Surveillance of low pathogenic avian influenza in layer chickens: risk factors, transmission and early detection

Low pathogenic avian influenza virus (LPAIv) of H5 and H7 subtypes are able to mutate to highly pathogenic avian influenza virus (HPAIv), which are lethal for most poultry species, can cause large epidemics and are a serious threat to public health. Thus, circulation of these LPAIv in poultry is undesirable. To date, these mutations cannot be reliably predicted and they can already take place within the first infected farm. Early detection and eradication of LPAI infected farms would minimise the probability of this mutations to happen. Because LPAI infections are associated with mild and unspecific clinical signs, detection of infected flocks is mainly based on serological surveillance.The goal of this thesis was to develop the framework of a serological surveillance programme for early detection of LPAI. Data of the current surveillance programmes in the European Union and the Netherlands were evaluated to quantify the risk of introduction of a LPAIv into the different poultry sectors. It was shown that among the different chicken production sectors in the Netherland, layer chicken flocks, in particular outdoor layers, have the highest risk of infection with LPAIv. In addition, chicken layers are often located in areas of high flock density, which increases the risk of spread of infection between flocks, and with that, the likelihood of mutation to HPAI. Thus, the need of implementing a science based early detection system was justified. However, early detection may lead to high (blood) sampling frequency, which increases the cost of the programme and affects animal welfare. Eggs could be used as sample alternative. Therefore, its use for surveillance was evaluated economically. It was shown that – under the current programme – surveillance based on egg sampling appears to be equally effective to and cheaper than blood sampling. To be able to calculate sample size and sampling frequency to design surveillance, the population dynamics of the infection have to be considered. The within and between farm transmission of LPAI was studied using transmission experiments and field data. This information together with information about the performance of the diagnostic test used for surveillance (testing either egg-yolk or sera samples) were combined to develop a mathematical model to design a risk based serological surveillance programme that aims at early detecting LPAI outbreaks. This model quantitatively identifies areas with flocks with high or low risk of transmission, where an specific sampling strategy (sample size and sampling frequency) is applied according to the area’s (flocks) risk of transmission. An output-based surveillance approach is recommended. This programme would target early detection in outdoor-layer flocks located in areas with high risk of transmission. In areas of low risk of transmission, surveillance could be performed aiming at substantiating absence of infection, for trading purposes. In this thesis, epidemiological and economical tools are provided that could aid the selection of an epidemiologically effective and economically efficient programme. A risk based programme, as suggested here, would be cheaper than the current surveillance programme and would contribute to minimise the risk of mutations of LPAIv to HPAIv.