Physical distancing measures may need to be in place intermittently until 2022, scientists have warned in an analysis that suggests there could be resurgences of Covid-19 for years to come.

The paper, published in the journal Science, concludes that a one-time lockdown will not be sufficient to bring the pandemic under control and that secondary peaks could be larger than the current one without continued restrictions.

One scenario predicted a resurgence could occur as far in the future as 2025 in the absence of a vaccine or effective treatment.

Marc Lipsitch, a professor of epidemiology at Harvard and co-author of the study, said: “Infections spread when there are two things: infected people and susceptible people. Unless there is some enormously larger amount of herd immunity than we’re aware of … the majority of the population is still susceptible.

Predicting the end of the pandemic in the summer [of 2020] is not consistent with what we know about the spread of infections.”

In its daily briefings, the UK government has not outlined plans beyond the current restrictions, but the latest paper adds to a building scientific consensus that physical distancing may be required for considerably longer in order to keep case numbers within hospitals’ critical care capacity.

Papers released by the government’s scientific advisory group for emergencies (Sage) in March suggested that the UK would need to alternate between periods of more and less strict physical distancing measures for a year to have a plausible chance of keeping the number of critical care cases within capacity.

The prospect of intermittent distancing raises difficult questions about what guidance will be given to high-risk groups, including over-70s and those with compromised immune systems.

It may be possible to relax distancing measures periodically while maintaining cases within a volume that health services can cope with, but the grave health risks of infection to some people will remain the same until a vaccine or highly effective treatments are available.

New treatments, a vaccine, or increasing critical care capacity could alleviate the need for stringent physical distancing, according to the paper in Science. “But in the absence of these, surveillance and intermittent distancing may need to be maintained into 2022,” the authors conclude.

The overall numbers of cases in the next five years, and the level of distancing required, were found to depend crucially on the overall current levels of infection and whether all those who are infected gain immunity and, if so, for how long. The authors cautioned that these are big unknowns and that a precise prediction of the long-term dynamics is not possible.

If immunity is permanent, Covid-19 could disappear for five or more years after the first outbreak, the paper suggests. If people have immunity for about a year, as is seen for some other circulating coronaviruses, an annual outbreak cycle would be the most likely outcome.

Asked to speculate on which of these scenarios was more likely, Lipsitch said: “Reasonable guesses are that there might be partial protection for close to a year. On the long end, it might be several years of good protection. It’s really speculative at this point.”

Under all scenarios considered, however, the models found that a one-time lockdown would result in a resurgence after restrictions are lifted.

Serological surveys, assessing the proportion of the population carrying protective antibodies, will be crucial to establish whether people have long-lasting immunity.

Other teams have found evidence that the immune response varies across people, with those who only have mild or no symptoms showing a far weaker response.

Ms. Marion Koopmans, head of virology at the Erasmus University Medical Centre in Rotterdam, whose team is studying the antibody response of those infected, said complete and permanent protection would be unusual for a respiratory virus.

“What you would expect to see – hope to see – is that people who have had it once … the disease would get milder,” she said before the latest paper was released.

Mark Woolhouse, a professor of infectious disease epidemiology at Edinburgh University, said: “This is an excellent study that uses mathematical models to explore the dynamics of Covid-19 over a period of several years, in contrast to previously published studies that have focused on the coming weeks or months.

“It is important to recognise that it is a model; it is consistent with current data but is nonetheless based on a series of assumptions – for example about acquired immunity – that are yet to be confirmed. The study should therefore be regarded as suggesting possible scenarios rather than making firm predictions.”

The Guardian