110.05 Mortality by external causes in the intra-urban space in Brazil

Wednesday, April 29, 2009
Refik Saydam (The Hilton Istanbul Hotel )
Edna M. Araujo Universidade Estadual de Feira de Santana, Bahia, Brazil, Brazil
Maria Conceição Nascimento Costa Universidade Federal da Bahia, Brazil
Nelson F. Oliveira Universidade Estadual de Feira de Santana, Bahia, Brazil, Brazil
Vijaya K. Hogan University of North Carolina, Chapel Hill, USA
Francisco Santana Divisão de Vigilância Epidemiológica da Secretaria de Saude do Estado da Bahia, Brazil
Tania Maria Araujo Universidade Estadual de Feira de Santana, Bahia, Brazil, Brazil
Introduction: In Brazil, deaths by external causes have been ranking the first in the statistics concerning mortality. Nevertheless, studies which investigate the relationship between that group of causa mortis and race/skin color are scarce. Objectives: To evaluate the relative contribution of race/skin color when determining the special distribution of mortality by external causes in Salvador, Brazil from 1998 to 2003. Material and Methods: Study of spatial aggregate contribution whose analysis unit is the ponderation area (AP). The data were collected from the Nina Rodrigues Forensic Medicine Institute, the Brazilian Institute of Geography and Statistics. Annual average death rates by all the external causes and also by homicide by AP were estimated during the accomplishment of this study. Moran’s Global Index was used to evaluate the presence of spatial autocorrelation. The Conditional Auto Regressive (CAR) model was used to evaluate the referred effect by making use of R statistical package. The use of Negative Binomial Regressive stood out in order to obtain a punctual risk measure. Results: Moran’s test I shows that, the areas of high mortality rates by all the external causes as well as homicide tended to aggregate and in these areas greater proportion of Black male population aged between 15 and 49 years was observed. In the negative binomial modeling a statistically significant Relative Risk (RR) equal=1,18; P=0,0003; IC 1.011 – 1,368 was obtained for the association between the  proportion of the male population whose ages range from 15 to 49 years and the mortality rate by all the external causes adjusted by co-variables. Conclusion: The interrelation between race, violence and space is part of a long process of inequality; therefore, its understanding requires the junction of interdisciplinary efforts which contribute to improve the knowledge about theme and, as a result, lead to more specific public health interventions. 

Learning Objectives: Evaluate the relative contribution of race/skin color when determining the spatial distribution of mortality by external causes in Salvador, Brazil from 1998 to 2003.

Sub-Theme: Social determinants of health and disease