Government scientists are expecting Covid-19 hospital admissions in England to rise to between 1,000 and 2,000 a day — and deaths to 100 to 200 a day — a few weeks after almost all restrictions are removed in England on July 19.

But they emphasised in their latest documents that projections are more uncertain now than at any point in the pandemic and the outcome could be worse.

Sage, the government’s scientific advisory group, considered the results of three academic modelling teams — at Warwick university; Imperial College London; and the London School of Hygiene and Tropical Medicine — to produce a consensus view that was put to ministers late last week ahead of Monday’s unlocking statement.

All scenarios show what Sage documents call “a period of extremely high prevalence of infection lasting until at least the end of August”, as young people in particular resume social contacts.

Health secretary Sajid Javid had previously said that the number of new Covid cases, which has risen 15-fold since the beginning of May to more than 30,000 a day, was likely to surge to a level above 100,000 a day during August. The scientists point out, however, that the number of positive tests confirmed may be suppressed by limits to testing uptake and capacity.

Chart showing that a full reopening on July 19 could send daily hospital admissions back above 1,000

Although daily new cases are therefore expected to peak well above the 60,000 reached in early January, “the scale of the next wave in hospital admissions is highly uncertain”, said Sage’s Spi-M modelling group.

“While most modelled scenarios have peaks lower than in January 2021, a resurgence of this scale of hospitalisations [4,000 a day] cannot be ruled out,” it said. “It is almost certain that, with the Delta variant dominant, the peak in deaths will be well below the levels seen in January.”

The chance that infection by the Sars-Cov-2 virus will lead to hospitalisation with Covid-19 is about one-quarter of what it was before the UK began its vaccination campaign in December, while the risk of dying has fallen to about one-tenth of the previous level.

One of the biggest uncertainties facing scientists trying to predict the future of the outbreak in England is how quickly people will return to pre-pandemic behaviour after lockdown ends. Social mixing increased by less than some behavioural experts had expected after the last easing of restrictions on the government’s road map took effect on May 17, but “freedom day” on July 19 may have a bigger impact.

The Netherlands provides a warning of how fast Covid infections can spread when lockdown ends. Mark Rutte, the Dutch prime minister apologised on Monday for lifting most remaining restrictions too soon, after new cases rocketed 20-fold in less than two weeks, to reach 10,000 a day. In response, the country reimposed curbs on bars, restaurants and nightclubs on Friday.

Chart showing that the surge in cases in the Netherlands highlights the potential risk of a full reopening

The two other key variables are the effectiveness and uptake of vaccines. “Although the number of people who have been vaccinated is known with high accuracy, the number who have not been vaccinated is not,” Sage scientists pointed out. UK population levels are uncertain, with results of the 2021 census not yet fully processed.

Small changes in these assumptions, for example if vaccine uptake is 92 per cent rather than 96 per cent, can have a substantial impact on Covid case numbers.

Sage pointed out to ministers that, of the four main risks associated with high infection numbers, the modelling only attempts to predict one: increasing hospitalisations and deaths.

Quantifying the other three — long Covid, workforce absences and the risk of more dangerous new variants emerging — is even harder.

Chart showing that ahead of its reopening, the UK already has the highest Covid case levels of all major European countries

After the current wave peaks, probably in late August or September, many modelling scenarios show further Covid waves during the autumn and winter, though they all forecast hospitalisation rates below last winter’s. The modelled winter waves are larger if the summer wave is smaller, because fewer people would then have acquired immunity through natural infection as well as vaccination.

Looking at the most optimistic and pessimistic scenarios, the Imperial College modellers found that the number of additional Covid-19 deaths by June 2022 would be anywhere from 9,400 to 115,000.

“As we have seen at several points in the pandemic, the potential of factors such as hospitalisations and deaths to grow exponentially — meaning that they double after a certain number of days — means that such a new wave can quickly cause significant problems from a relatively low baseline,” said Thomas House, a statistician at the University of Manchester and Spi-M member who was not directly involved in any of the three modelling exercises.

“Therefore it is likely to be a good idea to have a plan in place for such an eventuality, while hoping that it does not come to pass,” House added.