CR 55:79-90 (2012)  -  DOI: https://doi.org/10.3354/cr01120

ABSTRACT: A retrospective ecological study was conducted using time-series and wavelet analyses to evaluate the impact of weather variables and climatic indicators on the incidence of leptospirosis in Jamaica between 1992 and 2007. Disease incidence was statistically linked to heavier rainfall and declining temperatures, with reported cases of leptospirosis peaking late in the year following a peak in rainfall. There was also some indication that disease spikes may be linked to the El Niño phenomenon. The climatic associations were used as the basis for the creation of statistical models for predicting disease incidence late in the year around the time it peaks. The models showed reasonable skill, explaining up to 72% of the variability in the disease record. The data also showed 1, 2, and 4 yr periodicities in the wavelet coherency spectrum. The results are significant for surveillance and prediction of the disease.

KEY WORDS: Climate · Disease · Caribbean · Temperature · Rainfall · ENSO