PRINCIPIOS METODOLÓGICOS PARA LA EVALUACIÓN DE IMPACTO DE LA VARIABILIDAD Y EL CAMBIO CLIMÁTICO EN LA SALUD HUMANA. UN ENFOQUE

Signal detection of climate variability or change and the evaluation of its specific effects, requires an understanding of the variations in the observed data, which describe the natural climate variability and change signals. It is also necessary to understand the complex interactions that make up the climate system. In the present work, an unusual methodological approach is taken to evaluate the effects and impacts of climate variability and change on the behavior of different diseases, on the basis of practical experience of its application in four countries of the Caribbean, Central and South America: Cuba, Panama, Bolivia and Paraguay. For the determination of the climate signal change multivariate analysis techniques (empirical orthogonal functions) were used, combined with robust methods of time series decomposition (decomposition by median). They allowed us to describe the changes observed in the seasonal patterns of climate and epidemiological diseases for the period 1991-1999, with respect to the period 1961-1990. These results were used to build an autoregressive model with non-constant variance, with a climate index based on the signals obtained from the decompositions, which enters the model as an exogenous variable in order to make projections of the diseases.