IJCS | Volume 31, Nº2, March / April 2018

124 Soares et al. Cardiovascular Mortality X Gross Domestic Product Int J Cardiovasc Sci. 2018;31(2)123-132 Original Article In 2010 and according to data from the Brazilian Institute of Geography and Statistics (IBGE), the Rio de Janeiro State, then divided into 92 municipalities, had 15,989,929 inhabitants, with a population density of 365.23 inhabitants/km 2 . The Gross Domestic Product (GDP) of the Rio de Janeiro State corresponds to 11.3% of the Brazilian GDP. 6 The Rio de Janeiro State municipalities have a very heterogeneous socioeconomic structure. Some municipalities, such as Porto Real, have a GDP per capita (GDPpc) that exceeds R$ 200,000.00, and others, such as Japeri, have a GDPpc of R$ 5,000.00, similar to that of some countries, such as Congo, Samoa and Swaziland, and much lower than that of the Brazilian mean of R$ 19,000.00. 6 Some Rio de Janeiro State municipalities, such as São Francisco de Itabapoana, have a poverty index greater than 36%, while others, such as Niterói and Volta Redonda, have a poverty index lower than 10%. The poverty index considers three variables: the short duration of life (the population percentage that does not reach the age of 40 years), the lack of elementary education (the illiterate percentage of the population), the lack of access to public and private resources (the population percentage that has access to neither health care service nor potable water, and of malnourished children). 7 Some studies have assessed the evolution of mortality fromDCS and its major two subgroups in Brazil, ischemic heart diseases (IHD) and cerebrovascular diseases (CBVD), 5, 8-11 however, studies correlating that mortality with socioeconomic indicators per municipality are rare. Therefore, a study with the Rio de Janeiro State municipalities, which have a varied and heterogeneous socioeconomic structure, will allow us to build models about the evolution of the mortality rates from DCS and of GDPpc, estimating correlations between those variables aiming at suggesting factors involved in reducing the mortality rates from DCS, IHD and CBVD. Methods This study collected data on GDPpc and mortality in Rio de Janeiro State municipalities, which were analyzed according to the geopolitical structure of the year 1950, gathering the emancipated municipalities with their original headquarters from that date on. Those aggregates of municipalities caused a reduction in the total number of Rio de Janeiro State municipalities from 92 in 2010 to 56 aggregates for this study analysis. In addition, those aggregates of municipalities were analyzed by region. This study used the regional division proposed by the Rio de Janeiro State Secretariat of Health with a change, subdividing the Metropolitan region into theMetropolitan Belt, which comprises all municipalities in the region except for the municipalities of Rio de Janeiro and Niterói, which constituted two autonomous regions. The other regions, Mid-Paraíba, Mountain, Northern, Coastal Lowlands, Northwestern, Southern- Central, and Ilha Grande Bay, are those defined by the Rio de Janeiro State Secretariat of Health. 12 The GDP data were obtained from the Applied Economic Research Institute ( Instituto de Pesquisa Econômica Aplicada ) 13 for the years 1949, 1959, 1970, 1975, 1980 and 1985 to 2010. The population data were obtained from the IBGE 6 for the general census years (1950, 1960, 1970, 1980, 1991, 2000 and 2010) and population counting (1996). Intercensal population estimates were calculated with the arithmeticmethod by use of the census years or population counting immediately before or after. Those estimateswere performed for the fractions corresponding to the agegroups, at 10-year intervals, for each sex. TheGDPpcwas calculated by dividing the absolute and the municipality GDP by the population in the corresponding year. Then the GDPpc was converted into dollars (1 dollar = 3.2  reais , currency exchange rate of April 2015). To calculate the mortality rates, the mortality data restricted to adults aged 20 years and older from the database DATASUS-MS were analyzed. 14 Such data were divided into the major fractions of interest in this study: DCS, corresponding to the codes listed in chapter VII of ICD-9 15 or chapter IX of ICD-10; 16 IHD, corresponding to the codes 410-414 of ICD-9 or codes I20-I25 of ICD‑10; CBVD, corresponding to the codes 430-438 of ICD-9 or codes I60-I69 of ICD-10. In addition, the deaths from ill-defined causes (IDC), listed in chapter XVI of ICD‑9 and chapter XVIII of ICD-10, as well as the total of all‑cause (AC) deaths were used in the analysis. The ICD-9 was in force until 1995, while ICD-10 has been since 1996. The crude and sex- and age-adjustedmortality rates were calculated by use of the direct method 17 per 100,000 inhabitants. The mortality rates from IDC in Rio de Janeiro State have increased significantly since 1990, 8 thus, compensation was performed, consisting in assigning to deaths fromDCS, IHDandCBVD their part of deaths from IDC, corresponding to the fractions observed in the defined deaths, that is, excluded those from IDC. After compensation of the deaths from DCS, IHD and CBVD for those from IDC, sex- and age-adjustedmortality rates were estimated. The standard population for the adjustments was that of Rio de Janeiro State registered