Design and results of an intensive monitoring programme for avian influenza in meat-type turkey flocks during four epidemics in northern Italy.

Surveillance programmes for low pathogenicity (LPAI) and high pathogenicity avian influenza (HPAI) infections in poultry are compulsory for EU Member States; yet, these programmes have rarely been evaluated. In Italy, following a 1999 HPAI epidemic, control measures, including vaccination and monitoring, were implemented in the densely populated poultry area (DPPA) where all epidemics in Italy have been concentrated. We evaluated the monitoring system for its capacity to detect outbreaks rapidly in meat-type turkey flocks. The evaluation was performed in vaccination areas and high-risk areas in the DPPA, in 2000-2005, during which four epidemics occurred. Serum samples and cloacal swabs were taken from vaccinated birds and unvaccinated (sentinel) birds. We compared the detection rate of active, passive and targeted surveillance, by vaccination status, using multinomial logistic regression. A total of 13 275 samplings for serological testing and 4889 samplings for virological testing were performed; 6315 production cycles of different bird species were tested. The outbreaks detection rate in meat-type turkeys was 61% for active surveillance (n = 222/363 outbreaks), 32% for passive surveillance and 7% for targeted surveillance. The maximum likelihood predicted values for the detection rates differed by vaccination status: in unvaccinated flocks, it was 50% for active surveillance, 40% for passive surveillance and 10% for targeted surveillance, compared to respectively 79%, 17% and 4% for vaccinated flocks. Active surveillance seems to be most effective in detecting infection, especially when a vaccination programme is in place. This is the first evaluation of the effectiveness of different types of surveillance in monitoring LPAI infections in vaccinated poultry using field data.