At the recent press briefing, we were warned “we must learn to live” with this virus, with people continuing to die. So is there is an “acceptable” number of Covid-19 deaths? Prof Chris Whitty, the government’s chief medical adviser, highlighted that we live with seasonal flu, which kills thousands each year. Like deaths from air pollution, these deaths are estimated from statistical models and this can introduce volatility. The number of flu deaths in England over the winter of 2017-18 was estimated as 26,400 in 2019, but dropped to 22,000 by 2020. On average, over the five pre-pandemic years, there were about 10,000 seasonal flu deaths. Is that a reasonable standard for acceptability? For British road deaths, the worst year was during the blackout in 1941 when 9,000 died, almost matched by 8,000 deaths in 1966. That was clearly not considered acceptable, with every effort made to reduce mortality to current levels of fewer than 2,000 a year. One organisation has dared to define acceptable fatal risks. The Health and Safety Executive says life cannot be safe and an “acceptable” lethal public risk is one in a million a year. For the UK, that would mean about 67 deaths each year. However, that does not mean any risk above that level is unacceptable: there is a second, higher threshold of “unacceptable” risks to workers of one in 1,000 a year. That would correspond to about 67,000 UK deaths, exceeded in a year of pandemic. Within this very broad band, HSE says tolerable risks should be made “as low as reasonably practicable”, adopting measures to push down risks, but ensuring actions are commensurate with costs. With an infectious disease, risks are contagious. Measles offers one potential comparator; the UK achieved elimination but that turned out to be fragile and the country lost its measles-free status in 2019 . It is challenging to balance the harms and benefits of interventions unless communities are in a stable situation. Fortunately, vaccinations and improving treatments are moving us towards such managed stability. David Spiegelhalter is chair of the Winton Centre for Risk and Evidence Communication at Cambridge. Anthony Masters is statistical ambassador for the Royal Statistical Society