The verdict is in: Boris Johnson's handling of coronavirus has been disastrous. Britain's death toll remains one of the highest in the world, sick patients were discharged back into care homes at the height of the epidemic; and millions of pounds blown on useless PPE which can't be used by NHS staff. But while it's true that some things have gone wrong – as Boris himself has conceded – it's also true that many of those loudly condemning Boris for the way he has dealt with the pandemic have form in criticising Boris. So are they right? And is it fair to blame Britain's coronavirus blunders on the PM?
The UK is now fourth in the inglorious table of coronavirus deaths, with India sadly catching up fast. The United States, Brazil and Mexico have all fared worse, but then, of course, these countries are much larger. Deaths per head of population would be the more reasonable indicator, it would seem. On this measure, the UK is currently third, and the USA tenth (or respectively second and eighth, when very small countries such as Andorra and San Marino are reasonably excluded). Belgium ‘tops’ the larger nation table; Chile and Peru are, however, moving up through the top ten.
But even these population-adjusted figures are untrustworthy. There are major differences in how countries record Covid deaths – assumed diagnosis vs. proven, died with vs. died of, nursing home deaths included or excluded, home deaths counted or not, to name but four. PHE (Public Health England) counts deaths at any time after a Covid diagnosis, including deaths from unrelated causes. Germany and France include deaths in care homes; England only did so after 29 April. But Germany only counted care homes deaths when confirmed by a positive test result, while in Belgium, a clinician's suspicion of coronavirus was sufficient, likely seriously inflating their figures. Deaths at home are another cause of insecure statistics: France did not include these originally, but did from June.
Some countries deal especially in 'excess mortality': the total number of deaths over a fixed period compared to the average total over the last five years for that same period. The UK has this as its preferred index, though confusingly, the NHS, the Department of Health and Social Care, and the Office for National Statistics (ONS) for many months released separate daily figures for Covid-related deaths, each calculated in quite different ways. All have their value – but to different observers. The excess mortality rate – which will include individuals dying from Covid-19, as well as those dying because of cancelled operations, or scans, or through not coming to hospitals for fear of contracting Covid-19, over-burdening the NHS, etc., together with those non-Covid individuals from any other unusual transient factors – is of small passing interest to virologists. But most of these deaths are Covid-related deaths, even if the virus is only indirectly responsible (by precipitating our reaction) – and so these numbers are still interesting to health economists and planners.
But the point here is that if other countries use different measures, international data comparisons are rendered almost impossible. And if they use the same measure, as the ONS has lately done, producing an international comparison of excess mortality figures, then we are knowingly looking at numbers that include all deaths, Covid-caused and others. This immediately begs the question of time frame – is it valid to compare January to June in the UK with the same period in the USA? January to early June was (so far) the UK’s worst time: if (say) a whole 12 month period were studied, would other countries still look ‘better’? After all, the most up-to-date ONS figures (released on 4 August and relating to the week ending 24 July) show that for the sixth consecutive week the total number of UK deaths was below the five-year average.
Then there are other factors: China and Russia continue to report 'unusually low' numbers of deaths: interpret this how you will, but note that a Russian state-employed health demographer was fired after claiming publicly that the figures were artificially low. China altered its criteria for counting covid-related deaths several times in January and February. Aside from politics, what also matters is money: resource-poor nations have enormous difficulty testing large numbers of patients, which may explain the apparently low numbers in many African states. Then there's war and conflict. Syria reports just 48 deaths; Yemen 508: are these figures correct?
And then, again, it's worth pausing to remember that this pandemic is far from over. Some countries have barely got started: Central and South American countries are currently experiencing high case numbers and still 'moving up the table'. Countries previously held up as examples of how to do things well, including Australia and Japan, also appear more recently to be experiencing substantial increases in case numbers (though as yet not, interestingly, in numbers of deaths). Who knows who will end up with the most or fewest deaths?
So comparison is not impossible but just meaningless, at least for now. In comparative international terms, we have as yet little real idea how high the UK Covid mortality rate has been. But setting aside this inconvenient truth, even if the UK should ultimately prove to have done ‘badly’, the idea that the Prime Minister must therefore be responsible is unsustainable.
Some countries have more, and some fewer deaths. Almost all countries have had some form of lockdown: all could probably have done this a bit sooner – but in the UK at least, there is evidence that the infection rate was declining before compulsory lockdown commenced. Japan had no lockdown but appeared at least initially to have an extremely low number of deaths. But Sweden had a notoriously soft, ‘voluntary lockdown’, and is currently eighth in the ‘deaths per head of population’ table.
Population density seems really important – but then how to explain Sweden doing ‘worse’ than Japan? A major international airport hub seems very important (Belgium seems to have done poorly). The age profile of a country appears to be important too – and may, though again, really it’s too early to say, explain why Africa might not do as badly as initially feared. Multiple generations living together also has an impact. Race is a susceptibility factor. And it's clear that a capital city like London – a global hub, where thousands of people come and go from around the world each day – will have hardly helped keep numbers down in Britain.
In perhaps twelve months, after some considerable research effort has been expended, some good epidemiologists may – just may – be able to tell us which countries fared worst. Some time later, maybe, they may even offer some clues why. Until then, it's just too early to say whether Britain – and Boris's – handling of this pandemic really has been as bad as the PM's critics claim.
Professor Neil Scolding is the Burden Professor and Director of the Bristol Institute of Clinical Neurosciences. This article reflects his personal view not those of the University of Bristol