Stronger research leads on AI transmission to humans

The research by the European Centre for Disease Prevention and Control was conducted in a village where there had been extensive poultry deaths attributed to AI, some lab-confirmed poultry infections, and one human death from H5N1.

Blood sampling of 351 people, carried out within a 1 km radius of the man's house, showed that although the participants had frequent direct contact with poultry, none of the participants had neutralising antibodies to H5N1. More than 70% of the households involved in the surveying reported poultry deaths in the weeks before the man's death.

The findings from this Cambodian study strongly suggest that asymptomatic and mild H5N1 virus infections did not occur in this human population and that H5N1 infection was difficult to acquire.

This presumption has been the basis of risk assessments carried out by the European Centre for Disease Prevention and Control in the past year, although the evidence base for this has been weak.

There had previously been no published population surveys with serological support and the little published data that there was (mostly concerning contacts of cases or occupational groups) had not specified where the laboratory testing had been undertaken, or the methods used.

The study's findings will need to be replicated elsewhere before conclusions can be drawn, but the public health advice issued by the ECDC is now better supported scientifically. The results also emphasise that the risks to humans from the many different avian influenzas, even highly pathogenic ones, must each be considered individually.