Challenges related to records and quality of information in the Amazon

One of the United Nations’ sustainable development goals is the establishment of high-quality, valid, and reliable civil registry and vital statistics systems necessary for the design, evaluation, and implementation of social, economic and health programs, especially in the context of changes in the pattern of mortality that many countries have been experiencing. Although several methods developed in demography ^1,2 allow for following the evolution of the quality of the records and to indirectly estimate the levels of mortality in different localities, the quality of the system of vital statistics allows tracking the development of health conditions in a population, evaluating the public health policies, and assisting the health managers ^3,4,5.

Recently, the quality of mortality data in Brazil has improved substantially. Estimates of the degree of coverage of death registries in Brazil increased from about 80% in the 1980s to about 95% in 2010 ^6,7,8. However, regional variability in the quality of the degree of coverage of the death registry and the quality of information on the causes of death in the country is still high ^6,7,9. Since 2010, the South and Southeast regions show a complete coverage of adult mortality records. The states of the Northeast and North, even with the trends of improvements in recent years, have locations with low coverage ^6,10. The average coverage of mortality in the North Region rose from 65% to 76% from 1980 to 2010, whereas coverage in the South Region increased from 95% to 98% in the same period ¹⁰. In the case of smaller areas, such as mesoregions, the impacts of data quality can be even greater. The estimated degree of coverage for the São Paulo metropolitan area is 100%, high quality of data, with the probability of adult death of 0.2364, for individuals aged from 15 to 60 years. In the case of the South Amazonian mesoregion, the estimated degree of coverage is 68%. The observed data indicate a probability of adult death of 0.1102; however after applying the correction it is estimated at 0.1621 ¹⁰. An exercise with data from the 2010 Demographic Census indicates, for infant mortality, coverage of the events registration of practically 100% in the São Paulo metropolitan area compared with about 50% in the mesoregion of the South Amazon. In summary, directly analyzing the data, without applying correction methods, can lead to erroneous results directly affecting health policies. An important conclusion, however, is that the different estimation methods applied to limited data may present quite different results, complexifying the definition of health strategies. A study comparing different estimates of mortality for Brazil and its regions shows the great discrepancy of these estimates and highlights the importance of continuous investment in the quality of records and in the replicability of the methods used ¹¹.

In the specific case of the Amazon Region, the under-registration of deaths is high ^9,10. The comparative study of the registration of deaths from different sources (Demographic Census, Civil Registry, and Mortality Information System - SIM) for 2010 ¹² shows that in almost 65% of the municipalities of the North Region, the enumeration of deaths in the 2010 Demographic Census is higher than the registration of deaths in the SIM and in the Civil Registry. The study also shows that the census data have a better enumeration of events in locations with a higher rate of deprivation and worse social and economic conditions in the North Region. The release of the 2022 Demographic Census data will allow for similar analyses and assessments for the most recent period.

The improvement in the recording of events and the regional differentials tell only part of the story. In addition to enumerating the events, collecting additional information, especially the underlying causes of death is essential ^12,13,14. Over the last three decades, the percentage of deaths recorded as ill-defined causes in Brazil has decreased from 27.5% in 1980 to 8.5% in 2010. In the states of the Amazon, in the same period, about 13% of deaths are registered as ill-defined causes ¹⁰. In 2019, those were still 10% compared with less than 4% in the South Region.

Brazil is marked by great regional differences in the levels and trends of causes of death, reflecting different stages of the epidemiological transition process ^3,15,16. Considering the monitoring of the objectives of sustainable development ¹⁷, especially maternal and child health and the reduction of maternal mortality, regional variability is high ¹⁸. For example, in 2016, maternal mortality reached 64.4 per 100,000 live births in Brazil, ranging from 44.2 in the South to 84.5 per 100,000 live births in the North Region. In the state of Amapá, the rate reaches 141.7 per 100,000 live births. One of the great challenges of improving actions aimed at reducing maternal mortality is the lack of quality estimates that have regional details and that consider the temporal trend. The differential in the quality of the record of the specific cause of maternal death hinders the correct measurement of the level and trends ^19,20,21. The various methods applied based on Brazilian data may present very different results that have implications for the definitions of policies aimed at maternal and child health. Estimates based entirely on statistical models may present even more distinct results and in some cases have very varied trends among the studies ^22,23. Another example of the importance of the quality of the registry of causes of death in the Amazon Region refers to the monitoring and measurement of the impact of neglected tropical diseases, with a significant weight on mortality and years of life lost ^24,25.

Regarding more specific population groups, the limitations of the quality of vital statistics are particularly important for the adequate monitoring of the health conditions and mortality of indigenous populations. Since information on this group is very limited, most recent studies end up being based on information on deaths in households collected by the 2010 Demographic Census^26,27,28. Although the censuses offer an opportunity to study the original populations in the Amazon Region, the temporal frequency limits the adequate monitoring of the health condition of this population, further reinforcing the need to invest in the quality of the system of recording vital statistics for the different population sub-groups in Brazil.

Despite the improvement in the quality of death records in recent decades, the regional differential is still large and the limitations in the registration of underlying cause, age declaration, and various socioeconomic information are significant. The various methodological advances in the estimates presented by researchers and agencies, despite their importance, have limitations and do not replace quality information systems. In short, all efforts should be directed toward producing high-quality records, including the cause of death. Advances in registries can produce better and more reliable mortality estimates for understanding trends and differentials for several developing countries ^5,29. And, especially in the Amazon Region, these advances are indispensable for knowing the real panorama of health of the populations.

Acknowledgments

I thank Professor Carla Machado Jorge for her comments and suggestions. B. L. Queiroz is a fellow in productivity from the Brazilian National Research Council (CNPq).

Referências

• ¹
  Hill K, Choi Y, Timaeus IM. Unconventional approaches to mortality estimation. Demogr Res 2005; 13:281-99.
• ²
  Hill K, You DZ, Choi YJ. Death distribution methods for estimating adult mortality: sensitivity analysis with simulated data errors. Demogr Res 2009; 21:235-54.
• ³
  Aburto JM, Calazans J, Lanza Queiroz B, Luhar S, Canudas-Romo V. Uneven state distribution of homicides in Brazil and their effect on life expectancy, 2000-2015: a cross-sectional mortality study. BMJ Open 2021; 11:e044706.
• ⁴
  AbouZahr C, de Savigny D, Mikkelsen L, Setel PW, Lozano R, Nichols E, et al. Civil registration and vital statistics: progress in the data revolution for counting and accountability. Lancet 2015; 386:1373-85.
• ⁵
  Silva R. Population perspectives and demographic methods to strengthen CRVS systems: introduction. Genus 2022; 78:8.
• ⁶
  Queiroz BL, Freire FHMA, Gonzaga MR, Lima EEC. Completeness of death-count coverage and adult mortality (45q15) for Brazilian states from 1980 to 2010. Rev Bras Epidemiol 2017; 20:21-33.
• ⁷
  Paes NA. Avaliação da cobertura dos registros de óbitos dos estados brasileiros em 2000. Rev Saúde Pública 2005; 39:882-90.
• ⁸
  Adair T, Lopez AD. Estimating the completeness of death registration: an empirical method. PLoS One 2018; 13:e0197047.
• ⁹
  Queiroz BL, Lima EEC, Freire FHMA, Gonzaga MR. Temporal and spatial trends of adult mortality in small areas of Brazil, 1980-2010. Genus 2020; 76:36.
• ¹⁰
  Lima EEC, Queiroz BL. Evolution of the deaths registry system in Brazil: associations with changes in the mortality profile, under-registration of death counts, and ill-defined causes of death. Cad Saúde Pública 2014; 30:1721-30.
• ¹¹
  Queiroz BL, Gonzaga MR, Vasconcelos AMN, Lopes BT, Abreu DMX. Comparative analysis of completeness of death registration, adult mortality and life expectancy at birth in Brazil at the subnational level. Popul Health Metr 2020; 18:11.
• ¹²
  Diógenes VHD, Pinto Júnior EP, Gonzaga MR, Queiroz BL, Lima EEC, Costa LCC, et al. Differentials in death count records by databases in Brazil in 2010. Rev Saúde Pública 2022; 56:92.
• ¹³
  Borges GM. Health transition in Brazil: regional variations and divergence/convergence in mortality. Cad Saúde Pública 2017; 33:e00080316.
• ¹⁴
  França E, Abreu DX, Rao C, Lopez AD. Evaluation of cause-of-death statistics for Brazil, 2002-2004. Int J Epidemiol 2008; 37:891-901.
• ¹⁵
  Baptista EA, Queiroz BL, Pinheiro PC. Regional distribution of causes of death for small areas in Brazil, 1998-2017. Front Public Health 2021; 9:601980.
• ¹⁶
  França EB, Passos VMA, Malta DC, Duncan BB, Ribeiro ALP, Guimarães MDC, et al. Cause-specific mortality for 249 causes in Brazil and states during 1990-2015: a systematic analysis for the global burden of disease study 2015. Popul Health Metr 2017; 15:39.
• ¹⁷
  Teixeira RA, Ishitani LH, França E, Pinheiro PC, Lobato MM, Malta DC. Mortalidade por causas garbage nos municípios brasileiros: diferenças nas estimativas de taxas pelos métodos direto e Bayesiano de 2015 a 2017. Rev Bras Epidemiol 2021; 24:e210003.
• ¹⁸
  Silva BGC, Lima NP, Silva SG, Antúnez SF, Seerig LM, Restrepo-Méndez MC, et al. Mortalidade materna no Brasil no período de 2001 a 2012: tendência temporal e diferenças regionais. Rev Bras Epidemiol 2016; 19:484-93.
• ¹⁹
  Laurenti R, Mello-Jorge MHP, Gotlieb SLD. Reflexões sobre a mensuração da mortalidade materna. Cad Saúde Pública 2000; 16:23-30.
• ²⁰
  Laurenti R, Jorge MHPM, Gotlieb SLD. A mortalidade materna nas capitais brasileiras: algumas características e estimativa de um fator de ajuste. Rev Bras Epidemiol 2004; 7:449-60.
• ²¹
  Szwarcwald CL, Escalante JJC, Rabello Neto DL, Souza Junior PRB, Victora CG. Estimação da razão de mortalidade materna no Brasil, 2008-2011. Cad Saúde Pública 2014; 30 Suppl:S71-83.
• ²²
  Peterson E, Chou D, Moller AB, Gemmill A, Say L, Alkema L. Estimating misclassification errors in the reporting of maternal mortality in national civil registration vital statistics systems: a Bayesian hierarchical bivariate random walk model to estimate sensitivity and specificity for multiple countries and years with missing data. Stat Med 2022; 41:2483-96.
• ²³
  Leal LF, Malta DC, Souza MFM, Vasconcelos AMN, Teixeira RA, Veloso GA, et al. Maternal mortality in Brazil, 1990 to 2019: a systematic analysis of the Global Burden of Disease Study 2019. Rev Soc Bras Med Trop 2022; 55:e0279.
• ²⁴
  Martins-Melo FR, Carneiro M, Ramos Jr AN, Heukelbach J, Ribeiro ALP, Werneck GL. The burden of neglected tropical diseases in Brazil, 1990-2016: a subnational analysis from the Global Burden of Disease Study 2016. PLoS Negl Trop Dis 2018; 12:e0006559.
• ²⁵
  Martins-Melo FR, Ramos AN, Alencar CH, Heukelbach J. Trends and spatial patterns of mortality related to neglected tropical diseases in Brazil. Parasite Epidemiol Control 2016; 1:56-65.
• ²⁶
  Santos RV, Borges GM, Campos MB, Queiroz BL, Coimbra Jr. CEA, Welch JR. Indigenous children and adolescent mortality inequity in Brazil: what can we learn from the 2010 National Demographic Census? SSM Popul Health 2020; 10:100537.
• ²⁷
  Marinho GL, Borges GM, Paz EPA, Santos RV. Mortalidade infantil de indígenas e não indígenas nas microrregiões do Brasil. Rev Bras Enferm 2019; 72:57-63.
• ²⁸
  Campos MB, Borges GM, Queiroz BL, Santos RV. Diferenciais de mortalidade entre indígenas e não indígenas no Brasil com base no Censo Demográfico de 2010. Cad Saúde Pública 2017; 33:e00015017.
• ²⁹
  AbouZahr C, Boerma T. Health information systems: the foundations of public health. Bull World Health Organ 2005; 83:578-83.

Publication Dates

• Publication in this collection
  11 Aug 2023
• Date of issue
  2023

History

• Received
  28 May 2023
• Reviewed
  19 June 2023
• Accepted
  23 June 2023