Indian scientists estimate how bird flu could spread to humans

One of the worst concerns for epidemiologists is that the bird flu virus, which is spreading in poultry farms around the world, could spread to human hosts. Scientists have long warned that bird flu, often known as H5N1, could eventually spread dangerously from birds to people and cause a worldwide health emergency.

Now, Indian scientists have used an advanced artificial intelligence (AI)-based model to predict and map the potential spread of the H5N1 bird flu virus in human populations.

Ashoka University researchers Philip Cherian and Gautam Menon published a peer-reviewed study titled “Modeling potential zoonotic spillover events of H5N1 influenza” in BMC Public Health. The model simulates how the H5N1 virus might spread to humans and what early interventions might stop it.

“The risk of an H5N1 pandemic in humans is real, but we can hope to contain it through better surveillance and a more agile public-health response,” Professor Menon told the BBC.

In the study, the team used BharatSim – an agent-based simulation framework for infectious diseases – to describe the sequential stages of zoonotic spillover.

“We modeled the probability of initial spread events of H5N1 from birds to humans, followed by sustained human-to-human transmission. Our model describes the two-stage nature of the onset of the outbreak, showing how important epidemiological parameters controlling transmission can be used to calibrate data to the distribution of numbers of primary and secondary cases at early times,” the researchers wrote in the paper.

Avian flu – a type of influenza – has occasionally infected humans since originating in China in the late 1990s. It is widespread in South and Southeast Asia due to the world’s fastest growing poultry markets. The World Health Organization (WHO) reported 990 human H5N1 infections in 25 countries between 2003 and August 2025, resulting in 475 deaths, representing a 48% mortality rate.

Culling birds is effective in preventing spread into human populations, provided it is done before primary human infection occurs. The study said that the sooner the killing is done, the greater the chances of stopping the spread.

“Our simulations account for various interventions. These include (i) culling of all birds on the farm, (ii) isolating primary and secondary contacts once the threshold of cases is crossed, and (iii) a vaccination campaign where primary and secondary contacts are targeted. We find that culling of birds is effective provided no primary infection has occurred,” it read.

Experts underlined that control efforts are most effective in the early stages of the outbreak. “Once community transmission begins, drastic public-health measures such as lockdowns, mandatory masking and mass vaccination campaigns are the only options left. At this stage, the large number of cases ensure that stochastic effects should play a minor role and traditional compartmental models should provide appropriate guidance,” he said.

The study demonstrates how such models enable systematic real-time assessment of policy options that can limit disease spread as well as guide better knowledge of disease epidemiology for an emerging disease.

According to the Cleveland Clinic, symptoms of avian influenza include fever, cough, body aches, fatigue, muscle aches, sore throat and conjunctivitis.