Scientists have warned the NHS is still under significant pressure as the number of people infected with coronavirus in England is at the highest level recorded since the end of the first wave in May.

According to the influential React study, one in 64 people were infected in England between January 6 and 22. Infections have flattened but are at the highest level recorded in the series of reports.

There is an indication that infections were declining at the end of the reporting period, but scientists were quick to point out the drop is not as steep as seen at the end of the first wave.

The study, which has been conducted every month since May, excluding December, is the latest indicator that while the third national lockdown appears to be getting the spread of the disease under control – a point scientists admit was not a certainty – the UK still has a long way to go.

According to official government figures, daily positive Covid-19 tests reached an all-time high on December 29 with 81,522 – spiking again on January 4 to 76,090,  The most recent complete set of data available lists 35,106 cases on January 20. It’s a significant fall – around 50% down since the late December peak – but still stubbornly high. There was not a day in May when daily cases topped 5,000.

The “snapshot” React study, conducted by Imperial College London and Ipsos Mori, monitors the “prevalence” of the disease – how many people actually have it – by taking nose and throat swabs from 167,000 randomly selected people. It is a different measure to the infection data reported daily by the government, which records new cases.

The main findings from the eighth round of the study suggest national prevalence was 1.57%, or 157 per 10,000 people infected. Prevalence increased nationally in all adult age groups and was highest in 18 to 24-year-olds at 2.44%, while prevalence in the over-65s is 0.93% – almost one in 100.

It also estimated the national R number – a measure of coronavirus transmission – to be at 0.98 with a range of 0.92 to 1.04. Any number above one indicates the number of infections is increasing. Regional prevalence was highest in London at 2.83%, while in the South West it was 0.87%.

It comes after prime minister Boris Johnson announced lockdown measures will remain in place in England until at least March 8, and said travellers returning to the UK from “red list” countries will be sent to quarantine hotels for 10 days.

Professor Paul Elliott, director of the programme at Imperial, said: “The number of people infected with the virus is at the highest level that we’ve recorded since we began testing last May.

“We’re not seeing the sharp drop in infections that happened under the first lockdown and if infections aren’t brought down significantly, hospitals won’t be able to cope with the number of people that need critical care.

“We all need to stay at home wherever possible and help bring the virus under control and protect our already over-stretched health system.”

Prof Elliott said: “I think the suggestion now that there is a decline happening, particularly in some regions may reflect now that the restrictions through lockdown are beginning to have some effect on the prevalence.”

He added: “We really do need to keep an eye on what’s happening because even though we’re seeing this suggestion of a downtick now, which is really good news, it’s by no means as fast as we saw in lockdown one.

“When we were first in the field in May, there was a rapid decline in prevalence.”

The researchers say it will be some time before the impact of vaccination reduces pressures on hospitals.

Steven Riley, professor of infectious disease dynamics at Imperial College London, said the government would have to look at a number of different data sets – including information on hospital admissions and positive Covid-19 tests – in order to get the most reliable expectation of what will happen in the coming weeks.

He added: “We do know the R value at the end of the first lockdown and that was substantially lower than the R values that we’re measuring now, which is consistent with the other data.”

Prof Elliott said that the study’s mobility data – which looks at how much people are moving around – shows that there is more activity now than there was at the end of the first lockdown, but said the lockdown rules are not the same.

He added: “We are definitely heartened by the fact that we’re now seeing what looks like a decline in the last week of our survey.”

Prof Riley also commented on data from around Christmas, saying the concern was that social mixing would cause a spike in infections, but this was not the case.

He explained: “You would have expected a peak a week to 10 days after that increased mixing, and in our data we don’t see evidence of that and it’s certainly not in the ONS and some of the other data sources.

“I would note that we have seen substantial increases in infections in the older age groups who are most at risk.

“So, even though there was not an obvious spike in the absolute number, we are seeing considerably higher levels of infection in the most at risk older, so the intergenerational mixing during Christmas may have contributed to that.”