%0 Journal Article
%4 sid.inpe.br/mtc-m21b/2015/01.14.16.10
%2 sid.inpe.br/mtc-m21b/2015/01.14.16.10.35
%@doi 10.1016/S1473-3099(14)71031-X
%@issn 0140-6736
%T Interpretation of probabilistic forecasts of epidemics
%D 2015
%9 journal article
%A Lowe, Rachel,
%A Carvalho, Marilia Sá,
%A Coelho, Caio Augusto dos Santos,
%A Barcellos, Christovam,
%A Bailey, Trevor C.,
%A Stephenson, David B.,
%A Rodó, Xavier,
%@affiliation Institut Català de Ciéncies del Clima
%@affiliation Fundação Oswaldo Cruz (FIOCRUZ)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Fundação Oswaldo Cruz (FIOCRUZ)
%@affiliation Institució Catalana de Recerca i Estudis Avançats
%@affiliation
%@affiliation Institut Català de Ciéncies del Clima
%@electronicmailaddress racheol.lowe@ic3.cat
%@electronicmailaddress
%@electronicmailaddress caio.coelho@cptec.inpe.br
%B The Lancet
%V 15
%N 1
%P 20
%K Forecasts.
%X We are grateful to Eduardo Massad and colleagues1 for discussing the results of our study2 that addressed the potential for a dengue epidemic during the World Cup in Brazil. We believe our results are not comparable to point estimates obtained using deterministic models; however, we welcome the opportunity to discuss and clarify the interpretation of probabilistic forecasts of dengue risk. The approaches (including aim, methodological framework, data, and population) that we used2 differ in several ways from those used by Massad and colleagues3 to estimate the risk of acquiring dengue fever during the 2014 FIFA World Cup in Brazil. Massad and colleagues used a mathematical modelling approach, which was based on weekly notified cases in previous weeks, to estimate the number of cases of dengue in foreign visitors to Brazil. We used a spatiotemporal statistical model, driven by climate information and dengue incidence 4 months previously, to predict the probability of exceeding given thresholds for the whole Brazilian population.
%@language en
%3 Interpretation of probabilistic.pdf