A statistical model that identifies regions across Asia with high risk for infection with avian influenza H7N9 is reported in Nature Communications this week. The model may help current efforts for surveillance and control of this important disease.
Two epidemic waves of bird flu of the H7N9 type have so far affected China and most human cases are thought to be the result of exposure to infected birds in live-poultry markets. The potential for a new H7N9 epidemic is an important public health issue. Active surveillance of markets is hampered by the mild effects of H7N9 infection in poultry and by the limited resources available to veterinary services across Asia.
Marius Gilbert and colleagues compiled an extensive census of such markets in China and the most up-to-date records on confirmed cases in humans and poultry. They then used these data, together with a variety of geographical and environmental parameters, to develop a thorough statistical model that accurately predicts the risk of H7N9 infection in live-poultry markets in China. Extrapolation of the model across Asia indicated that the areas with highest risk include certain urban areas in China where the disease has not yet been detected, an extensive area in Bengal, the Vietnam river deltas, and areas of Indonesia and Philippines.
The present work goes beyond previous studies by using more comprehensive datasets and statistical analyses, and should facilitate the prioritisation of surveillance systems in the region.