CSMFs differ) across all 45 causes of death was only 10%. Finally, when they used VR data for the main urban area of Mexico (where access to hospitals is good) to estimate CSMFs for the three least developed states of Mexico, the average relative errors were 20%, 23%, and 31%.
What Do These Findings Mean?
These findings indicate that the researchers' method can provide accurate estimates of population CSMFs using ICD-coded cause-of-death data from deaths in hospital and VR data that cover part of the population. Even when the VR data from a developed area are used to calculate the CSMFs in a poorly developed area, the method produces a more accurate estimate than in-hospital death data used alone. Because the researchers have only tested their method for one country, additional “validation studies” need to be done using data from other countries with a good-quality VR system. If the method does work in these other settings, then existing data on in-hospital deaths could be used to determine the leading causes of death in countries with poor VR systems. Such information would be invaluable in establishing effective health policies.
Please access these Web sites via the online version of this summary at http://dx.doi.org/10.1371/journal.pmed.0040326.
• An accompanying paper by the same authors describes an alternative approach to collecting accurate cause-of-death data in developing countries
• The World Health Organization provides information on health statistics and health information systems, on the International Classification of Diseases, and on the Health Metrics Network, a global collaboration focused on improving sources of vital statistics and cause-of-death data
• Grand Challenges in Global Health provides information on research into better ways for developing countries to measure their health status