ABSTRACT:

Objective:

To assess time trends in maternal mortality in Brazil and its five geographical regions from 2001 to 2012, as well as to describe its main causes.

Methods:

This is a time series analysis, from data obtained in the Mortality Information System (SIM) and in the Live Births Information System (SINASC). Maternal mortality ratio (MMR) and causes of maternal death were described according to the categories of the Tenth Revision of the International Classification of Diseases, in the years 2001, 2006, and 2011. To estimate time trends, linear regression was used.

Results:

The highest MMR in Brazil was observed in 2009 (77.31 per 100,000 live births). A significant decreasing trend was observed in the Northeast and South regions and a significant increasing trend was found in the Midwest. There was an increase in deaths from other obstetric conditions and a decrease in deaths from edema, proteinuria and hypertensive disorders.

Conclusion:

Although there was a decreasing trend in the MMR in the Northeast and South regions, the high ratio observed in Brazil reveals the need to improve health care in prenatal, childbirth, and puerperium periods.

Keywords:
Maternal mortality; Causes of death; Health Information Systems; Women's health; Prenatal care; Epidemiology

INTRODUCTION

According to the International Statistical Classification of Diseases and Related Health Problems (ICD-10) and the World Health Organization (WHO), maternal mortality is defined as:

After 1980, with the adoption of policies and health actions for women, maternal mortality started to gain prominence in the scenario of health policies in Brazil²2. Victora CV, Aquino EM, Leal MC, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. Lancet 2011; 377(9780): 1863-76.. Given its magnitude and challenges for its reduction, especially in countries with low and middle incomes, the United Nations (UN) in 2000, in the Millennium Declaration, approved the "millennium development goals" of which the fifth is "improve the health of women," one of its components being the reduction of maternal mortality by three quarters until 2015³3. United Nations. General Assembly - Fifty-fifth session - Resolution adopted by the General Assembly: United Nations Millennium Declaration. New York. ONU; 2000.. In a study with data from approximately 180 countries between 1990 and 2013, Kassebaum et al. estimated that only 16 countries, of these only 7 being low- and middle-income countries, would reach the fifth "millennium development goal."⁴4. Kassebaum NJ, Bertozzi-Villa A, Coggeshall MS, Shackelford KA, Steiner C, Heuton KR, et al. Global, regional, and national levels and causes of maternal mortality during 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014; 384(9947): 980-1004. Victora et al. reported, in 2011, that Brazil possibly would not reach this reduction target of three quarters of maternal mortality rate in the period of 1990 to 2015²2. Victora CV, Aquino EM, Leal MC, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. Lancet 2011; 377(9780): 1863-76.. Also, in the recently published final report of the "Countdown to 2015," it can be seen that Brazil has not reached that goal⁵5. Countdown to 2015. A decade of tracking progress for maternal, newborn and child survival: the 2015 report. Washington, DC: World Health Organization e UNICEF; 2015..

In Brazil, maternal mortality rates are extremely high and have wide disparities between geographical regions⁶6. Laurenti R, Jorge MH, Gotlieb SL. A mortalidade materna nas capitais brasileiras: algumas características e estimativa de um fator de ajuste. Rev Bras Epidemiol 2004; 7(4): 449-60.. Some estimates suggest a decline in the rates²2. Victora CV, Aquino EM, Leal MC, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. Lancet 2011; 377(9780): 1863-76.. However, over the years, information on maternal mortality rates in the country have suffered from underreporting of deaths and inadequate report of recorded causes of death⁷7. Barros FC, Matijasevich A, Requejo JH, Giugliani E, Maranhão AG, Monteiro CA, et al. Recent trends in maternal, newborn, and child health in Brazil: progress toward millennium development goals 4 and 5. Am J Public Health 2010; 100(10): 1877-89.. In recent years, a series of government initiatives were carried out in order to improve the notification system⁸8. Brasil. Manual dos comitês de mortalidade materna. Brasília: Ministério da Saúde; 2009.. Thus, studies of time trends in maternal mortality, which do not take into account these changes in the surveillance system, may have inaccurate results.

In order to improve the quality of information, Laurenti et al. and Luizaga et al. proposed correction factors to minimize the underreporting of maternal deaths. These factors are applied to the total of official maternal deaths, so that the estimated indicators are closer to actual values⁶6. Laurenti R, Jorge MH, Gotlieb SL. A mortalidade materna nas capitais brasileiras: algumas características e estimativa de um fator de ajuste. Rev Bras Epidemiol 2004; 7(4): 449-60.^,99. Luizaga CT, Gotlieb SL, Jorge MH, Laurenti R. Mortes maternas: revisão do fator de correção para os dados oficiais. Epidemiol Serv Saude 2010;19(1):7-14.. Knowing the urgency of this subject in Brazil, the possibility of using these correction factors to obtain more accurate estimates of how maternal mortality rates have behaved over the last few years, as well as to know the regional differences and the main causes, is extremely important to the public health policies of the country and for the health services planning. Therefore, the objective of the study is to assess time trends in maternal mortality in Brazil and in its five geographical regions from 2001 to 2012, as well as to describe the main groups of maternal death causes.

METHODS

A time series study with secondary data analysis was carried out, covering the period of 2001 - 2012, from health information available in the database of DATASUS (Information Technology Department of the Brazilian Public Health Care System)¹⁰10. Ministério da Saúde. DATASUS. Informações de Saúde (TABNET). Disponível em: http://www2.datasus.gov.br/DATASUS/index.php?area=02 (Acessado em: 15 de novembro de 2014).
http://www2.datasus.gov.br/DATASUS/index... . For each year studied, the data were obtained in the Mortality Information System (SIM) and on the Information System on Live Births (SINASC), that carry out systematic records of data on mortality and survival by means of the Declaration of Deaths¹¹11. Ministério da Saúde. DATASUS. Mortalidade. Notas Técnicas. Disponível em: http://tabnet.saude.es.gov.br/cgi/tabnet/sim/sim96/obtdescr.htm (Acessado em: 20 de novembro de 2014).
http://tabnet.saude.es.gov.br/cgi/tabnet... and the Declaration of Liveborn Infants¹²12. Ministério da Saúde. DATASUS. Nascidos Vivos. Notas Técnicas. Disponível em: http://tabnet.saude.es.gov.br/cgi/tabnet/sinasc/nvdescr.htm (Acessado em: 20 de novembro de 2014).
http://tabnet.saude.es.gov.br/cgi/tabnet... .

The maternal mortality rate was estimated from the definition recommended by the WHO¹1. World Health Organization, UNICEF, UNFPA and The World Bank. Trends in maternal mortality: 1990 to 2010 - WHO, UNICEF, UNFPA and The World Bank estimates. Geneva: World Health Organization; 2012., and corresponds to the Chapter XV of the ICD-10 - "Pregnancy, childbirth and puerperium" - and the other deaths considered maternal, but classified in other chapters. The number of deaths considered maternal was selected, according to place of residence (region), regardless of age. Maternal deaths from obstetric causes in the period between 42 days and less than a year after birth are considered late deaths (codes O96 and O97 of ICD-10) and were not addressed in this study.

The maternal mortality ratio (MMR) represents the obstetric risk associated with pregnancy and is defined by the ratio between the number of maternal deaths during the period of a year per 100,000 live births during the same period¹³13. World Health Organization. Health statistics and health information systems. Maternal mortality ratio (per 100 000 live births). Disponível em: http://www.who.int/healthinfo/statistics/indmaternalmortality/en/ (Acessado em: 29 de janeiro de 2015).
http://www.who.int/healthinfo/statistics... . To calculate the MMR, births (obtained through SINASC) and maternal deaths (obtained through SIM) were used according to the Brazilian region of the mother's residence. The MMR was calculated for the country and according to the five Brazilian regions (North, Northeast, Southeast, South, and Midwest). The causes of maternal death were also described regarding Brazil and its regions, in the years 2001, 2006, and 2011, according to the categories of ICD-10, as follows:

Maternal disorders predominantly related to pregnancy, HIV, behavioral syndromes associated with physiological disturbances, and physical factors as well as other endocrine glands disorders were grouped into a single category defined as "other causes." In addition, the main causes of maternal death in the categories of ICD-10 that had higher levels of maternal mortality in the 3 years observed were described.

To calculate the mortality rates, adjustment factors to maternal deaths were used. In the period from 2001 to 2007, correction factors proposed by Laurenti et al.⁶6. Laurenti R, Jorge MH, Gotlieb SL. A mortalidade materna nas capitais brasileiras: algumas características e estimativa de um fator de ajuste. Rev Bras Epidemiol 2004; 7(4): 449-60. were used for the Brazilian regions and Brazil; thus, 1.08 - North; 1.76 - Northeast; 1.10 - Midwest; 1.35 - Southeast; 1.83 - South; and 1.40 - Brazil. For the period from 2008 to 2012, the factors updated by Luizaga et al.⁹9. Luizaga CT, Gotlieb SL, Jorge MH, Laurenti R. Mortes maternas: revisão do fator de correção para os dados oficiais. Epidemiol Serv Saude 2010;19(1):7-14. were applied. The factor of 0.93 was applied for the North region, 1.17 for Northeast, 1.47 for Midwest, 1.29 for Southeast, 1.1 for South, and 1.19 for Brazil. The use of factors throughout the period was established in accordance with those used by the Ministry of Health¹⁴14. Ministério da Saúde. DATASUS. Indicadores de mortalidade. Razão de mortalidade materna. Disponível em: http://tabnet.datasus.gov.br/cgi/idb2011/C03b.htm (Acessado em: 19 de outubro de 2014).
http://tabnet.datasus.gov.br/cgi/idb2011... .

The change in the MMR was calculated through the absolute and relative difference by comparing the year 2012 to 2001. To estimate the time trend in MMR, from 2001 to 2012, linear regression was used. All the analyses were based on public data available on the Internet, with unrestricted access data, and without identifying individuals. Data were analyzed using the software Stata version 12.0.

Due to the use of data of public domain, with unrestricted access, the analysis on ethics which is carried out by the system Comitê de Ética em Pesquisa /CONEP is dispensed by the Comissão Nacional de Ética em Pesquisa (CONEP) to conduct this study.

RESULTS

The MMR values from 2001 to 2012 in Brazil, and in its regions, are shown in Figure 1. The highest MMR in Brazil was found in 2009 (77 per 100,000 live births), and the lowest in 2012 (65 per 100,000 live births).

Among the regions of the country, from 2001 to 2007, the Northeast and South showed the highest MMR values. From 2008 to 2012, the highest values of MMR were found in the Midwest and Northeast. Within the study period, the lowest values of MMR for each year varied between regions. From 2009 to 2012, the South and the North regions had the lowest values of MMR in the country. The highest MMR, among the five regions in the study period, was found in 2005 in the Northeast (118 per 100,000 live births), and the lowest in 2011 in the South (49 per 100,000 live births).

Analyzing the whole period from 2001 to 2012, no significant trend of increase or decrease in MMR in the North and Southeast regions and Brazil itself was found. However, a significant downward trend in the MMR in the study period for the Northeast and South regions was found, showing a reduction of 3 and 6 maternal deaths per 100,000 live births for each year, respectively. Besides, a significant increase of MMR trend was detected for the Midwest region in the study period, an increase of 3 maternal deaths per 100,000 live births every year.

The relative (Δ%) and absolute (Δ absolute) changes in MMR values throughout the study period can be observed in Figure 2. The Southern region was the one that presented the highest relative change (-45%), corresponding to a decrease of 44 deaths per 100,000 live births. On the other hand, the Midwest region was the one that presented the highest percentage increase of MMR in 2012 compared to the year 2001 (42%), which corresponded to an increase of 25 deaths per 100,000 live births. Brazil presented a percentage reduction in MMR of 9% in 2012 compared to the year 2001, corresponding to a decrease of 6 deaths per 100,000 live births.

Figure 3 presents the proportional mortality for the main groups of maternal mortality causes by regions of Brazil, categorized by codes of Chapter XV of ICD-10, in 2001, 2006, and 2011. Edema, proteinuria, and hypertensive disorders (O10-O16) appear to be the main causes of maternal death in the North in 2001 (24%), 2006 (30%), and 2011 (27%), in the Northeast and Southeast regions in the years 2001 (24% and 26%, respectively) and 2006 (30% and 25%, respectively), and in the Midwest region in 2001 (28%) and 2011 (32%). Other obstetric conditions (O94-O99) were the leading cause of maternal death in the South in the 3 years analyzed (24, 32, 38%), in the Northeast and Southeast in 2011 (26 and 32 %, respectively) and in the Midwest region in 2006 (34%).

Moreover, an increase in those 3 years in maternal deaths due to other obstetric conditions (O94, O95, O98, O99) in all regions, except the Midwest, was observed. It was also detected a decrease in maternal deaths linked to maternal care related to the fetus and amniotic cavity and possible delivery problems. In Brazil, there was an increase of maternal deaths for other obstetric conditions (O94, O95, O98, O99) during the study period, as well as a decrease in maternal deaths by edema, proteinuria, and hypertensive disorders (O10-O16) and complications related predominantly with the puerperium (O85-O92).

Analyzing the leading cause of maternal mortality classified within the ICD-10 groups with the highest number of deaths in 2001, 2006, and 2011, it was found that eclampsia (O15) was the leading cause of maternal mortality in the North in the 3 years studied, in the Northeast and Southeast in 2001 and 2006, and in the Midwest in 2001. Other diseases of the mother complicating the pregnancy, childbirth, and the puerperium (O99) were the leading cause of maternal mortality in the South region during the study period, in 2006 for the Midwest region, and 2011 for the Northeast and Southeast regions. Gestational hypertension with significant proteinuria (O14) was the main cause of maternal mortality in the Midwest in 2011. For Brazil, it was found that eclampsia (O15) was the major cause in 2001 and 2006; and other diseases of the mother (O99) in 2011.

DISCUSSION

According to the results of the study, between 2001 and 2012, there was an increase in the MMR in the Midwest region and a reduction in the Northeast and South. In the other regions and in Brazil, the values of MMR were stable. With regard to the main causes of maternal death, for all regions stood out those of the edema, proteinuria and hypertensive disorders group, and other obstetric conditions group.

The maternal mortality rate is an important indicator of health and it reflects the quality of attention to women's health. The evaluation and monitoring of this indicator is very important because high MMR values are associated with inadequate provision of health services to this group, from family planning and prenatal care to the puerperium¹⁵15. Rede Interagencial de Informação para a Saúde. Indicadores básicos para a saúde no Brasil: conceitos e aplicações. 2. ed. Brasília: Organização Pan-Americana da Saúde; 2008.. However, the time trends assessment in maternal mortality is difficult and complex, given the absence and/or underreporting of data in some locations, as well as the reliability of the information⁶6. Laurenti R, Jorge MH, Gotlieb SL. A mortalidade materna nas capitais brasileiras: algumas características e estimativa de um fator de ajuste. Rev Bras Epidemiol 2004; 7(4): 449-60.. Furthermore, the use of secondary data implies some limitations. In addition to regional differences and persistent underreporting¹⁶16. Soares VMN, Azevedo EMM, Watanabe TL. Subnotificação da mortalidade materna no Estado do Paraná, Brasil: 1991-2005. Cad Saude Publica 2008; 24(10): 2418-26., the information is not collected to assess the objective of the study in question, meaning that the researcher has little control over the data¹⁷17. Bittencourt SA, Camacho LA, Leal MC. Hospital Information Systems and their application in public health. Cad Saude Publica 2006; 22(1): 19-30..

Nevertheless, this scenario has been changing in recent years. As a result of political movements to increase the visibility of maternal deaths, a series of actions were carried out by providing improvements in the system of registration and notification of deaths, including the compulsory investigation of deaths of women in reproductive age²2. Victora CV, Aquino EM, Leal MC, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. Lancet 2011; 377(9780): 1863-76.. Moreover, in order to improve the quality of information and minimize underreporting, correction factors⁶6. Laurenti R, Jorge MH, Gotlieb SL. A mortalidade materna nas capitais brasileiras: algumas características e estimativa de um fator de ajuste. Rev Bras Epidemiol 2004; 7(4): 449-60.^,99. Luizaga CT, Gotlieb SL, Jorge MH, Laurenti R. Mortes maternas: revisão do fator de correção para os dados oficiais. Epidemiol Serv Saude 2010;19(1):7-14. have been proposed and used in the present study. These advances, though positive, implicate a cautious interpretation of time trend analysis and regional maternal mortality indicator differences.

At the global level, two major systematic reviews on maternal mortality in the world and its main causes were published. The first gathered data from 115 countries and has accounted 60,799 deaths between the years 2002 and 2009, with significant differences between the regions of the world¹⁸18. Say L, Chou D, Gemmill A, Tunçalp Ö, Moller AB, Daniels J, et al. Global causes of maternal death: a WHO systematic analysis. Lancet Glob Health 2014; 2(6): e323-33.. In Latin America, for example, it was observed that the greater proportion of maternal deaths occurred due to bleeding (23.1%; 95%CI 19.7 - 27.8), followed by hypertension. The second review showed maternal mortality trajectories between 1990 and 2013, detailing the data for countries within each region⁴4. Kassebaum NJ, Bertozzi-Villa A, Coggeshall MS, Shackelford KA, Steiner C, Heuton KR, et al. Global, regional, and national levels and causes of maternal mortality during 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014; 384(9947): 980-1004.. This study demonstrated that there are still high rates in low- and middle-income countries, although the absolute number of deaths worldwide has decreased from 376,034 (95%CI 343,483 - 407,574) in 1990 to 292,982 (95%CI 261,017 - 327,792) in 2013.

In Brazil, published data are discrepant with respect to trends in MMR. In a document on maternal mortality trends from 1990 to 2010, WHO, UNICEF (United Nations Children's Fund), and The World Bank Estimates indicate a decline in MMR in Brazil, from 120 deaths in 1990 to 56 deaths in 2010, per 100,000 live births, with a decrease of 3.5 deaths every year¹1. World Health Organization, UNICEF, UNFPA and The World Bank. Trends in maternal mortality: 1990 to 2010 - WHO, UNICEF, UNFPA and The World Bank estimates. Geneva: World Health Organization; 2012.. While the Brazilian Ministry of Health, based on official statistics, reports stable values of MMR since 1996, with a MMR around 50 deaths per 100,000 live births²2. Victora CV, Aquino EM, Leal MC, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. Lancet 2011; 377(9780): 1863-76., corroborating the results found in this study.

In this study, the Northeast and Midwest regions showed the highest values of MMR. The striking socioeconomic differences among Brazilian regions may explain part of the differences found in the MMR. The Northeast being one of the regions with the highest rate of underreporting of MMR¹⁵15. Rede Interagencial de Informação para a Saúde. Indicadores básicos para a saúde no Brasil: conceitos e aplicações. 2. ed. Brasília: Organização Pan-Americana da Saúde; 2008., having a higher percentage of illiteracy, the largest population in vulnerability for economic and social deprivation¹⁹19. Instituto Brasileiro de Geografia e Estatística (IBGE). Síntese de indicadores sociais - Uma análise das condições de vida da população brasileira 2013. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2013., in addition to the difficulties in the access and use of health services²⁰20. Andrade MV, Noronha KV, Menezes RM, Souza MN, Reis CB, Martins DR, et al. Equidade na utilização dos serviços de saúde no Brasil: um estudo comparativo entre as regiões brasileiras no período 1998-2008 - TD 445. Belo Horizonte: CEDEPLAR/UFMG; 2011., it is expected that the MMR could still be higher than that obtained by means of the data used in this study.

A reduction by nearly half of the 2012 rate (64,83 deaths per 100,000 live births) is necessary for Brazil to achieve the fifth "millennium development goal" that is, three-quarters reduction in maternal mortality from 1990 to 2015, corresponding to a rate of 35 maternal deaths for every 100,000 births. Unfortunately, the challenge of significantly reducing maternal mortality rates in the country remains.

Regarding the causes of maternal death, it is important to distinguish between the causes of direct obstetric death - that are the result of complications of pregnancy, childbirth, and the postpartum period due to interventions, the lack of or inadequate treatment, or the chain of events that can elapse during these - and the causes of indirect obstetric death²¹21. Cross S, Bell JS, Graham WJ. What you count is what you target: the implications of maternal death classification for tracking progress towards reducing maternal mortality in developing countries. Bull World Health Organ 2010; 88(2):147-53.. The latter are the result of previously existing diseases or diseases developed during pregnancy, which are not direct obstetric causes, but that have their situation aggravated by gestation²¹21. Cross S, Bell JS, Graham WJ. What you count is what you target: the implications of maternal death classification for tracking progress towards reducing maternal mortality in developing countries. Bull World Health Organ 2010; 88(2):147-53.. When analyzing the main cause within the group of classification of ICD-10 with higher proportional maternal mortality rate in the regions and in Brazil in the years 2001, 2006, and 2011, it was found that for Brazil and for the Northeast and Southeast regions, in the first two years, the main cause was by direct obstetric death, whereas in 2011 it changed to indirect obstetric death. However, in the North and Midwest in 2011 the main cause consisted of directly related events in pregnancy, childbirth, and postpartum. The interpretation of these data should be cautious; however, it is evident that the planning of public policies and health services in relation to maternal health must be specific to each region, considering the differences in the rates and causes of maternal mortality between them over the past years.

CONCLUSION

The maternal mortality rate must be a health priority in Brazil, in view of the high values of MMR found in all regions, especially in the Midwest, and the absence of significant reduction in the MMR in recent years, even after confirmation of its importance in establishing the "millennium development goals." Furthermore, it is known that the majority of maternal deaths could be prevented if there was a good quality in health services, if they were comprehensive and interdisciplinary, ranging from family planning, prenatal care, delivery, to the postpartum period, as well as an awareness of mothers for the importance of medical assistance in these periods.

References

Publication Dates

History