Geographical distribution of the association between El Nino South Oscillation and dengue fever in the Americas: a continental analysis using geographical information system-based techniques

El Nino South Oscillation (ENSO) is one climatic phenomenon related to the inter-annual variability of global meteorological patterns influencing sea surface temperature and rainfall variability. It influences human health indirectly through extreme temperature and moisture conditions that may accelerate the spread of some vector-borne viral diseases, like dengue fever (DF). This work examines the spatial distribution of association between ENSO and DF in the countries of the Americas during 1995-2004, which includes the 1997-1998 El Nino, one of the most important climatic events of 20th century. Data regarding the South Oscillation index (SOI), indicating El Nino-La Nina activity, were obtained from Australian Bureau of Meteorology. The annual DF incidence (AI(y)) by country was computed using Pan-American Health Association data. SOI and AI(y) values were standardised as deviations from the mean and plotted in bars-line graphics. The regression coefficient values between SOI and AIy (r(SOI, AI)) were calculated and spatially interpolated by an inverse distance weighted algorithm. The results indicate that among the five years registering high number of cases (1998, 2002, 2001, 2003 and 1997), four had El Nino activity. In the southern hemisphere, the annual spatial weighted mean centre of epidemics moved southward, from 6 degrees 31'S in 1995 to 21 degrees 12' S in 1999 and the r(SOI, AI) values were negative in Cuba, Belize, Guyana and Costa Rica, indicating a synchrony between higher DF incidence rates and a higher El Nino activity. The r(SOI, AI) map allows visualisation of a graded surface with higher values of ENSO-DF associations for Mexico, Central America, northern Caribbean islands and the extreme north-northwest of South America.