There are now signs of the growth in UK Covid cases slowing, according to Professor Neil Ferguson, who is emerging as the de facto chief strategist of the government response to the crisis. No government data has been issued to confirm this trend but Ferguson has access to other real-time data through SAGE, the medical emergency committee. He was on BBC Radio 4's Today programme and said:-
'In the UK we can see some early signs of slowing in some indicators. Less so deaths, because deaths are lagged by a long time from when measures come in force. But if we look at the numbers of new hospital admissions, that does appear to be slowing down a little bit now. It has not yet plateaued, so still the numbers can be increasing each day, but the rate of that increase has slowed.'
This is quite significant and raises the prospect of Britain being further along the epidemic curve than some had feared. Perhaps seeing a peak earlier than expected (given the ten-day time lag between infections and death). It also squares with what James Forsyth revealed over the weekend: that the government thinks the virus deaths will peak in early April, rather than May as they had earlier feared. A study last week from Prof Tom Pike of Imperial College's engineering department also pointed to a mid-April peak, with deaths dropping off rapidly after that peak is hit.
It’s unclear whether an earlier peak would be bad news (that we have not managed to ‘flatten the curve’ as much as hoped, meaning more deaths but a shorter duration) or whether the main Imperial College London group, which Ferguson chairs, is revising its assumptions on the potency of the virus.
Ferguson's interviews can disclose pieces of hugely important information not mentioned in his own reports let alone government statements. His original study, which promoted the government lockdown, said that this virus could claim 500,000 British lives: more than every other disease in Britain put together. But this was the top of a large range, the lower point (assuming various lockdown measures were followed) being 7,000. When giving evidence in parliament last week, Prof. Ferguson said social distancing means he’s now looking at fewer than 20,000 Covid deaths, two-thirds of which would have occurred anyway. This points to his rock-bottom figure of about 7,000, a death toll far other conditions like seasonal flu (which typically takes 17,000 British lives a year).
Ferguson’s model, like all models, seeks to estimate a great many things: how contagious the virus is and, crucially for NHS planning, how many people it is likely to hospitalise. And kill. For example, Ferguson's study assumes a death rate of 0.9 per cent, nine times higher than seasonal flu. The Italian experience indicates a 10 per cent death rate but a model last week from two professors at Stanford university posit a 0.01 per cent death rate. So highest figure in this range is, literally, a thousand times more than the lowest figure.
If UK Covid hospitalisations rates are slowing, as Prof Ferguson says, that is not a model: it's hard data. it could be a blip – or a sign that we might be a bit closer to turning the corner.
Prof Ferguson also told BBC Radio4 that between three per cent and five per cent of Londoners could be infected: given the capital's 8.9 million population that's almost 450,000 people. This is consistent with estimates from the Chief Medical Officer: that the real number of infections is 10 to 20 times higher than the figure for those who have tested positive. Why the gap? Because Covid can be fought off by most people's immune systems with mild or no symptoms: when that happens, antibodies are created.
If those with antibodies are immune to reinfection (which the government believes likely, but it's still unproven) then this suggests a small army of people in the capital ready to return to work. But to identify them, you'd need an antibody test. Prof Ferguson told the BBC that such a test is 'days rather than weeks' away. For what it's worth, that's not my understanding: my sources say the UK test is two or three weeks away, due to doubts about whether it works.