Triple burden of mortality in municipalities in Minas Gerais according to the Global Burden of Disease study, 2000 to 2018

Abstract

To describe the temporal evolution of the main causes of mortality in Minas Gerais (MG), Brazil, and to verify the association with socioeconomic indicators. This is a mixed ecological study in which age-standardized mortality rates were calculated per 100,000 inhabitants due to noncommunicable diseases (NCDs), communicable, neonatal and nutritional diseases (NNDs) and external causes (ECs) for 853 municipalities in MG, according to data from the Global Burden of Disease (GBD) study, in the three-year periods 2000 to 2002 (T1), 2009 to 2011 (T2) and 2016 to 2018 (T3). Between T1 and T3, mortality due to NCDs predominated; there was a 22.4% decrease in the rates for NCDs (553.6 to 429.9) and a 29% decrease in the rates for NCDs (83 to 58.9), and a 3.5% increase in EC (62.2 to 64.4). The correlation coefficients were positive (R > 0.70; p < 0.05) and higher mortality rates were found in areas with worse socioeconomic status.

Resumo

Descrever a evolução temporal das principais causas de mortalidade em Minas Gerais (MG), Brasil, e verificar a associação com indicadores socioeconômicos. Estudo ecológico misto em que foram calculadas taxas de mortalidade padronizadas por idade, por 100 mil habitantes, por doenças crônicas não transmissíveis (DCNT), doenças transmissíveis, neonatais e nutricionais (TNN) e causas externas (CE), para 853 municípios de MG, segundo dados do estudo Carga Global de Doenças (GBD), nos triênios 2000 a 2002 (T1), 2009 a 2011 (T2) e 2016 a 2018 (T3). Entre T1 e T3 predominou a mortalidade por DCNT; houve decréscimo de 22,4% das taxas por DCNT (553,6 para 429,9) e de 29% da s taxas por TNN (83 para 58,9), e acréscimo de 3,5% por CE (62,2 para 64,4). Os coeficientes de correlação foram positivos (R > 0,70; p < 0,05) e foram encontrados taxas mais elevadas de mortalidade em áreas de pior status socioeconômico.

Palavras-chave:
Mortalidade; Doenças crônicas não transmissíveis; Violência; Doenças transmissíveis; Tripla carga de doenças; Estudos ecológicos

Resumen

Describir la evolución temporal de las principales causas de mortalidad en Minas Gerais (MG), Brasil, y verificar la asociación con indicadores socioeconómicos. Estudio ecológico mixto en el que se calcularon las tasas de mortalidad estandarizadas por edad, por 100.000 habitantes, por enfermedades crónicas no transmisibles (ENT), enfermedades transmisibles, neonatales y nutricionales (ENN) y causas externas (CE), para 853 municipios de MG, según datos del estudio Carga Global de Enfermedad (GBD), en los trienios 2000 a 2002 (T1), 2009 a 2011 (T2) y 2016 a 2018 (T3). Entre T1 y T3 predominó la mortalidad por ENT; Se observó una disminución del 22,4% en las tasas de ENT (553,6 a 429,9) y una disminución del 29% en las tasas de TNC (83 a 58,9), y un aumento del 3,5% en las de EC (62,2 a 64,4). Los coeficientes de correlación fueron positivos (R > 0,70; p < 0,05) y se encontraron tasas de mortalidad más altas en zonas de peor nivel socioeconómico.

Introduction

The demographic and epidemiological transitions observed on a global scale and currently underway in Brazil are causing changes in the morbimortality patterns of human populations and, consequently, requiring a reorganization of health agendas within health systems. The demographic transition theory builds on the change in the pattern of high to low birth and mortality rates¹. In developed countries, the epidemiological transition evolves with a substantial decline in infectious diseases and an increase in chronic non-communicable diseases (NCDs). In contrast, developing countries more frequently observe three concomitant causes of death groups, configuring the triple burden of diseases^2-4.

Different epidemiological transition profiles are also observed on a subnational scale. Social inequalities, cultural differences and different forms of organization and capacity of regional and local health systems⁵ can explain them. They are manifested by gender, age, ethnicity, and socioeconomic status⁶. In the 1990-2016 period, Brazil observed a 34% decline in the mortality rates from all causes (from 1,116.6 to 737.0 deaths per 100,000 inhabitants), albeit with significant variations between its federative units⁷. The most negligible decreases were observed in the North and Northeast⁷. Therefore, monitoring health indicators and identifying inequalities that widen social and health gaps within the same territory would be the first step in planning programmatic actions to reduce these events^7,8.

Geographic Information Systems (GIS) have been widely used in health data analysis because they enable the investigation of spatial patterns of events and their temporal development. Analyzing the behavior of health events over time - decline, increase, or stagnation - also provides important information regarding epidemiological variations resulting from historical sociocultural, demographic, environmental, and political changes. GIS also enable the analysis of impact, effectiveness, and the proposal of public policies.

Barreto and Carmo⁹ investigated the main causes of death and hospitalizations in Brazil in the 20^th century and explained overlapping patterns of illness and death in the country based on an analysis of the evolution of health problems over the years and decades⁹. Based on the analysis, the authors developed a reflection on the challenges and strategies within the Brazilian health system.

Soares Filho et al.¹⁰ described the temporal evolution of mortality rates for the three principal groups of causes of illness and death in Brazil between 1990 and 2019, identifying the progressive epidemiological transition and improved quality of death data in the country, with a change in this backdrop imposed by the COVID-19 pandemic. However, little literature addresses the triple burden of disease and mortality patterns in small Brazilian areas, which refers to a population subgroup or a geographic region¹¹.

Minas Gerais is a state with a large number of municipalities, and its territory reflects the regional heterogeneity observed in Brazil regarding economic and social development¹². Due to different socioeconomic development stages, such inequalities can result in the coexistence of different morbimortality settings¹⁰.

This study adds to the scientific efforts to analyze the spatial and temporal mortality pattern in small Brazilian areas. It aims to describe the distribution and temporal evolution of mortality by the primary cause groups in the municipalities of Minas Gerais (MG), Brazil, and verify the association with socioeconomic indicators.

Methods

Study design

This mixed, ecological, descriptive, analytical, epidemiological study estimated mortality in the 853 municipalities of the state of Minas Gerais (MG), Brazil, and the 13 intermediate geographic regions of the state, namely, Montes Claros, Teófilo Otoni, Governador Valadares, Ipatinga, Belo Horizonte, Barbacena, Juiz de Fora, Divinópolis, Pouso Alegre, Varginha, Uberaba, Uberlândia, and Patos de Minas (Figure 1).

Figure 1
Intermediate Geographic Regions of Minas Gerais.

Intermediate geographic regions correspond to a political-administrative section on an intermediate scale between the Federation Units (UF) and the Immediate Geographic Regions. They organize the territory to articulate the immediate regions through a higher hierarchy pole differentiated by the private and public management flows and more complex urban functions¹³.

This classification, established in 2017, corresponds to a review of the old mesoregions, in force since 1989¹³. In this work, we opted to use this regional division given by the Brazilian Institute of Geography and Statistics (IBGE), which is configured as a territorial base for collecting and disseminating official statistics and contributes to the action of state and municipal governments in implementing and managing public policies and administering investments^14,15.

Study location

Minas Gerais (MG) is a Brazilian state with a large territorial extension with 853 municipalities. It has significant disparities concerning the distribution of its population and demographic and socioeconomic characteristics^16,17. We can observe marked economic and social heterogeneity between the regions of Minas Gerais is similar to that observed among the Brazilian macro-regions, for which a clear division between North-Northeast and South-Southeast¹². The North of Minas Gerais and the Jequitinhonha/Mucuri Valley have the lowest development indicators in contrast to the Central, South, and Triângulo regions^12,16.

Mortality and socioeconomic disparity indicators and data source

Mean mortality rates per 100,000 inhabitants were standardized using the direct method and refer to the 2000-2002 (T1), 2009-2011 (T2), and 2016-2018 (T3) triennia. The three-year analysis was conducted to minimize fluctuations caused by small figures since 68.5% of the municipalities have less than 20,000 inhabitants, considering the initial and final years of the available estimates and half of the period.

The calculation of mortality rates included the mean number of deaths in the numerator and the mean population for each triennium in the denominator. The number of municipal deaths was extracted from a database for the 2000-2018 period, which shows estimates from the Institute for Health Metrics and Evaluation (IHME) produced within the Global Burden of Disease Study (GBD) at the request of the Rede GBD Brasil. The population estimates we used were of the Ministry of Health¹⁸.

The rates were age-standardized by the direct method, using the standard population of the Global Burden of Disease study¹⁹, and expressed based on 100,000 inhabitants. To measure the rates’ temporal evolution, we calculated the percentage change in mortality rates from the first triennium (T1) compared to the last triennium (T3). The descriptive analysis of mortality distribution was conducted by constructing choropleth maps evaluating mortality rates in the three triennia, considering quintiles as cutoff points.

The per capita GDP, the Brazilian deprivation index (BDI), and the percentage of young people with incomplete primary education in the municipalities of Minas Gerais were adopted to characterize the socioeconomic disparity between Minas Gerais intermediate regions. The description and analysis of the distribution of health events focusing on social inequalities in health aims to identify vulnerable groups by investigating indicators related to the social determinants of health. These measures encompass the so-called “inequality dimensions”, such as socioeconomic status indicators²⁰, like those measuring income and education, widely used in epidemiological studies investigating associations with mortality²¹, which guided the choice of the indicators and index mentioned above and detailed below.

The per capita GDP is an economic index that shows the value of final goods and services produced in a given geographic area and year, in current currency, and at market prices per individual. It is calculated by dividing GDP by the number of inhabitants in the region. It measures how much of the GDP would be shared by each individual in a country if everyone received equal shares. The index values ​​are in the public domain and were extracted from the Brazilian Institute of Geography and Statistics (IBGE) database²⁴.

The BDI was launched in 2020 and considers the combination of z-scores of three deprivation indicators on the census tract scale, based on data from the 2010 Demographic Census: percentage of households with income less than half the minimum wage, percentage of individuals aged 7 or over who are illiterate, and percentage of individuals with inadequate access to water, sanitation, waste collection, and no bathroom²⁵. The BDI generates a final score in which the lowest value represents the area with the least deprivation, while the highest value represents the area with the greatest deprivation^25,26.

The education data used in this study derived from the 2010 IBGE census and were retrieved from the Health Information System (TABNET) of the Unified Health System (DATASUS).²⁷

GBD study methodological aspects

The GBD organizes the underlying cause of death into a four-level hierarchy. Level 1 stratifies diseases into three principal groups: communicable, maternal, neonatal, and nutritional diseases, chronic noncommunicable diseases, and external causes. Levels 2, 3, and 4 detail the diseases in these three primary groups, disaggregating them into 21, 168, and 369 diseases, respectively.

In this article, the causes of death were analyzed per the most aggregated Level 1, except for maternal diseases, whose deaths were not calculated for the municipal base. Details on the general methodology of the GBD study can be found in other publications^19,28,29.

A summary of the data processing steps is described below. First, the data were assigned a cause under the GBD list of causes and corresponding International Statistical Classification of Diseases and Related Health Problems (ICD) code, and the aggregated data by gender or age were divided into detailed groups¹⁹. Next, a correction for misclassification of dementia, Parkinson’s disease, and atrial fibrillation was applied to the data¹⁹. The data were then submitted to a process whereby garbage codes (ICD codes that cannot be reliably attributed to a specific cause of death, such as codes for senility or back pain) and redistributed to actual causes from the GBD cause list.²⁸ Finally, the data were smoothed to account for stochastic variation over the year in a process called noise reduction¹⁹.

Data presentation and analysis

Municipal mortality rates were presented in choropleth maps. Conditionally formatted tables displayed aggregated municipal mortality rates in each intermediate region of Minas Gerais by group of causes for each triennium. The percentage changes in these rates between the triennia were also presented in these tables, in which the color formatting varied from dark red to dark blue. The darker the shade of red, the higher the mortality rates, and the smaller the reduction or the more significant the increase in the change percentages. Conversely, the darker the shade of blue, the lower the rates and the more significant the reduction in change percentages.

The median per capita GDP of each intermediate geographic region was calculated for the years of the third triennium. The median BDI refers to data from the 2010 Census. Education level was assessed for young people aged 18-24. The percentage of this population in each intermediate region with incomplete primary education was also considered, referring to data from the 2010 Census. A correlation analysis of these indicators was performed with the percentage rate variations between T3 and T1. The correlation analysis included the calculation of Pearson’s correlation coefficient (R), and a significance level of 5% was considered. Data were presented and analyzed through the R software³⁰ with the Rgeoda package.

Ethical aspects

The research complies with the provisions of Resolution Nº 466/12 of the National Health Council. All data used derive from secondary public domain databases, whose estimates do not allow the identification of individuals. The study was approved by the Research Ethics Committee of the Federal University of Minas Gerais (Opinion Nº 3.258.076). It is nested in the project “Inequalities in small geographic areas of indicators of chronic non-communicable diseases, violence, and their risk factors”.

Results

Minas Gerais had a mean number of deaths equivalent to 93,556 in the 2000-2002 triennium (T1), 114,380 in the 2009-2011 triennium (T2), and 130,095 in the 2016-2018 triennium (T3), with a predominance of NCDs (76.0%, 76.6% and 78.3% in each triennium, in this order) (Table 1).

Table 1
Distribution of deaths by groups of causes, Minas Gerais, in the triennia T1 (2000 to 2002), T2 (2009 to 2011), and T3 (2016-2018).

	T1					T2					T3
Cause	2000	2001	2002	Mean	%	2009	2010	2011	Mean	%	2016	2017	2018	Mean	%
CNND	11489	11568	11329	11462	12.3	11722	12035	12155	11971	10.5	13426	13653	13464	13514	10.4
NCD	69155	71647	72371	71058	76.0	84556	87992	90226	87591	76.6	100613	102434	102352	101800	78.3
EC	10609	11050	11449	11036	11.8	14301	14715	15437	14818	13.0	15605	14973	13764	14781	11.4
Total	91253	94265	95149	93556	100.0	110579	114742	117818	114380	100.0	129644	131060	129580	130095	100.0

• CNND = communicable, neonatal, and nutritional diseases; NCD = chronic noncommunicable diseases; EC = external causes.
• Source: Authors.

There was a significant decrease in mortality rates due to communicable, neonatal, and nutritional diseases from T1 to T3 (-29%) throughout the state, down from 83 to 58.9 per 100,000 inhabitants. In T3, the highest rates were in the intermediate regions of Montes Claros, Patos de Minas, Uberaba, and Teófilo Otoni (Figure 2A and Figure 3). Noncommunicable diseases (NCD) also showed a rate reduction (-22.4%), from 553.6 to 429.9 per 100,000 inhabitants. In T3, the highest rates were in municipalities in the Barbacena and Varginha regions (Figure 2B and Figure 3). In turn, external causes increased from 62.2 to 64.4 per 100,000 inhabitants (3.5%) between T1 and T3, with a reduction of 9.2% in the T2 and T3 stretch (Figure 2C and Figure 3).

Figure 2
Age-standardized mortality rates due to communicable, neonatal, and nutritional diseases (A), chronic non-communicable diseases (B), and injuries (external causes) (C) per 100,000 inhabitants, in T1, T2, and T3. Municipalities and intermediate geographic regions of Minas Gerais, GBD.

Figure 3
Heat map of the age-standardized mortality rates by causes per 100,000 inhabitants, and of the percent changes in the rates across T1, T2, and T3. Minas Gerais and intermediate geographic regions of the state, GBD.

Among Minas Gerais intermediate regions, the Uberaba region had the highest mortality rates due to communicable, neonatal, and nutritional diseases, and the Montes Claros region had the lowest rate decline due to these causes between T1 and T3 (Figure 3). Regarding chronic non-communicable diseases, the Teófilo Otoni intermediate region stood out for its growth in mortality rates (4.1%). In contrast, the Belo Horizonte, Pouso Alegre, Ipatinga, and Varginha regions had the most significant decline, with more than 25% in the period. Mortality from external causes increased in most regions between T1 and T3, mainly in Teófilo Otoni (41.4%), Montes Claros (22.3%), and Juiz de Fora (19.4%). On the other hand, the most considerable reductions were in the intermediate regions of Belo Horizonte (-9.2%), Pouso Alegre (-6.3%), and Varginha (-5.7%) (Figure 3).

Among external causes (Figure 4), mortality rates from transport accidents were the highest. At the same time, there was an increase in mortality rates from suicide and interpersonal violence (homicide) between T1 and T3.

Figure 4
Age-standardized mortality rates due to self-harm (suicide) and interpersonal violence (homicide) (A), unintentional injuries (B), and transport injuries (C) by 100,000 inhabitants in T1, T2, and T3. Municipalities and intermediate geographic regions of Minas Gerais, GBD.

Figure 5 shows the division of the state of Minas Gerais into intermediate geographic regions by median per capita GDP and BDI values ​​and the percentage of young people with incomplete primary education. Considering the socioeconomic indicators, the regions with the highest median per capita GDP were, in decreasing order, Uberaba, Uberlândia, Patos de Minas, and Belo Horizonte. The least economically prosperous regions are Teófilo Otoni, Montes Claros, and Governador Valadares (Figure 5A). These intermediate regions were also classified as areas of more significant material deprivation, according to median BDI values, and, together with the Ipatinga region, had the highest percentages of young people with incomplete primary education (Figures 5B and 5C). Although it had a high median per capita GDP, the Patos de Minas region was among the worst categories of median BDI and education indicators. Table 2 details the values ​​of the socioeconomic indicators for each Intermediate Geographic Region.

Table 2
Gross domestic product (GDP) per capita. median Brazilian deprivation index (BDI). and percentage of young people with incomplete primary education in each intermediate geographic region of Minas Gerais.

Intermediate region	GDP per capita (median*)	BDI (median*)	Incomplete education** (%)
Barbacena	19.90	-0.46	17.10
Belo Horizonte	24.60	-0.74	14.20
Divinópolis	23.20	-0.86	19.40
Governador Valadares	10.80	0.10	25.80
Ipatinga	21.50	-0.23	18.30
Juiz de Fora	14.30	-0.45	20.90
Montes Claros	10.20	0.74	22.80
Pouso Alegre	21.00	-0.67	18.10
Patos de Minas	25.60	-0.68	18.80
Teófilo Otoni	9.90	0.56	29.70
Uberaba	43.80	-1.02	17.40
Uberlandia	36.40	-0.88	15.30
Varginha	21.30	-0.85	18.20

• * Median municipal values for the municipalities that comprise each intermediate region. GDP per capita: median value for the three years 2016-2018. BDI: median value for 2010. ** According to data from the 2010 Demographic Census.
• Source: Authors.

Figure 5
Intermediate geographic regions of Minas Gerais by median GDP per capita in T3 (A), median Brazilian deprivation index (BDI) values in 2010 (B), and percentages of young people with incomplete primary education in 2010 (C).

The correlation analysis of socioeconomic indicators with the percentage changes in mortality rates indicated negative correlations with per capita GDP, regardless of the cause of death group. The correlations were significant for communicable, neonatal, and nutritional diseases (-0.70; p = 0.0081) and external causes (-0.64; p = 0.019). Considering the BDI and the education indicator, the correlations were positive for the three groups of causes, statistically significant, and robust (Figure 6).

Figure 6
Correlation between median GDP per capita, median BDI, percentage of people with incomplete primary education, and percent change in age-standardized mortality rates by 100,000 inhabitants between T3 and T1. Intermediate regions of Minas Gerais.

Discussion

We underscore the predominance of mortality from NCDs, persistent neonatal and nutritional communicable diseases, and the increase in external causes, characterizing the triple burden of causes of illness and death in the state, with smaller reductions in mortality rates in areas of worse socioeconomic status.

NCDs accounted for the most significant burden of causes of death. There was a more pronounced reduction in mortality from these causes in the intermediate geographic regions belonging to the best and middle economic strata (intermediate regions Belo Horizonte, Ipatinga, and Pouso Alegre). The second largest mortality burden, the risk of death from communicable diseases, also decreased during the study period, although to a lesser extent in areas with low median per capita GDP (intermediate regions of Montes Claros and Teófilo Otoni). External causes increased during the period analyzed, and the most significant growth occurred in the three geographic regions with the lowest per capita GDP (Teófilo Otoni, Montes Claros, and Governador Valadares).

Minas Gerais shows a significant variation in per capita GDP, with lower rates concentrated in the intermediate geographic regions of the north and higher rates mainly found in the center and the Triângulo Mineiro region. In this study, the higher the per capita GDP, the more significant the decrease in mortality rates. One of the wealthiest regions of the state, the Belo Horizonte region, had the most pronounced reductions in mortality rates in the three groups of causes. Other notable reductions occurred in the regions of Ipatinga, Barbacena, Uberaba, and Uberlândia, which concentrate the main activities of the economic sectors of industry, services, and agriculture³¹. Regarding health, the largest concentration of specialized services, equipment, and human resources is found in municipalities in the center and south of the state³², with better performance in mortality indicators.

On the other hand, the North and Northeast of the state showed the lowest declines in mortality rates from NCDs and CNNDs, with the worst results recorded in the intermediate regions of Teófilo Otoni, Montes Claros, and Governador Valadares, which have the lowest per capita GDP in the state. These are socioeconomically less favored areas, with the most significant deficiencies and the lowest development indicators³². Mortality from NCDs is an indicator sensitive to variables related to living conditions and access to health services. The lower decline in mortality rates from NCDs in these territories can, therefore, be explained by the lower access of these populations to goods and services, such as basic sanitation, especially sewage disposal^17,31.

From this same perspective, one can also talk about regional disparities in the state of Minas Gerais related to the distribution of health service provision and care gaps. One of the indicators used to assess the promotion of health equity by measuring the level achieved by healthcare regionalization and, thus, the regional capacities for outpatient and hospital care of resident populations is the resolution rate³².

According to the Minas Gerais Integrated Development Plan (PMDI) 2019-2030, seven of the 13 health macro-regions in the state had a resolution rate above 80%, and six showed values from 56 to 67% in 2017³¹. The regions with the lowest resolution rate, such as Governador Valadares and Teófilo Otoni, showed worse performances here, which coincided in the study with the worst mortality results in all groups of causes. Notably, we underscore the marked growth in mortality rates from external causes in the state. The municipalities with the most significant increases are mainly concentrated in intermediate regions with the lowest per capita GDP, namely, Teófilo Otoni, Montes Claros, Juiz de Fora, and Governador Valadares.

The association between mortality and education was significant in both developed and developing countries, indicating that the higher the education level, the lower the mortality rate^33,34. Our study confirmed these findings, identifying a strong correlation between low education levels in young people and worse mortality indicators.

Another strong correlation with social vulnerability was the association between the BDI and mortality, indicating that the more significant the material deprivation, the smaller the decline in mortality rates. These results were also found in municipalities in Minas Gerais in the Paraopeba River basin. Those with greater vulnerability had a high correlation due to external causes³⁵.

Noteworthy is that these findings reinforce studies on the epidemiological transition, under which the health situation would illustrate a change in morbimortality patterns marked by replacing the burden of communicable diseases with the burden of chronic diseases and external causes^4,6,36-38. They also support the assertion that populations in less socioeconomically developed countries, such as Brazil, show a pattern of concomitant disease burdens and conditions resulting from the juxtaposed risks from different spheres of determination of the health-disease process³⁹. However, besides this, within the same society, the more affluent populations would show a pattern of morbimortality marked by chronic non-communicable diseases. In contrast, the poorest populations would more strongly display a multiple pattern derived from the triple burden of diseases⁴⁰. The most socially and economically vulnerable segments would be more exposed to the juxtaposed risks, which would result in an excess of morbimortality in these populations⁴⁰.

Finally, the findings also lead to reflecting on competing risks, defined as events whose occurrences exclude or modify the probability of another event of interest^41,42. This study shows that regions with low per capita GDP have high mortality rates from CNNDs and external causes but low NCD rates compared to richer regions of Minas Gerais, albeit with a smaller decline in NCDs over the years.

This last situation does not correspond to what is expected in the literature since high rates of NCDs are also expected in poorer regions and more vulnerable populations^43,44. It is noteworthy that deaths from external causes are more frequent in young adults.⁴⁵ In the case of deaths from NCDs, approximately one-third are premature (30-69 years of age)⁴⁶. Therefore, one hypothesis for lower mortality rates from NCDs in the state’s poorest regions may be the existence of a triple burden of disease and competing risks.

Knowing the causes of mortality and their epidemiology supports setting priorities in health policies. However, we should consider the progress in improving mortality estimates provided by methodological corrections and adjustments for underreporting and garbage causes in the GBD study at the municipal level. Furthermore, adopting a geographic/spatial analysis tool allowed visualizing the spatial distribution of mortality estimates in small areas and identifying health inequalities in the state of Minas Gerais.

This study has some limitations. One is adopting a state correction factor rather than a municipal correction factor for underreporting deaths. Furthermore, we should consider the small number of deaths at the municipal level, which generally leads to random fluctuations in mortality rates over time. This study aimed to mitigate this variability by analyzing the data by triennium.

The study shows the epidemiological transition and the different mortality patterns in the state of Minas Gerais. Studying the causes of death and the pattern of these events in territories, especially in small areas, is a fundamental tool for identifying health inequalities, planning and implementing actions, and assessing health. Promoting managers’, health professionals’, and the general population’s access to these studies means strengthening open science, social participation, and the management of the Unified Health System.

References

• ¹ Para E, Maria A, Vasconcelos N, Miranda M, Gomes F. Transição demográfica: a experiência brasileira. Epidemiol Serv Saude 2012; 21(4):539-548.
• ² Omran AR. The Epidemiologic Transition: A Theory of the Epidemiology of Population Change. Milbank Q 2005; 83(4):731.
• ³ Defo BK. Beyond the 'transition' frameworks: the cross-continuum of health, disease and mortality framework. Glob Health Action 2014; 7:24804.
• ⁴ Martins TCF, Silva JHCM, Máximo GC, Guimarães RM. Transição da morbimortalidade no Brasil: um desafio aos 30 anos de SUS. Cien Saude Colet 2021; 26(10):4483-4496.
• ⁵ Araújo JD. Polarização epidemiológica no Brasil. Epidemiol Serv Saude 2012; 21(4):533-538.
• ⁶ Omran AR. The epidemiologic transition theory revisited thirty years later. World Heal Stat Q 1998; 53(2,3,4):99-119.
• ⁷ GBD 2016 Brazil Collaborators. Burden of disease in Brazil, 1990-2016: a systematic subnational analysis for the Global Burden of Disease Study 2016. Lancet 2018; 392(10149):760-775.
• ⁸ Oliveira DC, Barreira AS, Trunk MT, Guzmán AF. Efecto de las desigualdades socioeconómicas en la mortalidad de la ciudad de Fortaleza, Ceará, Brasil durante el año 2007. Rev Esp Salud Publica 2010; 84(4):441-450.
• ⁹ Barreto MLL, Carmo EH. Padrões de adoecimento e de morte da população brasileira: os renovados desafios para o Sistema Único de Saúde. Cien Saude Colet 2007; 12(Supl.):1779-1790.
• ¹⁰ Soares Filho AM, França GVA, Malta DC. Tripla carga de doenças no Brasil, 1990-2021: mudanças, inflexões e o fator COVID-19. Rev Min Enferm 2022; 26:e-1475.
• ¹¹ Jiang J, Rao JS. Robust small area estimation: an overview. Annual Rev 2020; 7:337-60.
• ¹² Pales RC, Santos GR, Rodrigues SG. Minas Gerais, estado síntese do desenvolvimento regional brasileiro [Internet]. 2014. [acessado 2022 dez 9]. Disponível em: https://www.congressods.com.br/quarto/index.php/trabalhos-aceites/gt-06-indicadores-de/171-anais/gt-06/468-minas-gerais-estado-sintese-do-desenvolvimento-regional-brasileiro
  » https://www.congressods.com.br/quarto/index.php/trabalhos-aceites/gt-06-indicadores-de/171-anais/gt-06/468-minas-gerais-estado-sintese-do-desenvolvimento-regional-brasileiro
• ¹³ Instituto Brasileiro de Geografia e Estatística (IBGE). Divisão Regional do Brasil em Regiões Geográficas Imediatas e Regiões Geográficas Imeadiatas [Internet]. 2017. [acesado 2022 dez 9]. Disponível em: https://www.ibge.gov.br/apps/regioes_geograficas/
  » https://www.ibge.gov.br/apps/regioes_geograficas
• ¹⁴ Barros SFS. Da Zona da Mata/MG à Região Geográfica Intermediária de Juiz de Fora/MG: continuidades e descontinuidades nas propostas de regionalização do IBGE. Caminhos Geogr 2021; 22(80):15-33.
• ¹⁵ Instituto Brasileiro de Geografia e Estatística (IBGE). Divisão Regional do Brasil. O que é [Internet]. [acessado 2022 dez 9]. Disponível em: https://www.ibge.gov.br/geociencias/organizacao-do-territorio/divisao-regional/15778-divisoes-regionais-do-brasil.html?=&t=o-que-e
  » https://www.ibge.gov.br/geociencias/organizacao-do-territorio/divisao-regional/15778-divisoes-regionais-do-brasil.html?=&t=o-que-e
• ¹⁶ Malachias I, Antônio F, Leles G, Auxiliadora M, Pinto S, Luiz C, Alencar FB, Silva E, Almeida PD, Siqueira M. Plano Diretor de Regionalização da Saúde de Minas Gerais (PDR/MG) [Internet]. 2011. [acessado 2022 out 13]. Disponível em: http://www.saude.mg.gov.br
  » http://www.saude.mg.gov.br
• ¹⁷ Secretaria de Estado de Saúde de Minas Gerais. Plano estadual de educação permanente em saúde de Minas Gerais [Internet]. 2018. [acessado 2022 out 13]. Disponível em: https://www.conass.org.br/planos-estaduais-educacao-permanente/PEEPS-MG.pdf
  » https://www.conass.org.br/planos-estaduais-educacao-permanente/PEEPS-MG.pdf
• ¹⁸ Brasil. Ministério da Saúde (MS). DATASUS. População Residente - Estudo de Estimativas Populacionais por Município, Idade e Sexo 2000-2021 - Brasil [Internet]. [acessado 2022 dez 9]. Disponível em: http://tabnet.datasus.gov.br/cgi/deftohtm.exe?ibge/cnv/popsvsbr.def
  » http://tabnet.datasus.gov.br/cgi/deftohtm.exe?ibge/cnv/popsvsbr.def
• ¹⁹ Abbafati C, Abbas KM, Abbasi-Kangevari M, Abd-Allah F, Abdelalim A, Abdollahi M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020; 396(10258):1204-1222.
• ²⁰ World Health Organization (WHO). Handbook on health inequality monitoring: with a special focus on low- and middle-income countries. Geneva: WHO; 2013.
• ²¹ Katikireddi SV, Niedzwiedz CL, Dundas R, Kondo N, Leyland AH, Rostila M. Inequalities in all-cause and cause-specific mortality across the life course by wealth and income in Sweden: a register-based cohort study. Int J Epidemiol 2020; 49(3):917-925.
• ²² Rede Interagencial de Informações para a saúde (RIPSA). Indicadores básicos para a saúde no Brasil: conceitos e aplicações [Internet]. 2008. [acessado 2022 dez 9]. Disponível em: http://bvsms.saude.gov.br/bvs/publicacoes/indicadores_basicos_saude_brasil_2ed.pdf
  » http://bvsms.saude.gov.br/bvs/publicacoes/indicadores_basicos_saude_brasil_2ed.pdf
• ²³ Instituto de Pesquisa Econômica Aplicada (Ipea). Produto interno bruto (PIB) per capita [Internet]. 2021. [acessado 2022 dez 9]. Disponível em: http://www.ipeadata.gov.br/ExibeSerie.aspx?serid=38375
  » http://www.ipeadata.gov.br/ExibeSerie.aspx?serid=38375
• ²⁴ Instituto Brasileiro de Geografia e Estatística (IBGE). Produto Interno Bruto dos Municípios. Downloads [Internet]. [acessado 2022 dez 9]. Disponível em: https://www.ibge.gov.br/estatisticas/economicas/contas-nacionais/9088-produto-interno-bruto-dos-municipios.html?edicao=29720&t=downloads
  » https://www.ibge.gov.br/estatisticas/economicas/contas-nacionais/9088-produto-interno-bruto-dos-municipios.html?edicao=29720&t=downloads
• ²⁵ Allik M, Ramos D, Agranonik M, Pereira Júnior E, Ichihara MY, Barreto ML, Leyland AH, Dundas R. Developing a Small-Area Deprivation Measure for Brazil. Technical Report [Internet]. 2020. [cited 2022 dez 9]. Available from: https://cidacs.bahia.fiocruz.br/ibp/publicacao/
  » https://cidacs.bahia.fiocruz.br/ibp/publicacao
• ²⁶ Centro de Integração de Dados e Conhecimentos para a Saúde (CIDACS). O índice - IBP [Internet]. [acessado 2022 dez 9]. Disponível em: https://cidacs.bahia.fiocruz.br/ibp/indice/
  » https://cidacs.bahia.fiocruz.br/ibp/indice
• ²⁷ Brasil. Ministério da Saúde (MS). Educação - Censos 1991, 2000 e 2010 - DATASUS [Internet]. [acessado 2023 maio 14]. Disponível em: https://datasus.saude.gov.br/educacao-censos-1991-2000-e-2010
  » https://datasus.saude.gov.br/educacao-censos-1991-2000-e-2010
• ²⁸ Johnson SC, Cunningham M, Dippenaar IN, Sharara F, Wool EE, Agesa KM, Han C, Miller-Petrie MK, Wilson S, Fuller JE, Balassyano S, Bertolacci GJ, Davis Weaver N; GBD Cause of Death Collaborators; Lopez AD, Murray CJL, Naghavi M. Public health utility of cause of death data: applying empirical algorithms to improve data quality. BMC Med Inform Decis Mak 2021; 21(1):175.
• ²⁹ 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(Supl. 1):E210003.SUPL.1.
• ³⁰ Allaire JJ. RStudio: Integrated Development Environment for R [Internet]. [cited 2022 set 21]. 2012. Available from: http://www.rstudio.org/
  » http://www.rstudio.org
• ³¹ Minas Gerais. Plano Mineiro de Desenvolvimento Integrado (PMDI) 2019-2030 [Internet]. 2019. [acessado 2022 dez 9]. Disponível em: https://www.planejamento.mg.gov.br/pagina/planejamento-e-orcamento/plano-mineiro-de-desenvolvimento-integrado-pmdi/plano-mineiro-de
  » https://www.planejamento.mg.gov.br/pagina/planejamento-e-orcamento/plano-mineiro-de-desenvolvimento-integrado-pmdi/plano-mineiro-de
• ³² Secretaria de Estado de Saúde de Minas Gerais. Plano Diretor de Regionalização da Saúde de Minas Gerais (PDR/MG) [Internet]. 2011. [acessado 2022 dez 9]. Disponível em: https://www.saude.mg.gov.br/images/noticias_e_eventos/000_2016/Livro Plano Diretor de Regionalizao - PDR-SUS-MG.pdf
  » https://www.saude.mg.gov.br/images/noticias_e_eventos/000_2016/Livro Plano Diretor de Regionalizao - PDR-SUS-MG.pdf
• ³³ Ross CE, Masters RK, Hummer RA. Education and the Gender Gaps in Health and Mortality. Demography 2012; 49(4):1157-1183.
• ³⁴ Halpern-Manners A, Helgertz J, Warren JR, Roberts E. The Effects of Education on Mortality: Evidence from Linked U.S. Census and Administrative Mortality Data. Demography 2020; 57(4):1513.
• ³⁵ Malta DC, Reis GM, Veloso GA, Cardoso LSM, Hartz ZMA, Cunningham M, Naghavi M. Mortality patterns in municipalities of a mining region before the Brumadinho dam failure, state of Minas Gerais, Brazil. Rev Bras Epidemiol 2023; 26(Supl. 1):e230010.
• ³⁶ Marinho F, Passos VMA, França EB, Marinho F, Passos VMA, França EB. Novo século, novos desafios: mudança no perfil da carga de doença no Brasil de 1990 a 2010. Epidemiol Serv Saude 2016; 25(4):713-724.
• ³⁷ Borges GM. Health transition in Brazil: regional variations and divergence/convergence in mortality. Cad Saude Publica 2017; 33(8):e00080316.
• ³⁸ Peixoto SV. A tripla carga de agravos e os desafios para o Sistema Único de Saúde. Cien Saude Colet 2020; 25(8):2912.
• ³⁹ Duarte EC, Schneider MC, Sousa RP, Ramalho WM, Sardinha LMV, Silva Júnior JB, Castillo-Salgado C. Epidemiologia das desigualdades em saúde no Brasil: um estudo exploratório [Internet]. 2002. [acessado 2022 dez 9]. Disponível em: https://bvsms.saude.gov.br/bvs/publicacoes/funasa/epi_desigualdades.pdf
  » https://bvsms.saude.gov.br/bvs/publicacoes/funasa/epi_desigualdades.pdf
• ⁴⁰ Paes-Sousa R. Diferenciais intra-urbanos de mortalidade em Belo Horizonte, Minas Gerais, Brasil, 1994: revisitando o debate sobre transições demográfica e epidemiológica. Cad Saude Publica 2002; 18(5):1411-1421.
• ⁴¹ Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999; 18:695-706.
• ⁴² Carvalho MS, Pacheco AG. Eventos competitivos. In: Carvalho MS, Andreozzi VL, Codeço CT, Campos DP, Barbosa MTS, Shimakura SE. Análise de sobrevivência teoria e aplicações em saúde. Rio de Janeiro: Fiocruz; 2011. p. 432.
• ⁴³ Marmot M, Bell R. Social determinants and non-communicable diseases: time for integrated action. BMJ 2019; 364:1251.
• ⁴⁴ Di Cesare M, Khang YH, Asaria P, Blakely T, Cowan MJ, Farzadfar F, Guerrero R, Ikeda N, Kyobutungi C, Msyamboza KP, Oum S, Lynch JW, Marmot MG, Ezzati M; Lancet NCD Action Group. Inequalities in non-communicable diseases and effective responses. Lancet 2013; 381(9866):585-597.
• ⁴⁵ Malta DC, Minayo MCS, Cardoso LSM, Veloso GA, Teixeira RA, Pinto IV, Naghavi M. Mortalidade de adolescentes e adultos jovens brasileiros entre 1990 e 2019: uma análise do estudo Carga Global de Doença. Cien Saude Colet 2021; 26(9):4069-4086.
• ⁴⁶ Malta DC, Duncan BB, Schmidt MI, Teixeira R, Ribeiro ALP, Felisbino-Mendes MS, Machado IE, Velasquez-Melendez G, Brant LCC, Silva DAS, Passos VMA, Nascimento BR, Cousin E, Glenn S, Naghavi M. Trends in mortality due to non-communicable diseases in the Brazilian adult population: national and subnational estimates and projections for 2030. Popul Health Metr 2020; 18(Suppl. 1):16.

• Funding
  This study was funded by the following projects: FAPEMIG Universal Call Nº 01/2021 “Inequalities in mortality indicators due to chronic non-communicable diseases and COVID-19 in Brazil and Minas Gerais”; and the National Health Fund, through the Health Surveillance Secretariat, “Studies on the Global Burden of Diseases, Surveys, and Artificial Intelligence”.

• Chief editors:
  Maria Cecília de Souza Minayo, Romeu Gomes, Antônio Augusto Moura da Silva

Publication Dates

• Publication in this collection
  10 Feb 2025
• Date of issue
  Feb 2025

History

• Received
  11 Feb 2023
• Accepted
  12 Dec 2023
• Published
  14 Dec 2023