Revealing Chinese-virus excess-death graphs #coronavirus

By Christopher Monckton of Brenchley

In the United Kingdom, total excess deaths are now causing real alarm among statisticians. The 18,516 deaths recorded in England and Wales in the week to April 10 are the highest weekly total since winter 2000 – and we are not in winter now. There 7996 (or 76%) more deaths than the mean weekly death toll for the time of year.

Fig. 1. UK weekly excess deaths by the thousand, compared to the five-year mean, attributed to the Chinese virus (grey) and not currently attributed to it (green), weeks 8 to 15 of 2020.

It can no longer be argued that the Chinese virus is “no worse than the annual flu”. If these are the excess deaths even after a lockdown, one can imagine how much worse the position might have been without a lockdown.

Of the 7996 excess deaths, 6213 had the Chinese virus registered on the death certificate, leaving 1783 unexplained excess deaths. A handful of these are attributable to suicide and other adverse consequences of the lockdown: inferentially, nearly all the rest are uncounted Chinese-virus deaths.

Sir David Spiegelhalter, Professor of the Public Understanding of Statistics in the University of Cambridge, described the excess-deaths spike as “incredibly vivid”. He told the Daily Telegraph:

“I don’t think I’ve ever been as shocked when I’ve looked at something, particularly as just over half of that spike were death certificates with COVID written on them. We knew there was going to be a jump in COVID-registered deaths. I hadn’t expected such a huge number of deaths which didn’t mention it on the death certificate.”

Sir David would not have been so surprised if he had been tracking our daily graphs showing the compound daily growth rates in confirmed (i.e., usually more serious) cases and in deaths. These growth rates, though a lot less bad than before the world began to take the Chinese virus seriously, are still dangerously high, baking in substantial numbers of future deaths.

The unallocated deaths reveal yet another weakness in HM Government’s recording and publication of the figures. It was already known that the death statistics announced in Downing Street’s daily press conferences were underestimated by at least 52% nationally (41% in England and Wales, 70% in Scotland, 91% in Northern Ireland: Fig. 2) because the figures were for hospital deaths only, excluding all deaths in care homes and in people’s houses.

Now it seems that even after adding 52% the figure is a 76% underestimate (Fig. 1), because the Government has not taken the elementary step of issuing instructions that all fatalities where the virus is suspected to have caused suffocation should be tested for the virus and the results reported to it within 24 hours where possible.

Fig. 2. Daily death counts reported by hospitals and the total corrected by the Office for National Statistics to allow for deaths registered later.

The absence of credible death statistics compounds the difficulties caused by HM Government’s failure to give instructions to hospitals and doctors to report all cases where the patient was infected but has recovered. In the absence of these basic numbers, HM Government is visibly stumbling about in the dark.

Daily growth rates in new cases and in deaths are no longer falling much, but they need to be lower before it becomes safe to end the lockdowns in those nations that have them. Sweden, with no lockdown, continues to track a little above the global daily growth rate in cumulative cases, and appreciably above it in cumulative deaths. Sweden has 175 deaths per million population, compared with 64 per million in Denmark, 34 in Norway and 25 in Finland.

Fig. 1. Mean compound daily growth rates in cumulative confirmed cases of COVID-19 for the world excluding China (red) and for several individual nations averaged over the successive seven-day periods ending on all dates from March 28 to April 20, 2020.

Fig. 2. Mean compound daily growth rates in cumulative COVID-19 deaths for the world excluding China (red) and for several individual nations averaged over the successive seven-day periods ending on all dates from April 4 to April 20, 2020.