Show Hide image

The fight against Covid-19: what does winning look like when the enemy is a virus?

Until the worldwide search for a vaccine succeeds, winning is likely to mean making sure we lose as little as possible.

The tabloid talk at the start of the Covid-19 outbreak was of the need for “Blitz spirit”. It was probably inappropriate; it was certainly premature. The point about the Blitz wasn’t only the justified fear of harm, or the deprivation of goods and liberties. The point was also that it went on for eight months, and the war for six years. The UK has been in Covid-19 lockdown for precisely two weeks. 

You measure your “Blitz spirit” – whatever that might be – at the end, not the beginning. And the story of Covid-19 is still only beginning.

How are we faring in the fight? The data we have provides an imperfect picture. At the time of writing, there have been 51,608 confirmed cases across the UK. The number of confirmed cases, however, depends largely on the extent to which a country tests for the disease. The UK has largely limited testing to patients who are seriously ill in hospital. Since we generally haven't tested people who are only mildly symptomatic or don't go to hospital, we don't know how many have had the disease. 

The best we can do is estimate, based on a more robust figure: the number of deaths. If Covid-19 kills around one in 100 people, as some modelling suggests, and the UK has recorded 5,373 deaths, that implies around 537,300 have actually had the disease. The number could be much higher, or a fair bit lower, since we don't know exactly what proportion of people Covid-19 kills. But 51,608 is surely a massive underestimate – and a figure that will surely rise sharply as more testing takes place, without telling us anything about the actual spread of the disease. 

One can look at it this way: Germany has twice as many confirmed cases as the UK (101,178), but has seen less than a third as many deaths (1,612). German doctors are good, but they aren't superhuman. The figures chiefly reflect the fact Germany is testing far more people, and so detecting far more cases. 

The figures for Covid-19-related deaths also have their issues. The daily totals that are being reported by the British government include only those who have died in hospital and tested positive. Frail or elderly people who die in care homes, for example – who are disproportionately susceptible to the virus – aren't included. 

How much of a problem is this? The Office for National Statistics, which includes all deaths where Covid-19 is mentioned on the death certificate, counted 210 up to 20 March. The “official” figures counted 170. The ONS figures, however, take time to come through and are only retrospectively useful.

Then there is the problem of “lag”. The official figures are based on the day deaths are reported, not the day they took place, and there can be a gap of days – or sometimes even weeks – between those two events. Death certificates have to be completed and family members informed. Reporting can be delayed by something as simple as staffing levels in particular departments on particular days.

The death figures aren’t useless, though. For one thing we can average out the daily death total – looking at the average for the past three days or week, for example – to smooth out day-to-day volatility. More importantly, as long as the methodology remains consistent – even if consistently flawed – we can still draw some comfort from positive trends. The death totals may undercount and be days out of date; as long as they do so in a consistent manner, they can still tell us something about the direction of travel.

Let’s compare the number of Covid-19-related deaths in the UK and other hard-hit Western nations. We’ll start at the point where each country recorded its 50th cumulative death and instead of using the deaths announced on a particular day, use the average announced on each of the past three days:

The interesting comparison here is between the UK and Italy – two countries with similar population sizes. Italy passed 50 deaths 16 days before the UK, and instituted a national lockdown 14 days earlier – although regional lockdowns were in place in Italy earlier than that. Italy has appeared to be seeing a stable reduction in average daily deaths over the last week; if the UK continues to track the same course, 16 days behind, that plateau would be nine or so days away.

All this is highly speculative, but it does chime with last weekend’s prediction by Neil Ferguson  – the Imperial College epidemiologist credited with changing the government’s thinking on the need for a wide-reaching lockdown – that UK deaths would rise for the next seven to ten days, level off, then fall. He suggested the total number of deaths in the UK could potentially be kept at between 7,000 and 20,000. In many ways that would be an astonishing testament to the power of the lockdown. Seasonal flu, for example, was calculated by Public Health England to have been associated with 26,408 deaths in 2017/18, and 28,330 in 2014/15. Covid-19 is a far more deadly and apparently more easily transmissible disease. Professor Ferguson’s team, remember, predicted up to 510,000 Britons could die if nothing were done to stop the disease’s spread.

Let’s assume that cases in the UK plateau and fall relatively quickly. The question is: what happens then? 

The section no one quotes from the original Imperial College paper – the one that set us on this path  – suggested a “minimum policy” of social distancing, home isolation of cases and school and university closure would “need to be maintained until large stocks of vaccine are available to immunise the population – which could be 18 months or more.” The researchers suggested such measures would need to be in place at least two-thirds of the time, with only temporary relaxations until Covid-19 cases started to rise again. 

Coronavirus hasn’t suddenly become less transmissible, and if cases are falling it is because of the countermeasures that have been taken – remove them and what is to stop a second wave of cases spreading just as aggressively as the first? We already know how transformative those countermeasures have been, but we don’t yet know how destructive. Last week we learned 950,000 people had applied for Universal Credit in a fortnight – roughly ten times the number who do so in a normal two-week period. Other data will emerge in time, but the issues caused by a decade of austerity are hardly likely to have been improved by such a profound economic shock. Rising homelessness; cuts to social care; over-stretched mental health services; the rise of zero-hour contracts. How will the lockdown have affected those? 

If a complete, instant return to our previous life isn’t possible, what middle ground might there be? Some look to South Korea and Japan, which have been remarkably effective in minimising their outbreaks without the need for full lockdowns.

South Korea has used widespread testing – it was stockpiling kits as early as January and introduced drive-through test centres – as well as contact tracing, quarantine and invasive tracking technology to curb the virus. Even there, however, life isn’t quite as it was: data from Citymapper suggests the use of the app to navigate Seoul was less than 50 per cent of normal levels on all but three days in March. Japan, meanwhile, has seen a spike in cases and Prime Minister Shinzo Abe is reportedly on the brink of declaring a state of emergency. Many shops, bars and restaurants are reported to have pre-emptively closed, and figures released by Google using anonymised mobile phone location data showed visits to retail and recreation sites were down 26 per cent at the end of March.

A phased easing of the lockdown is entirely plausible – with greater restrictions remaining for those deemed most vulnerable. So is some form of passporting for those who have had – and recovered from – the virus. Better testing and contact tracking will be part of the long-term strategy, too. Meanwhile, the worldwide search for a vaccine will continue. Until that search succeeds, “winning” is likely to mean making sure we lose as little as we can.

Until the worldwide search for a vaccine succeeds, winning is likely to mean making sure we lose as little as we can.

David Ottewell is head of data journalism at New Statesman Media Group.