Três alternativas estocásticas para modelar morbimortalidade por doenças respiratórias e cardiovasculares via variáveis atmosféricas
GOMES, Ana Carla dos Santos. Três alternativas estocásticas para modelar morbimortalidade por doenças respiratórias e cardiovasculares via variáveis atmosféricas. 2015. 97f. Tese (Doutorado em Ciências Climáticas) - Centro de Ciências Exatas e da Terra, Universidade Federal do Rio Grande do Norte, Natal, 2015.
Gomes, Ana Carla dos Santos
Climate and air pollution, among others, are responsible factors for increase of health vulnerability of the populations that live in urban centers. Climate changes combined with high concentrations of atmospheric pollutants are usually associated with respiratory and cardiovascular diseases. In this sense, the main objective of this research is to model in different ways the climate and health relation, specifically for the children and elderly population which live in São Paulo. Therefore, data of meteorological variables, air pollutants, hospitalizations and deaths from respiratory and cardiovascular diseases a in 11-year period (2000-2010) were used. By using modeling via generalized estimating equations, the relative risk was obtained. By dynamic regression, it was possible to predict the number of deaths through the atmospheric variables and the betabinomial-poisson model was able to estimate the number of deaths and simulate scenarios. The results showed that the risk of hospitalizations due to asthma increases approximately twice for children exposed to high concentrations of particulate matter than children who are not exposed. The risk of death by acute myocardial infarction in elderly increase in 3%, 6%, 4% and 9% due to high concentrations CO, SO2, O3 and PM10, respectively. Regarding the dynamic regression modeling, the results showed that deaths by respiratory diseases can be predicted consistently. The beta-binomial-poisson model was able to reproduce an average number of deaths by heart insufficiency. In the region of Santo Amaro the observed number was 2.462 and the simulated was 2.508, in the Sé region 4.308 were observed and 4.426 simulated, which allowed for the generation of scenarios that may be used as a parameter for decision. Making with these results, it is possible to contribute for methodologies that can improve the understanding of the relation between climate and health and proved support to managers in environmental planning and public health policies.