West Nile Virus infection in Northern Italy: Case-crossover study on the short-term effect of climatic parameters

Abstract

Background: Changes in climatic conditions are hypothesized to play a role in the increasing number of West Nile Virus (WNV) outbreaks observed in Europe in recent years.

Objectives: We aimed to investigate the association between WNV infection and climatic parameters recorded in the 8 weeks before the diagnosis in Northern Italy.

Methods: We collected epidemiological data about new infected cases for the period 2010-2015 from the European Center for Disease Control and Prevention (ECDC) and meteorological data from 25 stations throughout the study area. Analyses were performed using a conditional Poisson regression with a time-stratified case-crossover design, specifically modified to account for seasonal variations. Exposures included weekly average of maximum temperatures, weekly average of mean temperatures, weekly average of minimum temperatures and weekly total precipitation.

Results: We found an association between incidence of WNV infection and temperatures recorded 5-6 weeks before diagnosis (Incidence Rate Ratio (IRR) for 1 °C increase in maximum temperatures at lag 6: 1.11; 95% CI 1.01-1.20). Increased weekly total precipitation, recorded 1-4 weeks before diagnosis, were associated with higher incidence of WNV infection, particularly for precipitation recorded 2 weeks before diagnosis (IRR for 5 mm increase of cumulative precipitation at lag 2: 1.16; 95% CI 1.08-1.25).

Conclusions: Increased precipitation and temperatures might have a lagged direct effect on the incidence of WNV infection. Climatic parameters may be useful for detecting areas and periods of the year potentially characterized by a higher incidence of WNV infection.

Keywords: Case-crossover; Lag-distributed models; Precipitations; Temperatures; West Nile Virus.