EPidemiology of Avian Influenza in Africa (EPIAAF) Survey : Final report. LoA PR 37212 between FAO and CIRAD

The objective of the EPIAAF (Epidemiology of Avian Influenza in AFrica) survey was to better understand the epidemiology of HPAI in Africa by assessing risk factors linked to the introduction, diffusion and persistence of HPAI outbreak foci in seven infected countries (Burkina Faso, Cameroon, Egypt, Ivory Coast, Niger, Nigeria and Sudan). CIRAD was contracted by FAO through a LoA signed in November 2007 to conduct the survey which was implemented in collaboration with four European research institutes (FLI, IZSVe, RVC and ULB), eight national survey consultants hired by FAO (two in Sudan and one in all other countries) and the veterinary services and veterinary diagnostic laboratories of each country. The EPIAAF survey protocol is fully described in the report entitled "Report of the protocol for epidemiological surveys LoA PR 37212 between FAO and CIRAD" published in December 2007. Shortly, descriptive studies were conducted to collect standardised information on HPAI outbreaks and analytic studies were conducted to compare risk factor information for outbreak sites and non infected sites. Activities implemented during the survey are partly described in the "EPIAAF Midterm report LoA PR 37212 between FAO and CIRAD" published in May 2008 and partly described in the present report. Briefly, field investigation missions were successfully conducted in a total of 43 sites (22 case sites, 21 control sites) located in the seven countries. They enabled to fill in 55 standardised questionnaires and to collect duplicate samples (tracheal swabs, cloacal swabs and sera) from 3 672 birds. Laboratory analysis including, RRT-PCR for the detection of AI and APMV1 virus, ELISA and/or HI tests for the detection of antibodies against type A influenza, H5 AI strains, H7 AI strains, and APMV1 virus, were performed by the FAO reference laboratory in Padova. All data collected (questionnaire data, data on risk factor at the national level collected by the NSCs, laboratory results) were stored in a specific database (EPIAAF database) that was created by FLI and distributed to representatives of all countries during the EPIAAF restitution workshop held in Yaoundé at the end of September 2008. Data analyses were performed using data entered in this EPIAAF database and were conducted by CIRAD, RVC and ULB. The main findings confirmed or mellowed some putative risk factors for H5N1 HPAI and brought light on possible new risks that need further investigations. With regards to introduction of H5N1 HPAI - AI viruses are commonly circulating in poultry but the proportion of seropositive birds (12% when considering only non-vaccinated birds) seems insufficient to confer satisfactory flock immunity for protection against HPAI; - Field and statistical evidences of a possible role of wild birds in H5N1 HPAI introduction were found but were not conclusive; - The possible role played by trade of poultry and poultry products was strongly emphasized by field findings although no statistical association with infection could be found, most probably because of a lack of statistical power and the difficulty to capture the diversity and complexity of trade relationships through this type of survey. With regards to the characteristics of H5N1 HPAI outbreaks - H5N1 HPAI may have been present on the African continent as early as October 2005 even thought the first outbreak was notified only in February 2006 ; - In 2006 and for the sites investigated in this survey, the time interval between detection and reporting of signs was not too long (4 days in average) and was even shorter between reporting of signs and investigation by veterinary authorities (1 day in average). On the opposite, EPIAAF survey Final report 4 the time interval between detection of signs and confirmation of the outbreak by the reference laboratories was very long (56 days in average); - The morbidity and mortality varied significantly depending on the species although the case fatality proportion was s