RESUMO
Dengue continues to be the most important viral vector-borne disease in the world, particularly in Asia and Latin America, and is significantly affected by climate variability. The influence of climate in an endemic region of Colombia, from 2010 to 2011, was assessed. Epidemiological surveillance data [weekly cases] were collected, and incidence rates were calculated. Poisson regression models were used to assess the influence of the macroclimatic variable ONI [Oscillation Nino Index] and the microclimatic variable pluviometry [mm of rain for Risaralda] on the dengue incidence rate, adjusting by year and week. During the study period, 13,650 cases were reported. In 2010, the rates ranged from 8.6 cases/100,000 pop. up to a peak of 75.3 cases/100,000 pop. for a cumulative rate of 456.2 cases/100,000 pop. in that week. The climate variability in 2010 was higher [ONI 1.6, El Nino to -1.5, La Nina] than in 2011 [ONI -1.4, La Nina to -0.2, Neutral]. The mean pluviometry was 248.45 mm [min 135.9-max 432.84]. During El Nino, cases were significantly higher [mean 433.81] than during the climate neutral period [142.48] and during the La Nina [52.80] phases [ANOVA F = 66.59; p < 0.001]. Regression models showed that the ONI [coefficient 0.329; 95%CI 0.209-0.450] and pluviometry [coefficient 0.003; 95%CI 0.002-0.004] were highly significant independent variables associated with dengue incidence rate, after adjusting by year and week [p < 0.001, pseudo r[2] = 0.6913]. El Nino significantly affected the incidence of dengue in Risaralda. This association with climate change and variability should be considered in the elements influencing disease epidemiology. In addition, predictive models should be developed further with more available data from disease surveillance