Premature mortality due to cardiovascular disease and social inequalities in Porto Alegre: from evidence to action

Abstracts

FUNDAMENTO: Duas perspectivas, a econômica (doença causando empobrecimento) e a social (pobreza causando adoecimento), têm disputado internacionalmente a justificação de políticas públicas de saúde. OBJETIVO: Investigar a relação entre mortalidade precoce por doenças cardiovasculares (DCV) e condições socioeconômicas (SE) em Porto Alegre (PA), e discutir fundamentos e estratégias para a prevenção das DCV. MÉTODOS: Análise ecológica da associação entre mortalidade por DCV aos 45-64 anos e condições SE de 73 bairros de PA. Estimou-se o risco relativo (RR) e a fração do risco atribuível (FRA) às desigualdades entre bairros agrupados em 4 estratos SE. RESULTADOS: A mortalidade precoce por DCV foi 2,6 vezes maior nos bairros classificados no pior comparado ao melhor de 4 estratos SE. Entre bairros extremos, o RR chegou a 3,3 para as DCV e 3,9 para as doenças cerebrovasculares. Comparada à mortalidade no melhor estrato, 62% dos óbitos precoces do pior estrato e 45% dos da cidade como um todo seriam atribuíveis à desigualdade socioeconômica. CONCLUSÃO: Quase a metade da mortalidade por DCV antes do 65 anos pode ser atribuída à pobreza. A doença, por sua vez, contribui para a pobreza e reduz a competitividade do país. É preciso reduzir o adoecimento e recuperar a saúde dos mais pobres com investimentos que resultem em desenvolvimento econômico-nacional e melhoria das condições sociais da população.

Mortalidade; doenças cardiovasculares; prevenção de doenças; iniqüidade social

ORIGINAL ARTICLES

Premature mortality due to cardiovascular disease and social inequalities in Porto Alegre: from evidence to action

Sérgio Luiz Bassanesi; Maria Inês Azambuja; Aloyzio Achutti

Universidade Federal do Rio Grande do Sul, Academia Sul-Riograndense de Medicina, Porto Alegre, RS - Brazil

Mailing address

SUMMARY

BACKGROUND: Two perspectives, the economic (disease causing impoverishment) and social (poverty causing illness), have internationally disputed the justification for public health policies.

OBJECTIVE: To investigate the relationship between early mortality by cardiovascular disease (CVD) and socioeconomic (SE) conditions in the city of Porto Alegre (PA), and discuss bases and strategies for the prevention of CVD.

METHODS: An ecological analysis of the association between mortality by CVD at 45-64 years of age and SE conditions of 73 districts/neighborhoods in PA. The relative risk (RR) and the fraction of risk (FRA) attributable to inequality among the districts grouped into 4 SE strata were estimated.

RESULTS: Early mortality by CVD was 2.6 times higher in the districts classified in the worst compared to the best of the 4 SE strata. Among the extreme districts, the RR reached 3.3 for CVD and 3.9 for cerebrovascular disease. Compared to the mortality in the best stratum, 62% of the early deaths in the worst stratum and 45% of those in the city as a whole could be attributed to socioeconomic inequality.

CONCLUSION: Almost half of the mortality by CVD before 65 years of age can be attributed to poverty. Disease, on the other hand, contributes towards poverty and reduces competitiveness of the country. It is necessary to reduce illness and recover the health of the poorest inhabitants with investments that result in national economic development and improvement of the social conditions of the population.

Key words: Mortality; cardiovascular diseases; disease prevention; social inequity.

Introduction

Two perspectives, the macroeconomic (disease causing impoverishment)¹ and the social (poverty causing disease)^2-4, have been disputed as justifications for public health policies.

In the area of chronic diseases, as of 2001, international organizations have alerted to the risk of migration of the CVD epidemic from central countries to those of middle and low incomes due to populational aging, urbanization, and the increased capacity for consumption of the inhabitants of these countries^5-7. To prevent these chronic illnesses, they have recommended restructuring health care services in order to prioritize the early identification of individuals at risk and their ongoing treatment^5-7.

The macroeconomic argument has been emphatically used to stimulate immediate action. Based on demographic estimates, Leeder et al¹ suggest that there is a two-decade window of opportunity to establish awareness and avoid catastrophic consequences for countries twenty to forty years from now.

The perspective of social determinants in illness had little expression in the central countries during the 20^th century, and reemerged during the transition to the 21^st century because of economic globalization⁸. Nevertheless, attention given to chronic diseases in this agenda is still irrelevant⁹.

This study intends to investigate the association between early mortality by cardiovascular disease and socioeconomic inequalities in Porto Alegre, and suggest paths to prevention based on the interpretation of the findings.

Methods

This is a cross-section ecological study, which has as analysis units the districts of the Porto Alegre municipality. The city was chosen because of its high average Human Development Index (HDI = 0.865 in 2000¹⁰) despite marked inequalities in social development indicators (from 0.46 to 0.93) among its regions¹¹.

Data relative to the variables "age", "gender", "basic cause of death (ICD10)", and "place of residence" were extracted from a databank with 51,562 georeferenced deaths during the period between 2000 and 2004, the number of liveborns in the information system of liveborns in Porto Alegre from 2000 to 2004, and information relative to demographic and socioeconomic variables of microdata from the year 2000 Census. Based on the available primary variables, those used in this study and listed in ^{Panel 1} were selected or constructed. The decision to use early deaths instead of all deaths was made in order to increase the specificity of cardiovascular diagnoses and to avoid the possibility of the greater longevity of the richest individuals artificially inflating the risk of death by cardiovascular disease in this population stratum.

All data were pooled by district of residence. Data relative to all the original 9 districts with less than 3,000 inhabitants in 2000 were incorporated into contiguous districts, resulting in a universe of 73 study units.

In order to minimize random fluctuation in the estimates of annual mortality by CVD, ischemic heart disease (IHD), and cerebrovascular disease (CeVD) in the district, in addition to the mean 5-year mortality rate, empirical Bayesian attenuation was used. Assuming the existence of a spatial tendency in event distribution, the occurrences were reestimated taking into consideration the annual means of cases in the neighboring districts.

Analyses were carried out using as units the districts and their distribution in 4 SE risk strata. Initially, mortality cartograms were made stratifying the districts into 4 levels of each of the seven socioeconomic variables selected (Fig. 1). The cartograms suggested a strong spatial correlation among the independent variables.

In order to optimize stratification as per the set of independent variables, 3 different strategies were used: cluster analysis, analysis of the principal components, and Moran spatial autocorrelation. The cluster analysis defined, for the set of variables, four strata as the groupings with the least variability within each one and the greatest difference among them. The distribution of the districts in the strata defined by the K-means method is shown on Figure 2A. Analysis of the principal components sought to reduce the number of independent variables (due to the high spatial autocorrelation suggested by the initial cartogram) prior to the stratification of the districts. The technique resulted in the production of a single component, with the power to explain 74% of the variation among the study units and highly correlated with all the variables. This component was parameterized in order to be read as a score with a mean equal to zero and standard deviation equal to one, in which the more negative values were associated with the better districts and the more positive values with the worst districts. Stratification of the districts in four levels according to this score can be seen on the cartogram in Figure 2B. This score was also used to estimate the difference in risk and the relative risk among the extreme districts (see below). For Moran's autocorrelation, a matrix of adjacent neighborhoods was used. Each one of the independent variables was tested and the score was generated by the analysis of the principal components. The objective was to assess if the neighboring areas are similar regarding the variable in question compared to the standard that would be expected in a situation of complete spatial randomness. The autocorrelation was considered significant when p<0.05. The Moran I Global Indices consistently showed highly significant global spatial autocorrelation for all the independent variables and for the principal component. The Moran I Local Indices enabled mapping of the districts according to whether they were surrounded by neighborhoods with equivalent (low-low or high-high) or different (low-high or high-low) rates of risk, guiding the delimitation of borders among the strata. The splatter plot for the score generated in the principal component analysis may be seen in Figure 2C. The final district SE stratification proposal took into account the coincidences among the three methods and the borders established by Moran's method. In the few cases with unclear district classification, reference was made to the original cartograms of the education and income variables, and personal knowledge of the city was used. The final stratification of the districts into strata 1, 2, 3, and 4, respectively, as high, medium-high, medium-low, and low socioeconomic levels is presented in Figure 2D.

The 4 strata reflect approximate quartiles of the quality of life levels into which the districts are distributed. Based on them, estimates were made of the relative risk (RR) and fraction of risk (FRA) of early cardiovascular death attributable to SE inequality between stratum 1 (best) and 4 (worst) SE level, and between stratum 1 and the city as a whole.

With data referring to each district, a matrix was produced to correlate all the variables. In order to summarize the association between the SE condition of the district and early mortality by CVD, IHD, and CeVD, multiple linear regression analyses were performed. There was a spatial correlation among the independent variables and among the dependent variables, but not in the associations among them, i.e., the association in one district occurred independently from the association in the neighboring districts. Thus, there is no need to report spatial regression techniques (carried out with similar findings).

In order to describe early CVD mortality inequality among the districts with extreme SE conditions, the Slope Index of Inequality (SII) and the Relative Index of Inequality (RII) were estimated. Calculation of these estimates requires the creation of an intermediate variable (Ridit), an indicator of the position of mortality relative to the socioeconomic variable, which in this case was the score of the principal component previously described. The districts were ordered according to their SE scores. Next, calculations were made of the Relative Frequency (RF) and Cumulative Frequency (CF) of early deaths weighted by the size of the population, followed by the Ridit = (CF + (CF-RF))/2. SII is the linear regression coefficient of the coefficient of mortality over the Ridit, and RII is the difference between the extreme values of the regression line.

Excel software was used to organize the databank: SPSS 13.0 for Windows and Statistica for non-spatial analyses, GeoDa and SigEpi for spatial analyses, Tabwin for thematic maps, and Brechas 1.0 for measurements of inequalities among the districts. The project was approved by the Ethics Committee of the Hospital Moinhos de Vento, where it was based. The study was partially financed by the Initiative for Cardiovascular Health, CDC-Delhi.

Results

Between 2000 and 2004, mortality by CVD for 45-64 years of age corresponded to 28.5% of the total deaths and 22.8% of all deaths by CVD. In this age bracket, 40% of the cardiovascular deaths were due to IHD and 30% were due to cerebrovascular disease (CeVD).

Table 1 shows the number of districts, populations, and means of the independent and dependent variables per socioeconomic stratum. There is a linear gradient in the measurements of the independent variables between the best stratum (stratum 1) and the worst stratum (stratum 4). There is also a linear tendency towards increased cardiovascular mortality between stratum 1 and stratum 4. The consistency of the data enables an easy and direct interpretation of these results.

Figure 3 shows the distribution of early mortality by CVD, IHD, and CeVD in the four strata, and figure 4 illustrates the inequalities among the strata. Early mortality by cardiovascular disease is 2.6 times greater in stratum 4 compared to stratum 1. For the stratum with the worst indicators and for the city as a whole, the fractions of mortality (FRA) attributable to social inequality using as reference stratum 1 were 62% and 44.7%, respectively. In other words, 62% of early deaths by CVD in stratum 4 and 44.7% in the city could have been avoided if the respective population had the same socioeconomic conditions as stratum 1. For IHD and CeVD, the fractions of risk attributable to SE inequality were 39.8 and 47.6%, respectively.

The matrix of correlations in figure 5 suggests a high linear-type correlation between the variables. The independent variables that showed the most pronounced relationship with the cardiovascular mortality indicators were income, education, and violence. In the multiple linear regression analysis, violence and income remained independent and significantly associated with the three outcomes analyzed. The final mathematical model, considering all the districts and mortality by all CVD, was Cardiovascular Mortality 45-64 years = 233.77 + 184.79 Violence - 2.6 Income

The coefficient of determination (r²) resulting from this equation is 0.61, i.e., 61% of the variability in the distribution of the coefficient of early mortality by CVD can be explained by the variability in the distribution of Violence and Income. Equivalent findings were noted for the outcomes IHD and CeVD.

The estimated risk of mortality by early CVD in the district with the best socioeconomic situation (corresponding to one Ridit = 1) was 123.1/100,000, and in the district with the worst situation (Ridit = 0) it was 402.5/100,000. The estimated difference (SII) was -279.5/100,000 and the Relative Inequality Index (RII) between the extremes corresponded to 402.5/123.1, or 3.3. In other words, mortality by cardiovascular disease is 3.3 times greater in the worst district relative to the best district. RII was 2.5 for ischemic heart disease, and 3.9 for cerebrovascular disease (Figure 6).

Discussion

As is true for the United States and Europe, since 1980, mortality rates (deaths/100 thousand inhabitants) due to IHD and cerebrovascular disease dropped significantly in Brazil, for all age groups^12-13. Even so, the demand for treatment of chronic diseases at health care centers is expected to climb in the next decades, accompanying the dislocation of cohorts born during the period of highest fertility rates (up until approximately 1965-70)¹⁴. Until 2020, the greatest growth in number of adults will occur between the ages of 45 and 64 years. After 2020, growth will shift progressively towards the older population¹⁴. Mortality by CVD in the 45-64 years of age population is high in Brazil compared to developed countries, especially among the women^15-16. This difference increases with the decrease in age. In the younger group, aged 35 to 44 years, mortality by acute myocardial infarct and cerebrovascular disease in 1988 was 3 and 5 times higher, respectively, in Brazil than in the USA among men and 4 and 6 times higher, respectively, among women¹⁶. The findings of this study support the idea that this difference is associated with the poor quality of life of the populations in the large urban centers compared to those of developed countries.

Even in a city with a relatively high HDI such as Porto Alegre, almost half of the early deaths (45%) could be avoided if all the districts had the best conditions of the 4 SE strata, which means 80% of the excessive early deaths in the city attributable to this inequality of socioeconomic conditions among districts. Two variables, the average of years of education of the household heads - an indicator of the antecedents of social inclusion/exclusion of the residents - and mortality by external causes - an indicator of the current exposure to risk, were capable of together explaining 61% of the distribution of mortality by CVD among the districts.

These findings are consistent with others published recently in Brazil^17-19. Ishitani et al¹⁷ in studying the deaths of adults 35-64 years old between 1999 and 2001 in 98 municipalities with more than 100,000 inhabitants, 90% of them residents of the urban area, showed, by simple and multiple linear regression, a negative correlation between CVD and income/educational level, and a positive association with poverty and poor living conditions. In São José do Rio Preto, SP, Godoy et al¹⁸ showed that the principal component extracted from the variables that reflects income and education in the domiciles explained 87% of the variation among the census strata, and stratification in quartiles based on this component was associated with a mortality by CVD 40% higher in the worst compared to the best stratum. Melo et al¹⁹ in studying the spatial distribution of mortality by acute myocardial infarct in Rio de Janeiro, described a spatial distribution associated with a strong socioeconomic gradient.

Differences in the occurrence of illness and death may be attributed to differences in access to treatment of patients²⁰ and the difference in exposure to risk factors^1,21,22. Similarly, intervention proposals emphasize the amplification of investments in health care services geared towards early identification and ongoing accompaniment of individuals with chronic diseases and their risk factors²³. Nevertheless, individual treatment of ill or high-risk patients, while fulfilling an important humanitarian role, has a small impact in reducing populational rates of illness, since action is taken in reference to the cause of those cases, but not to the cause of the occurrences²⁴.

Nancy Krieger has proposed that social inequality in health care is the result of an embodiment of unequal social experiences - embodying inequality²⁵. Among the conditions biologically incorporated but with strong social determinants are birth weight, height, and immune responses to acquired infections²⁵, which are known to be associated with the acquisition of metabolic and immune phenotypes of risk for some causes of illness²⁶. In the case of atherosclerosis, the transition from the degenerative to the inflammatory paradigm²⁷ would enable these SE differences in the occurrence of CVD to be attributed to the interaction between unequal accumulation of biological vulnerability over a lifetime (low birth weight, infections, and stress) and unequal levels of current exposure to environmental factors^26,28. Investment in prevention, in this case, would be an investment in the improvement of living conditions of the population.

Even though history never repeats itself, much can be learned from it. There are significant coincidences between the present situation of the large Brazilian cities and European cities at the beginning of the 19^th century^29-32. As is the case here, industrialization promoted a rapid increase in the urban population with the appearance of shanty towns/slums and great social inequality in mortality^30-31. During the 100 years of the industrial revolution, the population of Paris and London increased 5 times, and that of Berlin, 10 times³¹. In 50 years, between 1950 and 2000, the Brazilian urban population went from 19 million to 146 million inhabitants, i.e., it increased more than 7 times³³. Between 1830 and 1840, Villermé in France and Chadwick in England showed that mortality was greater in the large cities. Villermé showed that mortality was 50% higher in the poorest districts³¹. Tuberculosis, followed closely by pneumonia and influenza, was the primary cause of death in European cities³¹. The question debated then was if poverty caused disease or disease caused poverty^29-32. The interrelationship between disease and social development was so significant in Europe that the famous phrase spoken by Virchow dates back to this period (1848): "Medicine is a social science, and politics is nothing more than medicine on a large scale." Also in 1948, after documenting the horrific health conditions of the working population in Great Britain, Edwin Chadwick – the father of English Health Care³² - defended and promoted the approval of the "Public Health Act." Chadwick's argument was that generalized disease resulted in poverty for the workers and a significant loss of economic productivity for England. The solution he proposed was to carry out investments in sanitary engineering projects, which besides benefiting health directly, would stimulate economic growth, create jobs, and widen access of the poorest to food (reduction of poverty), thus contributing to maintaining social order³².

In Brazil today, as was the case in Europe 150 years ago, the great challenge is the reduction of excessive illness among poor adults. Tuberculosis was substituted by CVD as the main cause of death. The expected growth in the number of cases of early CVD over the next 15-20 years, and in older individuals after that, will have a significant impact on the budgets of the National Unified Health System [SUS] and the Ministry of Social Security and Social Assistance [Previdência Social]^28,34-35, and affect the productivity and competitiveness of the country in the globalized market.

Chadwick's example demonstrates that it is not necessary to chose between investing in the prevention of diseases in order to reduce the economic impact of illness¹ or intervening in poverty in order to reduce the rates of disease and early death². The best alternative would be to integrate the two strategies, i.e., prioritize health care policies, which besides resulting in direct gains in terms of health care, would stimulate national economic growth and improvement of social conditions. Moreover, it would advance intersector policies of economic development that value the promotion of health and the reduction of morbimortality by chronic illnesses in the population.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

This study was partially funded by IC-Health Delhi (Índia).

Study Association

This study is not associated with any graduation program.

References

• 1. Leeder S, Raymond S, Greenberg H; Liu H. A race against time: the challenge of cardiovascular disease in developing economies. The Earth Institute: Columbia University; 2004. [acesso em 2007 Aug 20]. Disponível em: http://www.eldis.org/go/display/?id=18750&type=Document
• 2. Marmot M. Social determinants of health inequalities. Lancet. 2005; 365: 1099-104.
• 3. Almeida-Filho M, Kawachi I, Pellegrini Filho A, Dachs JNW. Research on health inequalities in Latin America and the Caribbean: bibliometric analysis (1971-2000) and descriptive content analysis (1971-1995). Am J Public Health. 2003; 93: 2037-43.
• 4. Bassanesi SL. Mortalidade por doença isquêmica no coração no estado do Rio Grande do Sul: aspectos geográficos e sócio-econômicos. [Tese]. Porto Alegre: Universidade Federal do Rio Grande do Sul; 1997.
• 5. Epping-Jordan JA. The challenge of chronic conditions: WHO responds. [Editorial]. BMJ. 2001; 323: 947-8.
• 6. Lenfant C. Can we prevent cardiovascular disease in low and middle income countries? Bull WHO. 2001; 79: 980-2.
• 7. Strong K, Mathers C, Leeder S, Beaglehole R. Preventing chronic diseases: how many lives can we save? Lancet. 2005; 366: 1578-82.
• 8. Labonté R, Schrecker T. Globalization and social determinantes of health. Introduction and methodologic background. Global Health. 2007; 3: 7.
• ⁹
  Council of Science Editors. Global theme issue on poverty and human developments: October 22, 2007. [citado em 2006 jul 10]. Disponível em: http://www.councilscienceeditors.org/globalthemeissue.cfm?printPage=1&
• ¹⁰
  Observatório da Cidade de Porto Alegre. [acesso em 2006 dez 12]. Disponível em: http://www2.portoalegre.rs.gov.br/observatorio/default.php?p_secao=4
  » link
• 11. Furtado A, Costa BM, Macedo CEG, Germano LR, Macerata MA, Silva ME, et al. Mapa da inclusão e exclusão social de Porto Alegre. Prefeitura Municipal de Porto Alegre: Secretaria do Planejamento Municipal; 2004.
• 12. Lotufo PA. Por que não vivemos uma epidemia de doenças crônicas: o exemplo das doenças cardiovasculares? Cienc saúde coletiva. 2004; 9: 844-7.
• 13. de Souza MF, Alencar AP, Malta DC, Moura L, Mansur AP. Serial temporal analysis of ischemic heart disease and stroke death risk in five regions of Brazil from 1981 to 2001. Arq Bras Cardiol. 2006; 87: 735-40.
• 14. Oliveira, JC, Albuquerque FR, Lins IB. Projeção da população do Brasil por sexo e idade para o período 1980-2050: revisão 2004, metodologia e resultados. (IBGE). [citado em 2007 out 10]. Disponível em: http://www.ibge.gov.br/home/estatistica/populacao/projecao_da_populacao/metodologia.pdf
• 15. Lotufo PA. Premature mortality from heart diseases in Brazil: a comparison with other countries. Arq Bras Cardiol. 1998; 70: 321-5.
• 16. Chor D, Fonseca MJM, Andrade CR. Doenças cardiovasculares: comentários sobre a mortalidade precoce no Brasil. Arq Bras Cardiol. 1995; 64: 15-9.
• 17. Ishitani LH, Franco GC, Perpétuo IH, França E. Desigualdade social e mortalidade precoce por doenças cardiovasculares no Brasil. Rev Saúde Pública. 2006; 40: 684-91.
• 18. Godoy MF, Lucena JM, Miquelin AR, Paiva FF, Oliveira DLQ, Augustin Jr JL, et al. Mortalidade por doenças cardiovasculares e níveis socioeconomicos na população de São José do Rio Preto, Estado de São Paulo, Brasil. Arq Bras Cardiol. 2007; 88: 200-6.
• 19. Melo EC, Carvalho MS, Travassos C. Distribuição espacial da mortalidade por infarto agudo do miocárdio no Rio de Janeiro. Brasil. Cad Saúde Pública. 2006; 22: 1225-36.
• 20. Travassos C, Viacava F, Fernandes C, Almeida CM. Desigualdades geográficas e sociais na utilização de serviços de saúde no Brasil. Cienc saúde coletiva. 2000; 1: 133-49.
• 21. Duncan BB, Schmidt MI, Achutti AC, Polanczyk CA, Benia LR, Maia AA. Socioeconomic distribution of noncommunicable disease risk factors in urban Brazil: the case of Porto Alegre. Bull Pan Am Health Organ. 1993; 27: 337-49.
• 22. Yusuf S, Hawken S, Ounpuus S, Dans T, Avezum A, Lanas F, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet. 2004; 364: 937-52.
• ²³
  Plano de reorganização da atenção à hipertensão arterial e ao diabetes mellitus. Rev Saúde Pública. 2001; 35: 585-8.
• 24. Rose G. Sick individuals and sick populations. Int J Epidemiol. 1985; 14: 32-8.
• 25. Krieger N, Davey Smith G. "Bodies count," and body counts: social epidemiology and embodying inequality. Epidemiol Rev. 2004; 26: 92-103.
• 26. Azambuja MI, Levins R. Coronary heart disease (CHD) one or several diseases? Changes in the prevalence and features of CHD. Persp Biol Med. 2007; 50: 228-42.
• 27. Ridker PM. C-reactive protein and the prediction of cardiovascular events among those at intermediate risk: moving an inflammatory hypothesis toward consensus. J Am Coll Cardiol. 2007; 49: 2129-38.
• 28. Lessa I. Editorial. Cienc saúde coletiva. 2004; 9 (4): 828.
• 29. Szreter S. The population health approach in historical perspective. Am J Public Health. 2003; 93: 421-31.
• 30. Szreter S. Industrialization and health. Br Med Bull. 2004; 69: 75-86.
• 31. Cairnes J. Matters of life and death: perspectives on public health, molecular biology, cancer and the prospects for the human race. New Jersey: Priceton University Press; 1997.
• 32. Susser E, Bresnahan M. Origins of epidemiology. Ann NY Acad Sci. 2001; 954: 6-18.
• 33. Rocha RM. A ocupação e o processo de urbanização sem planejamento no eixo rodoviário do complexo territorial BrasíliaGoiânia. Brasília: Faculdade de Arquitetura e Urbanismo Programa de Pós Graduação; 2006. p. 3.
• 34. Achutti A, Azambuja MI. Chronic non-communicable diseases in Brazil: the health care system and the social security sector. Cienc saúde coletiva. 2004: 9: 833-40.
• 35. Azambuja MI, Foppa M, Achutti A. Economic burden of severe cardiovascular disease in Brazil: an estimation based on secondary data. Arq Bras Cardiol. (in press).

Publication Dates

• Publication in this collection
  25 June 2008
• Date of issue
  June 2008

History

• Received
  12 Nov 2007
• Reviewed
  09 Jan 2008
• Accepted
  15 Jan 2008