Assessment of seasonality of influenza in swine using field submissions to a diagnostic laboratory in Ontario between 2007 and 2012

Abstract

Seasonality of any infectious disease is important for its control and monitoring. While influenza seasonality in people has been evaluated extensively, this question has not been studied well in swine populations. The goal of this study was to investigate seasonality of influenza in swine, using diagnostic submissions to a diagnostic laboratory. Two thousand seven hundred and eleven virological tests within 685 submissions and 5471 serological tests within 193 submissions in Ontario swine between 2007 and 2012 were included in the study and converted to total monthly number of virological and serological submissions, and the number of positive submissions. Data were analyzed by time-series decomposition, fixed-effect Poisson, random-effect Poisson regression with month as uncorrelated and correlated random effects. All approaches identified seasonality in virological submissions (P < 0·02) with peak in January and April, and a trough in July, but were not able to detect seasonality of influenza-positive virological submissions (P > 0·13). Seasonality of positive serological submissions was identified only if independence between months was assumed (P < 0·03). Almost 50% of serological submissions had evidence of exposure to H3N2 and H1N1. Thus, this study identified evidence of seasonality in influenza-like disease in swine herds, but not in circulation of influenza virus. Evidence of seasonality in exposure to influenza was dependent on assumptions of between-month correlation. High exposure to H3N2 and H1N1 subtypes warrants more detailed investigation of within-herd influenza virus circulation. The study provides initial insight into seasonality of influenza in swine and should be followed with herd-level studies.