ABSTRACT

OBJECTIVE

To analyze the time trend of monthly mortality rates from chronic respiratory diseases in Brazil from 1996 to 2017, with forecasts for 2022, besides analyzing the possibility of achieving the goal of the Plano de Ações Estratégicas para o Enfrentamento das Doenças Crônicas Não Transmissíveis no Brasil (Strategic Action Plan to Tackle Chronic Noncommunicable Diseases in Brazil) from 2011 to 2022.

METHODS

This is an ecological study that uses data from Sistema de Informações sobre Mortalidade (SIM – Mortality Information System), Sistema de Informações Demográficas e Socioeconômicas (Demographic and Socioeconomic Information System) and Pesquisa Nacional por Amostra de Domicílios Contínua (PNAD Contínua – Continuous National Household Sample Survey). We established the age range between 30 and 69 years old and the evolution of the rates over time was made by autoregressive integrated moving average models in R statistical tool.

RESULTS

Premature mortality rates from chronic respiratory diseases are decreasing in Brazil as a whole, mostly in state capitals. There is also a trend to reach the Ministry of Health’s goal in most of the country. For capitals that tend not to reach the goal, there is an association between mortality and social indicators, healthcare network and frequency of smoking.

CONCLUSION

This study intends to improve planning of the public health system for the control of chronic respiratory diseases.

Respiratory Tract Diseases, mortality; Chronic Disease Indicators; Health Programs and Plans, trends; Ecological Studies

INTRODUCTION

The worldwide premature mortality rate from chronic noncommunicable diseases (NCDs) dropped by 18% at the beginning of this century¹1. World Health Organization. World Health Statistics 2020. Geneva (CH): WHO; 2020.. However, this progress was not sustained since the decline of mortality from NCDs slowed down¹1. World Health Organization. World Health Statistics 2020. Geneva (CH): WHO; 2020.. These diseases, mainly cardiovascular diseases, cancers, chronic respiratory diseases and diabetes, correspond to the largest number of deaths in the world (63% annually) and, once demographic changes compensate for the drop in mortality rates, NCDs are taking an even higher proportion of total deaths¹1. World Health Organization. World Health Statistics 2020. Geneva (CH): WHO; 2020.,²2. World Health Organization. Global Action Plan for the Prevention and Control of NCDs 2013-2020. Geneva (CH): WHO; 2013 [cited 2020 Oct 14]. Available from: https://www.who.int/publications/i/item/9789241506236
https://www.who.int/publications/i/item/... .

In the Americas, the percentage reduction rate of premature mortality from NCDs also decreased³3. Martinez R, Soliz P, Caixeta R, Ordunez P. Reflection on modern methods: years of life lost due to premature mortality: a versatile and comprehensive measure for monitoring non-communicable disease mortality. Int J Epidemiol. 2019;48(4):1367-76. https://doi.org/10.1093/ije/dyy254
https://doi.org/10.1093/ije/dyy254... . Despite leading as the cause of morbidity and mortality in Brazil, there was a 20% decline in the mortality rate from NCD between 1996 and 2007⁴4. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Análise de Situação de Saúde. Plano de ações estratégicas para o enfrentamento das doenças crônicas não transmissíveis (DCNT) no Brasil 2011-2022. Brasília, DF; 2011 [cited 2020 Oct 14]. (Série B. Textos Básicos de Saúde). Available from: https://bvsms.saude.gov.br/bvs/publicacoes/plano_acoes_enfrent_dcnt_2011.pdf
https://bvsms.saude.gov.br/bvs/publicaco... , and a reduction in the premature mortality rate (30 to 69 years old) from NCDs, from 2000 to 2013 (2.5% per year), with chronic respiratory diseases (CRD) leading the percentage variation (4.1%)⁵5. Malta DC, Andrade SSCA, Oliveira TP, Moura L, Prado RR, Souza MF. Probabilidade de morte prematura por doenças crônicas não transmissíveis, Brasil e regiões, projeções para 2025. Rev Bras Epidemiol. 2019;22:E190030. https://doi.org/10.1590/1980-549720190030
https://doi.org/10.1590/1980-54972019003... .

The CRD alone was the third cause of death (7%) in the world in 2017, after cardiovascular and neoplastic diseases, especially chronic obstructive pulmonary disease (COPD) and asthma⁶6. GBD Chronic Respiratory Disease Collaborators. Prevalence and attributable health burden of chronic respiratory diseases, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Respir Med. 2020;8(6):585-96. https://doi.org/10.1016/S2213-2600(20)30105-3
https://doi.org/10.1016/S2213-2600(20)30... . Between 1990 and 2017, the total number of deaths caused by CRD increased worldwide, whereas the age-standardized mortality rate decreased⁶6. GBD Chronic Respiratory Disease Collaborators. Prevalence and attributable health burden of chronic respiratory diseases, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Respir Med. 2020;8(6):585-96. https://doi.org/10.1016/S2213-2600(20)30105-3
https://doi.org/10.1016/S2213-2600(20)30... ,⁷7. Li X, Cao X, Guo M, Xie M, Liu X. Trends and risk factors of mortality and disability adjusted life years for chronic respiratory diseases from 1990 to 2017: systematic analysis for the Global Burden of Disease Study 2017. BMJ. 2020;368:m234. https://doi.org/ 10.1136/bmj.m234
https://doi.org/... . The CRD scenario in Brazil followed the global trend, with COPD being the fourth cause of death in 2019, with an absolute increase in the number of deaths from CRD and a reduction in age-adjusted mortality rates⁵5. Malta DC, Andrade SSCA, Oliveira TP, Moura L, Prado RR, Souza MF. Probabilidade de morte prematura por doenças crônicas não transmissíveis, Brasil e regiões, projeções para 2025. Rev Bras Epidemiol. 2019;22:E190030. https://doi.org/10.1590/1980-549720190030
https://doi.org/10.1590/1980-54972019003... ,⁸8. Leal LF, Cousin E, Bidinotto AB, Sganzerla D, Borges RB, Malta DC, et al. Epidemiology and burden of chronic respiratory diseases in Brazil from 1990 to 2017: analysis for the Global Burden of Disease 2017 Study. Rev Bras Epidemiol. 2020;23:e200031. https://doi.org/10.1590/1980-549720200031
https://doi.org/10.1590/1980-54972020003... .

The NCDs mainly affect low-income and low-education populations, as they are more exposed to risk factors and have reduced access to health services⁴4. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Análise de Situação de Saúde. Plano de ações estratégicas para o enfrentamento das doenças crônicas não transmissíveis (DCNT) no Brasil 2011-2022. Brasília, DF; 2011 [cited 2020 Oct 14]. (Série B. Textos Básicos de Saúde). Available from: https://bvsms.saude.gov.br/bvs/publicacoes/plano_acoes_enfrent_dcnt_2011.pdf
https://bvsms.saude.gov.br/bvs/publicaco... ,¹¹11. Malta DC, Bernal, RTI, Lima MG, Araújo SSC, Silva MMA, Freitas MIF, et al. Doenças crônicas não transmissíveis e a utilização de serviços de saúde: análise da Pesquisa Nacional de Saúde no Brasil. Rev Saude Publica. 2017;51 Supl 1:4s. https://doi.org/10.1590/S1518-8787.2017051000090
https://doi.org/10.1590/S1518-8787.20170... . In the case of CRD, it is not different. Smoking is the main risk factor associated with years of life adjusted for disability⁸8. Leal LF, Cousin E, Bidinotto AB, Sganzerla D, Borges RB, Malta DC, et al. Epidemiology and burden of chronic respiratory diseases in Brazil from 1990 to 2017: analysis for the Global Burden of Disease 2017 Study. Rev Bras Epidemiol. 2020;23:e200031. https://doi.org/10.1590/1980-549720200031
https://doi.org/10.1590/1980-54972020003... . In this context, studies show that less socioeconomically favored groups are more prone to use tobacco frequently¹²12. Allen L, Williams J, Townsend N, Mikkelsen B, Roberts N, Foster C, et al. Socioeconomic status and non-communicable disease behavioural risk factors in low-income and lower-middle-income countries: a systematic review. Lancet Glob Health. 2017;5(3):e277-89. https://doi.org/10.1016/S2214-109X(17)30058-X
https://doi.org/10.1016/S2214-109X(17)30... . Moreover, data from Vigilância de Fatores de Risco e Proteção para Doenças Crônicas por Inquérito Telefônico (Vigitel – Surveillance System for Risk and Protective Factors for Chronic Diseases by Telephone Survey)¹³13. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Vigilância de Doenças e Agravos não Transmissíveis e Promoção da Saúde. Vigitel Brasil 2017: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico. Brasília, DF; 2018 [cited 2020 Oct 14]. Available from: https://bvsms.saude.gov.br/bvs/publicacoes/vigitel_brasil_2017_vigilancia_fatores_riscos.pdf
https://bvsms.saude.gov.br/bvs/publicaco... indicate lower frequency of smoking with higher schooling. Furthermore, the poorest are less diagnosed with CRD and those who use fewer medications¹⁴14. Leal LF, Bertoldi AD, Menezes AMB, Borges RB, Mengue SS, Gazzana MB, et al. Indicação, acesso e utilização de medicamentos para doenças respiratórias crônicas no Brasil: resultados da Pesquisa Nacional sobre Acesso, Utilização e Promoção do Uso Racional de Medicamentos no Brasil (PNAUM), 2014. Cad Saude Publica. 2018;34(10):e00202817. https://doi.org/10.1590/0102-311X00208217
https://doi.org/10.1590/0102-311X0020821... .

Facing the impact of NCDs, global and national policies are committed to addressing these diseases. The first High-Level Meeting on the Prevention and Control of NCDs in 2011; target 3.4 from Sustainable Development Goals (SDG), of reducing by one-third premature mortality from these diseases by 2030; and the Global Action Plan for the Prevention and Control of Noncommunicable Diseases 2013–2020 were highlights on the world agenda²2. World Health Organization. Global Action Plan for the Prevention and Control of NCDs 2013-2020. Geneva (CH): WHO; 2013 [cited 2020 Oct 14]. Available from: https://www.who.int/publications/i/item/9789241506236
https://www.who.int/publications/i/item/... ,¹⁵15. NCD Countdown 2030 collaborators. NCD Countdown 2030: worldwide trends in non-communicable disease mortality and progress towards Sustainable Development Goal target 3.4. Lancet. 2018;392(10152):1072-88. https://doi.org/10.1016/S0140-6736(18)31992-5
https://doi.org/10.1016/S0140-6736(18)31... .

In Brazil, the implementation of the Surveillance System for NCDs was relevant and allowed structuring actions to monitor risk factors, morbidity and mortality of NCDs⁴4. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Análise de Situação de Saúde. Plano de ações estratégicas para o enfrentamento das doenças crônicas não transmissíveis (DCNT) no Brasil 2011-2022. Brasília, DF; 2011 [cited 2020 Oct 14]. (Série B. Textos Básicos de Saúde). Available from: https://bvsms.saude.gov.br/bvs/publicacoes/plano_acoes_enfrent_dcnt_2011.pdf
https://bvsms.saude.gov.br/bvs/publicaco... ,¹⁶16. Malta DC, Silva MMA, Moura L, Morais Neto OL. A implantação do Sistema de Vigilância de Doenças Crônicas Não Transmissíveis no Brasil, 2003 a 2015: alcances e desafios. Rev Bras Epidemiol. 2017;20(4):661-75. https://doi.org/10.1590/1980-5497201700040009
https://doi.org/10.1590/1980-54972017000... . The Ministry of Health (MS) also built the Plano de Ações Estratégicas para o Enfrentamento das Doenças Crônicas Não Transmissíveis (Strategic Action Plan to Tackle NCDs) in Brazil, 2011–2022, having as one of the goals reducing the premature mortality rate from NCDs by 2% per year⁴4. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Análise de Situação de Saúde. Plano de ações estratégicas para o enfrentamento das doenças crônicas não transmissíveis (DCNT) no Brasil 2011-2022. Brasília, DF; 2011 [cited 2020 Oct 14]. (Série B. Textos Básicos de Saúde). Available from: https://bvsms.saude.gov.br/bvs/publicacoes/plano_acoes_enfrent_dcnt_2011.pdf
https://bvsms.saude.gov.br/bvs/publicaco... .

The country has been considered an example in the fight against smoking, recognized by world organizations, with emphasis on the increase in the tax on cigarettes and the increase in warnings on packaging, by Law No. 12,546 of 2011¹⁶16. Malta DC, Silva MMA, Moura L, Morais Neto OL. A implantação do Sistema de Vigilância de Doenças Crônicas Não Transmissíveis no Brasil, 2003 a 2015: alcances e desafios. Rev Bras Epidemiol. 2017;20(4):661-75. https://doi.org/10.1590/1980-5497201700040009
https://doi.org/10.1590/1980-54972017000... .

However, globally, CRD received fewer resources for research and less public attention than other diseases, such as cardiovascular and neoplastic ones¹⁷17. Labaki WW, Han MK. Chronic respiratory diseases: a global view. Lancet. 2020;8(6):531-3. https://doi.org/10.1016/S2213-2600(20)30157-0
https://doi.org/10.1016/S2213-2600(20)30... . The present study intends to analyze the time trend of mortality from CRD in Brazil, forecasting rates for 2022, besides analyzing, for each state capital, the possibility of achieving the goal of the Strategic Action Plan to Tackle NCDs in Brazil, 2011–2022.

METHODS

This is a time-series ecological study of monthly mortality rates from CRD (ICD-10: J30–J98) between 1996 and 2017, with forecast for 2022, in Brazilian capitals and in Brazil as a whole. Data on deaths were obtained from the Brazilian Unified Health System Computer Department (Datasus) Mortality Information System – Ministry of Health (SIM/MS), and population data from 1996 to 2015 were acquired from the Demographic and Socioeconomic Information System, also from Datasus/MS. The estimated total population for 2016 and 2017 was made based on the continuous Pesquisa Nacional por Amostra de Domicílios (PNAD – National Household Sample Survey), using weighed samples to calculate the total population.

The monthly general crude mortality rates from CRD and monthly age-standardized rates were calculated using the method of direct standardization, which enabled the comparison of the results of the different populations by age group. The method selects a standard population and applies the mortality rate of all units to that same population. The Brazilian population in 2010 (census year) was used; mortality rates were calculated and represented by 100,000 inhabitants.

The chosen age range was 30 to 69 years and allowed us to compare the time trend of mortality from CRD in Brazil and the goal proposed by the Strategic Action Plan to Tackle NCDs.

Descriptive data analysis was used to explore the behavior of monthly mortality rates from CRD, obtaining values of minimum, first quartile, median, mean, third quartile, maximum and standard deviation. Based on the mean and standard deviation, one can calculate the coefficient of variation of the rates, which is important to express the dispersion of the data and compare values without the influence of the differences in magnitude of the variable.

The evolution of mortality rates in relation to time (months/years) was made by autoregressive integrated moving average (ARIMA) models. The time-series models of this type allow the dependent variable, mortality from CRD, to be represented by Y_t and explained by lagged values (Y_t-n), from Y_t itself, different from the regression models, in which Y_t is explained by the regressors μ, X1, X2, X3, ... Xn, without influence of time¹⁸18. Gujarati DN, Porter DC. Basic econometrics. 5. ed. New York: The McGraw-Hill Education; 2008.. Thus, μ is the constant of the model and X1, X2, X3, ... Xn are explanatory variables. To complement ARIMA models, the model with autoregressive distributed lag (dynamic regression) was also used. The variables used were the standardized mortality rate, the time and this same rate delayed in time (t-1).

Assuming that the mortality rate follows a normal distribution (Gaussian) and that it is a realization of a stationary stochastic process, we applied the dynamic regression for each capital, as well as for Brazil. The lag order for each capital and Brazil (t-1) was determined by the autocorrelation function. Being the model in general given by:

Where Y_t is the value of the dependent variable at instant t; γ is the parameter attached to the autoregressive operator; θ is the parameter attached to the moving average operator; and Θ and Γ are seasonal factors. For ARIMA models, the first difference was denoted as ∆zt = zt – zt-1 being Y_t = ∆z_t for the purpose of writing the estimated models. The components trend, seasonality, cycle and error were decomposed by the X12-ARIMA methodology and the trend evaluated as increasing or decreasing. The Dickey-Fuller test increased was performed for all series, which presented unit root, and, for this reason, the first difference was used. Autocorrelation functions also presented a slow decay.

The model was estimated by the function of maximum likelihood. For the modeling process, the first three years of Rio de Janeiro were suppressed.

Regarding 19 capitals (Belém, Boa Vista, Macapá, Manaus, Palmas, Porto Velho, Rio Branco, João Pessoa, Maceió, Salvador, São Luís, Teresina, Brasília, Campo Grande, Cuiabá, Goiânia, Vitória, Curitiba and Florianópolis), the outliers had to be replaced with the mean mortality rates of their respective years to achieve better adequacy of the models.

The evaluation of the overall performance of the model was based on the prediction of monthly mortality rates from CRD for 2017 and the comparison with the actual values of that year, which made it possible to calculate the percentage of the mean absolute percentage error (MAPE). In this procedure, the rates for 2017 were deleted from the model and estimated in the remainder. The maximum MAPE value for the model to be considered valid for forecasting was defined as 50%. After the validating the model, forecast until December 2022 was made.

The predicted percentage change for the mortality rate was calculated based on the value of the 2011 standardized annual rate, the year of preparation of the Ministry of Health plan, and the expected mortality rate for 2022. The result of each Brazilian capital was compared, as well as the one of all Brazil, with the document proposal (a 19.93% drop in the premature mortality rate until 2022, value obtained from the calculation of the accumulated rate of 2% per year), evaluating which units show a trend or not to reach the target.

Data were analyzed in R statistical tool¹⁹19. R Core Team. R: a language and environment for statistical computing. Vienna (AT). R Foundation for Statistical Computing; 2020 [cited 2020 Oct 14]. Available from: https://www.R-project.org/
https://www.R-project.org/... , version 3.6.3, using the Forecast²⁰20. Hyndman R, Athanasopoulos G, Bergmeir C, Caceres G, Chhay L, O’Hara-Wild M, et al. Forecasting functions for time series and linear models: R package version 8.12. 2020 [cited 2020 Oct 14]. Available from: http://pkg.robjhyndman.com/forecast
http://pkg.robjhyndman.com/forecast... and Imtest²¹21. Zeileis A, Hothorn T. Diagnostic checking in regression relationships. R News. 2002;2(3):7-10 [cited 2020 Oct 14]. Available from: https://www.r-project.org/doc/Rnews/Rnews_2002-3.pdf
https://www.r-project.org/doc/Rnews/Rnew... packages.

RESULTS

From the analysis of minimum and maximum values from mortality rates, presented in Table 1, we observed the lowest monthly rates in Boa Vista, Palmas, Porto Velho, Rio Branco, Aracaju, João Pessoa, São Luís, Teresina, Campo Grande, Vitória and Florianópolis; the highest value was found in Boa Vista (15.04%).

The comparison between the mean and median data in Table 1, for the mortality rates of each capital, showed that Palmas has the highest number of outliers.

We observed a significant variation in the monthly mortality rates from CRD in all units studied, which can be noted from their mean and standard deviation values presented in Table 1. All capitals, and Brazil as a whole, presented a coefficient of variation higher than 25%, with the lowest value calculated for the country (25.14%). In turn, Boa Vista and Palmas showed coefficients of variation higher than 100% (respectively, 124.41% and 149.49%).

Regarding the accuracy of the forecasts, we observed in the MAPE, in Table 2, that the models of Brazil (2.94%), São Paulo (12.93%) and Rio de Janeiro (13.81%) have the lowest average percentage error in relation to the others. The models of Boa Vista, Macapá, Manaus, Aracaju, João Pessoa, Campo Grande and Vitória presented high error values, showing a low-accurate forecasting capacity. The error values of the capitals Porto Velho, Rio Branco, Natal, Teresina and Cuiabá show that their models are not suitable for forecasting. Cuiabá, besides the medium-high percentage error, showed a model with behavior in levels, making it difficult to analyze the trend.

Based on Table 2 and Figures 1 and 2, we note that premature mortality rates are decreasing in Brazil and in 15 of the 22 capitals that had their models validated. Belém, Palmas, Recife, São Luís, Campo Grande, Vitória and Curitiba, on the other hand, show these rates in an increasing trend. For Rio de Janeiro, mortality from CRD data between 1996 and 1998 were much lower than the others, indicating a probable underreporting during this period and, therefore, they were disregarded.

According to Table 3, we also verified that Brazil as a whole and 16 of the 22 capitals tend to reach the goal of the Strategic Action Plan to Tackle NCDs, as they present a higher-than-expected predicted percentage variation for the premature mortality rates from CRD (19.92%). For Boa Vista, Recife, São Luís, Campo Grande and Vitória, the results show a trend not to reach the goal. The capital Boa Vista showed the lowest rate in 2011 (10.04), whereas the highest was seen in Rio Branco (36.76).

DISCUSSION

The downward trend in premature mortality rates from CRD in most of Brazil coincides with results of global research, which show the largest reductions in the burden of CRD in Latin America, and other Brazilian studies that indicate a decrease in those rates in the five regions of the country⁵5. Malta DC, Andrade SSCA, Oliveira TP, Moura L, Prado RR, Souza MF. Probabilidade de morte prematura por doenças crônicas não transmissíveis, Brasil e regiões, projeções para 2025. Rev Bras Epidemiol. 2019;22:E190030. https://doi.org/10.1590/1980-549720190030
https://doi.org/10.1590/1980-54972019003... ,⁶6. GBD Chronic Respiratory Disease Collaborators. Prevalence and attributable health burden of chronic respiratory diseases, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Respir Med. 2020;8(6):585-96. https://doi.org/10.1016/S2213-2600(20)30105-3
https://doi.org/10.1016/S2213-2600(20)30... ,⁸8. Leal LF, Cousin E, Bidinotto AB, Sganzerla D, Borges RB, Malta DC, et al. Epidemiology and burden of chronic respiratory diseases in Brazil from 1990 to 2017: analysis for the Global Burden of Disease 2017 Study. Rev Bras Epidemiol. 2020;23:e200031. https://doi.org/10.1590/1980-549720200031
https://doi.org/10.1590/1980-54972020003... . This decline has been attributed, among other factors, to advances in the fight against smoking⁵5. Malta DC, Andrade SSCA, Oliveira TP, Moura L, Prado RR, Souza MF. Probabilidade de morte prematura por doenças crônicas não transmissíveis, Brasil e regiões, projeções para 2025. Rev Bras Epidemiol. 2019;22:E190030. https://doi.org/10.1590/1980-549720190030
https://doi.org/10.1590/1980-54972019003... . Actions in this sense are related to Eixo II – Promoção da Saúde – do Plano de Ações Estratégicas para o enfrentamento das DCNT (Axis II – Health Promotion – of the Strategic Action Plan to Tackle NCDs) in Brazil. The increase in the access and use of health services also seems to corroborate, according to a study that assessed data from the 2013 Pesquisa Nacional de Saúde (PNS – National Health Survey)²²22. Stopa SR, Malta DC, Monteiro CN, Szwarcwald CL, Goldbaum M, Cesar CLG. Acesso e uso de serviços de saúde pela população brasileira, Pesquisa Nacional de Saúde 2013. Rev Saude Publica. 2017;51 Supl 1:3s. https://doi.org/10.1590/S1518-8787.2017051000074
https://doi.org/10.1590/S1518-8787.20170... .

According to the Síntese de Indicadores Sociais (SIS – Summary of Social Indicators), from the Instituto Brasileiro de Geografia e Estatística (IBGE – Brazilian Institute of Geography and Statistics), there was a drop in the proportion of people living below the poverty line between 2012 and 2014, an important social determinant in the health-disease process, but it grew again since 2015²³23. Instituto Brasileiro de Geografia e Estatística. Síntese de indicadores sociais: uma análise das condições de vida da população brasileira. Rio de Janeiro: IBGE; 2019 [cited 2020 Oct 14]. (Estudos e pesquisas. Informação demográfica e socioeconômica; nº 40). Available from: https://biblioteca.ibge.gov.br/visualizacao/livros/liv101678.pdf
https://biblioteca.ibge.gov.br/visualiza... . Another significant determinant is schooling, which presented an increase in the level of education over the last decades that may be related to the reduction of exposure to CRD risk factors, although this higher schooling occurred unequally among different socioeconomic classes²³23. Instituto Brasileiro de Geografia e Estatística. Síntese de indicadores sociais: uma análise das condições de vida da população brasileira. Rio de Janeiro: IBGE; 2019 [cited 2020 Oct 14]. (Estudos e pesquisas. Informação demográfica e socioeconômica; nº 40). Available from: https://biblioteca.ibge.gov.br/visualizacao/livros/liv101678.pdf
https://biblioteca.ibge.gov.br/visualiza... . In this context, there seems to be a relation between asthma mortality and factors such as schooling and income distribution, explaining the low mortality from this disease in high-income countries²⁴24. GBD 2015 Chronic Respiratory Disease Collaborators. Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Respir Med. 2017;5(9):691-706. https://doi.org/10.1016/S2213-2600(17)30293-X
https://doi.org/10.1016/S2213-2600(17)30... . For COPD, this association is less monotonic and results from increased air pollution and smoking in countries in demographic and epidemiological transition, a causality that deserves specific research in a future study²⁴24. GBD 2015 Chronic Respiratory Disease Collaborators. Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Respir Med. 2017;5(9):691-706. https://doi.org/10.1016/S2213-2600(17)30293-X
https://doi.org/10.1016/S2213-2600(17)30... .

Knowing that the highest mortality rates from CRD were found in South Asia, a study that analyzed the trend of prevalence, lethality and risk factors for CRD in India allows comparing the evolution of tackling these chronic diseases in another developing country to the situation in Brazil⁶6. GBD Chronic Respiratory Disease Collaborators. Prevalence and attributable health burden of chronic respiratory diseases, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Respir Med. 2020;8(6):585-96. https://doi.org/10.1016/S2213-2600(20)30105-3
https://doi.org/10.1016/S2213-2600(20)30... ,²⁵25. India State-Level Disease Burden Initiative CRD Collaborators. The burden of chronic respiratory diseases and their heterogeneity across the states of India: the Global Burden of Disease Study 1990–2016. Lancet Glob Health. 2018;6(12):e1363-74. https://doi.org/10.1016/S2214-109X(18)30409-1
https://doi.org/10.1016/S2214-109X(18)30... . India’s 2017 National Health Policy recommends reducing mortality from NCDs, including CRD, by 25% by 2025, whereas the study showed decreases greater than 30% in crude mortality rates from COPD and asthma between 1990 and 2016²⁵25. India State-Level Disease Burden Initiative CRD Collaborators. The burden of chronic respiratory diseases and their heterogeneity across the states of India: the Global Burden of Disease Study 1990–2016. Lancet Glob Health. 2018;6(12):e1363-74. https://doi.org/10.1016/S2214-109X(18)30409-1
https://doi.org/10.1016/S2214-109X(18)30... .

The trend of reaching the goal in most of Brazil corroborates a study by Malta (2016), who described the advances in MS’s plan after five years of implementation, in which monitoring by Vigitel demonstrated a reduction in premature mortality from the four main groups of NCDs, and a decrease in tobacco consumption²⁶26. Ministério da Saúde (BR), Departamento de Informática do SUS (DATASUS). Informações de Saúde: Rede Assistencial. Brasília, DF; 2017 [cited 2020 Oct 14]. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=0204
http://www2.datasus.gov.br/DATASUS/index... . The study highlights other actions in surveillance such as the performance of the PNS in 2013 and the Pesquisa Nacional de Saúde do Escolar (PeNSE – National Survey on Students’ Health) since 2009; besides initiatives in health and comprehensive care promotion, such as the redefinition of the Rede de Atenção à Pessoa com Doença Crônica (Care for People with Chronic Diseases Network), the release of the Programa Nacional de Melhoria do Acesso e da Qualidade da Atenção Básica (PMAQ – National Program for Access and Quality Improvement in Primary Care) and the Programa Farmácia Popular do Brasil (Brazilian Popular Pharmacy Program)²⁶26. Ministério da Saúde (BR), Departamento de Informática do SUS (DATASUS). Informações de Saúde: Rede Assistencial. Brasília, DF; 2017 [cited 2020 Oct 14]. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=0204
http://www2.datasus.gov.br/DATASUS/index... . The partnerships with educational and research institutions in structuring NCDs surveillance and funding, in 2011, the implementation of MS’s plan are also notewhorthy¹⁶16. Malta DC, Silva MMA, Moura L, Morais Neto OL. A implantação do Sistema de Vigilância de Doenças Crônicas Não Transmissíveis no Brasil, 2003 a 2015: alcances e desafios. Rev Bras Epidemiol. 2017;20(4):661-75. https://doi.org/10.1590/1980-5497201700040009
https://doi.org/10.1590/1980-54972017000... .

To better interpret the results of the trend to reach or not the goal, we investigated some factors: health care situation; social indicators; and exposure to smoking.

We studied the situation of health care by the Cadastro Nacional de Estabelecimentos de Saúde (CNES – National Registry of Health Facilities) and an article that estimated access to drug treatment of CRD¹⁴14. Leal LF, Bertoldi AD, Menezes AMB, Borges RB, Mengue SS, Gazzana MB, et al. Indicação, acesso e utilização de medicamentos para doenças respiratórias crônicas no Brasil: resultados da Pesquisa Nacional sobre Acesso, Utilização e Promoção do Uso Racional de Medicamentos no Brasil (PNAUM), 2014. Cad Saude Publica. 2018;34(10):e00202817. https://doi.org/10.1590/0102-311X00208217
https://doi.org/10.1590/0102-311X0020821... ,²⁶26. Ministério da Saúde (BR), Departamento de Informática do SUS (DATASUS). Informações de Saúde: Rede Assistencial. Brasília, DF; 2017 [cited 2020 Oct 14]. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=0204
http://www2.datasus.gov.br/DATASUS/index... . São Luís and Boa Vista, capitals that tend not to reach the goal, have a high relationship between population number and number of health facilities, the first one with 1,056.15 inhabitants for each establishment and the second one with 1,055.17 people per unit, what demonstrates possible health system overload in these municipalities²⁶26. Ministério da Saúde (BR), Departamento de Informática do SUS (DATASUS). Informações de Saúde: Rede Assistencial. Brasília, DF; 2017 [cited 2020 Oct 14]. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=0204
http://www2.datasus.gov.br/DATASUS/index... . However, considering only public establishments, the ratio between population and number of establishments ranks Boa Vista as the capital with the second-best relation, showing less overload of the public health system. Therefore, although this ratio is high, 3,485.26 inhabitants per public establishment, alone it does not justify the trend of the municipality not to reach the goal of MS’s plan, demonstrating probable influence of other factors.

Another capital that tends not to reach the goal is Recife, which has a ratio of 720.19 inhabitants per health facility, the 12th worst ratio, improving six positions when the restriction to public units is made. Vitória and Campo Grande have worse overload placements when disregarded private establishments, the first one presenting the best relation between all capitals when considered public and private establishments. Thus, the private sphere would be responsible for reducing the health system overload in these capitals, which could not possibly solve the issue of CRD, once most people depend on the Brazilian Unified Health System (SUS). Thus, it is possible that there is a relation between the ratio between population and number of health facilities and the time trend of mortality from CRD, which is best seen for São Luís and Boa Vista, but this influence is unclear and cannot be analyzed without considering other factors.

About access to medicines for CRD, Leal et al.¹⁴14. Leal LF, Bertoldi AD, Menezes AMB, Borges RB, Mengue SS, Gazzana MB, et al. Indicação, acesso e utilização de medicamentos para doenças respiratórias crônicas no Brasil: resultados da Pesquisa Nacional sobre Acesso, Utilização e Promoção do Uso Racional de Medicamentos no Brasil (PNAUM), 2014. Cad Saude Publica. 2018;34(10):e00202817. https://doi.org/10.1590/0102-311X00208217
https://doi.org/10.1590/0102-311X0020821... (2018) observed that, for individuals who claimed to have medical indication for pharmacological treatment, the prevalence of the use of medicines was lower among residents from North of Brazil (84%), followed by Northeast (92.8%)¹⁴14. Leal LF, Bertoldi AD, Menezes AMB, Borges RB, Mengue SS, Gazzana MB, et al. Indicação, acesso e utilização de medicamentos para doenças respiratórias crônicas no Brasil: resultados da Pesquisa Nacional sobre Acesso, Utilização e Promoção do Uso Racional de Medicamentos no Brasil (PNAUM), 2014. Cad Saude Publica. 2018;34(10):e00202817. https://doi.org/10.1590/0102-311X00208217
https://doi.org/10.1590/0102-311X0020821... . As two of the five capitals that tend not to reach the reduction target belong to the northeast region, and one to the north region, there may be a relation between this trend and low access to drug treatment, which is not valid for Campo Grande, because Midwest obtained the best prevalence (98%) of access to medicines for CRD. Southeast region of Brazil presented an 89.7% prevalence, ranking third in the country.

Regarding social indicators, on the one hand, São Luís is the second capital with the lowest mean monthly household income per capita (R$ 1,043.00), higher only than Macapá’s income. On the other hand, Vitória, which also tends not to reach the goal, has the highest mean income (R$ 2,988.00) among all capitals²²22. Stopa SR, Malta DC, Monteiro CN, Szwarcwald CL, Goldbaum M, Cesar CLG. Acesso e uso de serviços de saúde pela população brasileira, Pesquisa Nacional de Saúde 2013. Rev Saude Publica. 2017;51 Supl 1:3s. https://doi.org/10.1590/S1518-8787.2017051000074
https://doi.org/10.1590/S1518-8787.20170... . About concentration of wealth, we observed, from the Gini index, a classic inequality index that shows the distribution of per capita household income, in which Recife stands out for the greater inequality in income distribution²⁷27. Instituto Brasileiro de Geografia e Estatística. Síntese de Indicadores Sociais: uma análise das condições de vida da população brasileira. Rio de Janeiro: IBGE; 2017 [cited 2020 Oct 14]. (Estudos e pesquisas. Informação demográfica e socioeconômica; nº 37). Available from: https://renastonline.ensp.fiocruz.br/sites/default/files/arquivos/recursos/liv101459.pdf
https://renastonline.ensp.fiocruz.br/sit... .

São Luís and Recife, as well as Boa Vista, also present a high proportion of people below the poverty line, which corroborates the idea that social indicators influence the trend of mortality from CRD in these capitals²³23. Instituto Brasileiro de Geografia e Estatística. Síntese de indicadores sociais: uma análise das condições de vida da população brasileira. Rio de Janeiro: IBGE; 2019 [cited 2020 Oct 14]. (Estudos e pesquisas. Informação demográfica e socioeconômica; nº 40). Available from: https://biblioteca.ibge.gov.br/visualizacao/livros/liv101678.pdf
https://biblioteca.ibge.gov.br/visualiza... . The issue of schooling, an index in which Recife and São Luís showed lower rates than most capitals (98.80% and 98.50%, respectively),²⁸28. Instituto Brasileiro de Geografia e Estatística. Pesquisa Nacional por Amostra de Domicílios Contínua - PNAD Contínua. Rio de Janeiro: IBGE; 2017 [cited 2020 Oct 14]. Available from: https://www.ibge.gov.br/estatisticas/sociais/educacao/17270-pnad-continua.html?edicao=21073&t=sobre
https://www.ibge.gov.br/estatisticas/soc... among people aged between six and 14 years old, is also included. In an opposite way, Vitória and Campo Grande showed 100% schooling for this age group²⁸28. Instituto Brasileiro de Geografia e Estatística. Pesquisa Nacional por Amostra de Domicílios Contínua - PNAD Contínua. Rio de Janeiro: IBGE; 2017 [cited 2020 Oct 14]. Available from: https://www.ibge.gov.br/estatisticas/sociais/educacao/17270-pnad-continua.html?edicao=21073&t=sobre
https://www.ibge.gov.br/estatisticas/soc... .

The frequency of adults who smoke is the highest in Porto Alegre (14.40%) and lowest in São Luís (4.80%)¹³13. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Vigilância de Doenças e Agravos não Transmissíveis e Promoção da Saúde. Vigitel Brasil 2017: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico. Brasília, DF; 2018 [cited 2020 Oct 14]. Available from: https://bvsms.saude.gov.br/bvs/publicacoes/vigitel_brasil_2017_vigilancia_fatores_riscos.pdf
https://bvsms.saude.gov.br/bvs/publicaco... ; Campo Grande also has a high rate in relation to other capitals (10.80%), whereas the frequency in Boa Vista represents the median (7.20%), the same value observed in Recife and close to the one from Vitória (7.60%)¹³13. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Vigilância de Doenças e Agravos não Transmissíveis e Promoção da Saúde. Vigitel Brasil 2017: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico. Brasília, DF; 2018 [cited 2020 Oct 14]. Available from: https://bvsms.saude.gov.br/bvs/publicacoes/vigitel_brasil_2017_vigilancia_fatores_riscos.pdf
https://bvsms.saude.gov.br/bvs/publicaco... . The distribution is similar for the percentage of adults who smoke 20 or more cigarettes per day, with Porto Alegre leading, the frequency of São Luís being higher only than the one from Macapá¹³13. Ministério da Saúde (BR), Secretaria de Vigilância em Saúde, Departamento de Vigilância de Doenças e Agravos não Transmissíveis e Promoção da Saúde. Vigitel Brasil 2017: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico. Brasília, DF; 2018 [cited 2020 Oct 14]. Available from: https://bvsms.saude.gov.br/bvs/publicacoes/vigitel_brasil_2017_vigilancia_fatores_riscos.pdf
https://bvsms.saude.gov.br/bvs/publicaco... . Therefore, the association between the frequency of this risk factor and mortality from CRD is better seen in Campo Grande.

We based the choice of the SIM as the source of death data on studies that show its adequate performance, being well evaluated in usability, reliability, safety and quality in use²⁹29. Morais RM, Costa AC. Uma avaliação do Sistema de Informações sobre Mortalidade. Saude Debate. 2017;41 Nº Espec:101-17. https://doi.org/10.1590/0103-11042017s09
https://doi.org/10.1590/0103-11042017s09... . SIM is MS’s oldest information system, implemented in 1976 and, since then, expanding its stability and geographic coverage²⁹29. Morais RM, Costa AC. Uma avaliação do Sistema de Informações sobre Mortalidade. Saude Debate. 2017;41 Nº Espec:101-17. https://doi.org/10.1590/0103-11042017s09
https://doi.org/10.1590/0103-11042017s09... ,³⁰30. Pinto LF, Freitas MPS, Figueiredo AWS. Sistemas Nacionais de Informação e levantamentos populacionais: algumas contribuições do Ministério da Saúde e do IBGE para a análise das capitais brasileiras nos últimos 30 anos. Cienc Saude Coletiva. 2018;23(6):1859-70. https://doi.org/10.1590/1413-81232018236.05072018
https://doi.org/10.1590/1413-81232018236... . However, studies indicate limitations, mostly in interoperability and, especially in the North and Northeast, high under-registration and number of deaths from undefined causes²⁹29. Morais RM, Costa AC. Uma avaliação do Sistema de Informações sobre Mortalidade. Saude Debate. 2017;41 Nº Espec:101-17. https://doi.org/10.1590/0103-11042017s09
https://doi.org/10.1590/0103-11042017s09... ,³⁰30. Pinto LF, Freitas MPS, Figueiredo AWS. Sistemas Nacionais de Informação e levantamentos populacionais: algumas contribuições do Ministério da Saúde e do IBGE para a análise das capitais brasileiras nos últimos 30 anos. Cienc Saude Coletiva. 2018;23(6):1859-70. https://doi.org/10.1590/1413-81232018236.05072018
https://doi.org/10.1590/1413-81232018236... . Another limitation of the study is the generation of large confidence intervals of ARIMA models, for the forecast for 2022, even with a reasonable predictive capacity with 32.21% average percentage error for 2017, year left out of the estimate in order to test the forecasts.

CONCLUSIONS

The study demonstrates that early mortality rates from CRD are decreasing in Brazil as a whole and in most capitals, and we can observe in most of the country the trend to achieve the goal of the Plano de Ações Estratégicas para o Enfrentamento das DCNT (Strategic Action Plan to Tackle NCDs) in Brazil, 2011–2022, of reducing these rates by 2% per year. Such decline is consistent with the results of national and international studies, and may be assigned to initiatives in health, surveillance and comprehensive care promotion. The fight against smoking, the realization of the PNS, the monitoring by Vigitel and the redefinition of the Care for People with Chronic Diseases Network are noteworthy actions. The increased access to health services and the improvement of social indicators also seem to support the decreasing trend.

However, we observed that five capitals tend not to reach the goal: Boa Vista, Recife, São Luís, Campo Grande and Vitória. Among the factors that may explain this propensity are the health services overload (a relation better seen in São Luís and Boa Vista), low access to medicines for CRD in some regions, unfavorable social indicators and higher frequency of smoking (association more noted in Campo Grande). Nevertheless, it is possible to conjecture that there is influence of other factors in the time evolution of monthly mortality rates from CRD, especially for Vitória, capital with the highest average monthly household income per capita and schooling from six to 14 years of 100%.

Considering that CRD receive less investments in research and public attention, compared to other NCDs, this article contributes to a greater epidemiological knowledge of this group of diseases and to the monitoring of public health policies. We hope that this study allows a better planning of the public health system for the control of CRD and promotes initiatives in the research, surveillance and health care areas, especially in the capitals, and results show a trend not to achieve the goal of the Brazilian Ministry of Health.

REFERENCES

Publication Dates

History