ON MAY 10, the prime minister announced a limited relaxation of lockdown restrictions.

The next day he told parliament the new rules were “very good advice for the entire population of the United Kingdom”.

He was reacting to united opposition in the devolved administrations of Scotland, Wales and Northern Ireland to what was being proposed. The Isle of Wight wasn’t mentioned.

Perhaps it should have been, because confidence which underpins advice on which people’s lives may depend is supposed to be based on data which is both reliable and relevant to those receiving it.

Two of the most important coronavirus risk factors are prevalence of the disease (the number of active cases) and the extent to which it is under control (the “R number”, or rate at which infected people are passing it on).

Accurate calculations for both of these risks require mass testing, which was paused very early on in Britain. This is now acknowledged by leading scientists as a mistake.

But in truth it was inevitable, because the infrastructure for mass testing simply wasn’t in place when it was needed.

If it hadn’t stopped, it would have been overwhelmed.

So instead, the progression of the pandemic has had to be extrapolated from hospital admissions, which generally lag infections by around a fortnight. And the number of infections can best be approximated by looking at the fatality rate.

On the Island, at the end of last week, we had a total of 181 confirmed cases and 56 deaths, giving a crude fatality rate of 31 per cent.

This is an obvious nonsense, given a worldwide rate in low single digits, and it implies an actual infection rate many times higher than 181.

The government’s statisticians do their best with the data available, and have come up with an “R rate” for the South East of England of 0.71.

But stats for the South East are of no more use to the Island than those produced for regional house prices or jobless figures. Reliance on them is dangerous guesswork.

Some useful data is, though, available from NHS England’s daily spreadsheets for hospital deaths, adjusted retrospectively to reflect the date of death rather than the date of reporting. They have shown for some time that the national peak of deaths was reached on April 8.

But the data is also broken down into health trusts. The Island’s hospital fatalities are: week ending 25/3, 2; w/e 1 /4, 1; w/e 8/4, 5; w/e 15/4, 5; w/e 22/4, 6; w/e 29/4, 7; w/e 6/5, 4; w/e 13/5, 2; w/e 20/5, 2.

This suggests, albeit in a small sample, a substantially later peak than nationally, and a slow decline from that peak. Add to that the alarming recent rise in Island care-home deaths — from three to 18 in the fortnight ending May 8 — and a definite Island time-lag is indicated.

Yet the advice is blanket. Open the schools, enjoy the countryside, summer’s here.

I wouldn’t be so sure. There’s a dangerous dearth of data, competing ideological doctrines, and an à la carte science which can be tailored to what you want to believe.

The fat lady has barely begun to clear her throat, and an abundance of caution is still needed.

No one has the right to squander what we’ve achieved with the lockdown.