COVID VaccinesReturn to WorkMental Health

U.K. to use A.I. to spot dangerous side effects in the millions of COVID-19 vaccinations it will deliver

November 5, 2020, 4:15 PM UTC

The British government plans to use artificial intelligence software to spot any potential safety concerns with its planned mass vaccination program for COVID-19.

The U.K. government is expected to approve one or more coronavirus vaccines before the end of the year, and has told the government-run National Health Service to make preparations for a mass vaccination campaign “by Christmas,” according to news reports.

But vaccinating the entire British population of almost 68 million people against COVID-19 will pose a number of challenges—one of which will be monitoring those who receive the inoculation for any dangerous side effects.

Even though the vaccines will have undergone testing with thousands of volunteers before they are approved by the British medical regulator, it is possible that those trials will fail to pick up important safety concerns.

And the U.K. Medicines and Healthcare Products Regulatory Agency (MHRA), the country’s main medical regulator, is worried that its current process for reporting “adverse events” for medicines will be incapable of handling both the potential number of side effect reports—and too slow to pick up on any worrisome trends.

To automate the process, the MHRA has given a $1.96 million contract to the U.K. arm of Genpact, the U.S.- listed business process outsourcing firm, to design a machine learning system that can ingest reports of side effects and pick up on potential safety concerns.

The regulator told the Financial Times that, based on its previous experience with vaccination campaigns, it would expect that, over a six- to 12-month period, there might be 50,000 to 100,000 reports of suspicious side effects for every 100 million doses injected.

The possibility of using A.I. to monitor drug side effects has been the focus of academic research for several years, and the British regulator is not the only government considering the use of the technology in this way.

The U.S. Food and Drug Administration held a competition earlier this year for computer algorithms designed to identify trends in the massive volume of adverse event reports it processes for all drugs in the U.S. The competition gave out two prizes, one for the most innovative approach, which went to an A.I. system designed by two researchers with a small technology company called Enigma, and another for the system that would be most easy to integrate with the agency’s existing processes, which went to one of the agency’s own scientists.

It is not clear, however, whether or how quickly the FDA will implement either system or if it might use such a system to monitor U.S. COVID-19 vaccinations.