For a better experience, click the Compatibility Mode icon above to turn off Compatibility Mode, which is only for viewing older websites.

Understanding mortality data in Latin American cities

A conversation with Dr. Bernardo Lanza Queiroz and Dr. Usama Bilal

Posted on August 1, 2019
Bernardo Queiroz and Usama Bilal, headshots

How long we live and what we die from is affected by the neighborhood, city, and country where we live. Yet all too often, data on life expectancy and causes of death are only available at the national level. These summaries don’t show the significant variability among cities -both within and across countries- and as a result they make it hard to develop policies and plans that effectively address these inequalities.

Researchers Bernardo Lanza Queiroz and Usama Bilal are working to tackle issues with mortality data across Latin America. Katy Indvik of SALURBAL and LAC-Urban Health asked them about their experiences.

LAC-Urban Health: Where do data on life expectancy and causes of death come from in Latin America? Who is responsible for collecting this information?

Bernardo Lanza Queiroz: Most mortality data in Latin America come from vital registration systems and are organized by national statistical offices or Ministries of Health – or by both. Countries collect information on age, sex and causes of death, but time series vary a lot by country. For instance, Mexico and Chile have a very long series of mortality data, whereas Bolivia has a shorter and irregular series. Data for Brazil is available starting in 1979.

LAC-Urban Health: When we do have them, what can local level mortality data tell us about a given city or neighborhood? What might local authorities do with this information? 

Usama Bilal: Mortality data can serve as a useful indicator of the health status of a city or its neighborhoods, which can help local authorities prioritize resources to improve the health of populations and evaluate the effects of policies or other interventions.

BLQ: Small-area mortality estimates (for all causes or cause-specific) are relevant to understanding the potential impacts of public health measures and for investigating regional variations in health and mortality. Better public health planning is possible with adequate mortality estimates at the local level – for example, in Brazil there have been some interesting studies on the impact of policies on violent causes of death.

LAC-Urban Health: What are (some of) the greatest challenges to using data on mortality in Latin America?

UB: I would say the biggest challenge we are facing is undercounting, or the lack of complete coverage of all deaths - the phenomenon that leads to deaths not being registered for one reason or another. This can lead to a miscalculation of the health status of a city or a neighborhood, which means this information is of less value unless methods are used to correct for these issues.

BLQ: A main issue is still data completeness. This has been improving over recent years, but still needs to be addressed by governments, and should be taken into account by researchers.

LAC-Urban Health: Do issues with data quality (including undercounting, for example) tend to concentrate in specific areas? What do we know about these areas and what are the potential consequences of having poor quality data about them?

UB: We have seen that some cities in Peru and Colombia tend to have a lower coverage of deaths, or, in other words, that a higher proportion of their deaths (about 40% and 20% on average, respectively) do not end up in their vital registration system. If we do not correct for this, certain Peruvian and Colombian cities would appear to have a better health status than they do. Some of these cities are actually among the poorest in the region, and as a result of undercounting, we would not be able to see the cross-city inequalities that exist in reality.

BLQ: I know more about the case of Brazil, and I recently collaborated with colleagues from Ecuador, analyzing data quality at the regional level. There are large differences across regions in Brazil and Ecuador, although they are decreasing overtime – such as regions in north and northeast Brazil that had very poor data quality in the 1980s but have improved a lot over the years. The main issue is using appropriate demographic and statistical methods to adjust for data quality and perform a proper analysis. We do need to be very careful when using mortality data, but there are ways to generate good estimates.

LAC-Urban Health: Can you think of any examples of “best practices”- where a city or country has successfully made changes to improve mortality data collection?

BLQ: I think Brazil is a good example of best practices. First, all data is publicly available. Second, the Ministry of Health invests substantially in supporting data collection, to improve the quality of causes of death information and ensure that all events are registered. The Mortality Information System in Brazil is also available to health policy actors.

Interested in hearing more?

On September 4, 2019, Bernardo Lanza Queiroz, Usama Bilal and other experts will at Drexel University in Philadelphia, Pennsylvania as part of: “Levels and determinants of urban mortality: A focus on cities of Latin America”. The half-day of discussion will center on identifying health lessons from available mortality data across Latin American cities, and options for managing issues with data quality and completeness. Registration for the event is still open!


Picture of city and slum divided landscape  

Bernardo Lanza Queiroz, MA, PhD specializes in economic demography, population aging, and mortality and health. He is an Associate Professor at the Department of Demography and researcher at the Center for Development and Regional Planning (CEDEPLAR) at the Federal University of Minas Gerais in Brazil. Dr. Lanza Queiroz’s work focuses on demographic methods, indirect techniques, and regional and urban economics.  Currently, his research is centered on two main topics: how demographic changes are related to the changes in the labor market in developing countries; and small-area mortality estimations using defective data. He is also working to create a Latin America Human Mortality Database, with the goal of supporting researchers, students, policy makers and the general public in the study of mortality in the region.

Usama Bilal, MD, MPH, PhD specializes in the macrosocial determinants of health, including the health consequences of urban and neighborhood characteristics and challenges with using mortality data at the local, city level. Dr. Bilal is an Assistant Professor of Epidemiology at the Department of Epidemiology and Biostatistics and the Urban Health Collaborative at the Dornsife School of Public Health at Drexel University, where he was previously a postdoctoral research fellow. He obtained his Ph.D. in Epidemiology (with a concentration in Cardiovascular Epidemiology) at the Johns Hopkins Bloomberg School of Public Health, his MPH at the Escuela Nacional de Sanidad, and his MD at the Facultad de Medicina of the Universidad de Oviedo in Spain. Dr. Bilal is part of the Salud Urbana en América Latina(SALURBAL) research project, which focuses on understanding the complex drivers and systems that affect urban health in Latin American cities.

This post was written as a contribution to Cities, Sectors, and Health, run by SALURBAL. To contact the blog or learn more about the SALURBAL project email

Posted in localnews, Health Equity