An additional 30,000 coronavirus deaths are expected to occur in England alone by the end of June under a best-case scenario, according to a report presented by the government's scientific pandemic influenza group on modelling, operational sub-group (SPI-M-O).
On 22 February, Boris Johnson will outline the "cautious" four-part plan to lift England's lockdown, starting with all schools reopening on 8 March.
This lifting comes as cases continue to fall, with 9,834 people testing positive across the UK on 21 February – a 16.2% reduction on the previous seven days.
Emerging evidence also suggests coronavirus vaccines ward off hospitalisations, helping to prevent the NHS from being overwhelmed.
Watch: Scientists predicted at least 30,000 more coronavirus deaths 'under most optimistic model' for lifting lockdown
The SPI-M-O report implies, however, many more lives are set to be lost in the coming months.
Based on "very uncertain modelled assumptions" by the University of Warwick and Imperial College London, the report estimates at least a further 30,000 coronavirus-related deaths will occur from 12 February to 30 June "under the most optimistic" scenario.
This is assuming "baseline measures once restrictions are lifted significantly reduce transmission compared to pre-COVID-19 behaviours", like continued social distancing to some extent and the wearing of face coverings.
In a worst-case scenario, 81,200 fatalities may occur, according to the report, which informs the advice the scientific advisory group for emergencies (Sage) gives the government.
Given the substantial uncertainty surrounding these models, the SPI-M-O scientists have reiterated lockdown lifting should be driven by data not dates.
England's chief medical officer Professor Chris Whitty has previously pointed out up to 20,000 people die from flu in "high years", with it being up to politicians to decide an "acceptable" level of death that justifies reopening society.
Models are always uncertain, with scientists having to make assumptions on vaccine effectiveness in the real world, outside of a clinical trial.
Other assumptions include the vaccine programme's coverage and roll-out speed going forward, as well as the extent to which societal measures will continue to dampen transmission once lockdown is eased.
The uncertainty becomes more pronounced the further scientists look into the future, with the SPI-M-O team being "much more confident in the modelled outcomes to the end of May than beyond".
To learn more, the Imperial and Warwick scientists created models based on "rates of transmission under different autumn tiers" that were previously in place in England.
They also adjusted for the emergence of the so-called Kent variant towards the end of 2020, which is thought to spread up to 70% more easily than earlier versions of the virus.
The models are also based on "allowing indoor contact with one guest per household per day", a "major change" from the tier system that "has the potential of a resulting large increase in transmission".
Watch: Can you catch coronavirus twice?
The results of the Imperial model suggest an additional 33,200 to 81,200 coronavirus deaths may occur between 12 February and 30 June "as restrictions are relaxed".
This is "in line with" the Warwick team's findings.
Case numbers are expected to rise as lockdown lifts and a greater degree of social mixing is permitted.
Nevertheless, the SPI-M-O scientists have stressed resurgences could be "well below those seen in January 2021".
This relies on gradual restriction easing, vaccines being "highly effective and rapidly rolled out" and "longer term baseline measures achieving a significant reduction in transmission".
The SPI-M-O report comes as Public Health England scientists reveal the first Pfizer-BioNTech coronavirus vaccine dose was found to be 72% effective 21 days later, rising to 86% one week after the second jab among healthcare workers in a real-world study.
No vaccine is completely effective, with older people in particular tending to have a more muted immune response.
The threshold required to achieve herd immunity with the Kent variant is unclear, but "likely to be high". Experts have previously pinned the coronavirus' herd immunity level at around 80%.
The vaccines being rolled out across the UK are only approved for use in adults, who make up around 79% of England's population.
Even if 79% of adults are vaccinated, that only translates to 62% of the total population, pointed out the SPI-M-O scientists.
"As a result, herd immunity is not likely to be reached in these scenarios without a further resurgence of transmission," they said.
Maintaining "baseline measures" once restrictions are lifted is "almost certain to save many lives and minimise the threat to hospital capacity".
These measures could be "voluntary", or include an effective test, trace and isolate system.
The SPI-M-O scientists have warned hospital occupancy is still "very high and will remain so for a significant length of time".
On 17 February, the latest data available, 1,397 people were admitted to hospital with the coronavirus in the UK.
Speaking of England's lockdown easing, Sir Patrick Vallance – the UK's chief scientific adviser and Sage member – said: "We're not starting from a good position", but rather the situation is "getting better".
The Warwick and Imperial models do not account for variations in vaccine coverage between communities, with evidence suggesting fewer people from minority ethnic groups are taking up the jab.
They also assume the effectiveness of the vaccines against new coronavirus variants like that from South Africa "remains high", despite studies throwing up mixed results.
Waning immunity over time, which occurs with other viruses of the coronavirus class, was also not accounted for.
Watch: What is long COVID?