All-Cause Mortality

This discussion is another extension in a series reviewing issues surrounding the Covid-19 Pandemic and the Efficacy of the Lockdown Policy. In part, the scope and efficacy of any lockdown policy must be predicated on some assessment of the risks to life. However, aspects of this review process have been critical of some of the main sources of statistics, which appear to under-estimate infections and over-estimate deaths. As a consequence, the statistical risk of death as a ratio of deaths to infection presents a false impression, which is not only misleading, but might be questioned as misinformation, if it is intended to deceive the general public. As the following table illustrates, even if the figures were correct, statistical risk might still require some wider perspective.

Country Population % Death Deaths Infections %-Pop Deaths %-Inf %-Death %-Pop
UK 67,772,000 0.94% 637,057 250,908 0.370% 36,042 14.36% 5.66% 0.053%
Taiwan 23,570,000 0.77% 181,489 446 0.002% 7 1.57% 0.00% 0.000%
S.Korea 50,800,000 0.63% 320,040 11,142 0.022% 264 2.36% 0.08% 0.001%

If we compare the figures for the three countries shown, the UK is one of the worst affected by virtue of its percentage of deaths relative to infections (%-inf) being over 14.3%. However, this figure is questionable, when compared with a more general mortality estimate of 1% that has been associated with the Covid-19 virus. Of course, we might also compare the UK figure with Taiwan (1.5%) and South Korea (2.4%), which while both higher than the 1% suggested for the Covid-19 virus, are both considerably less than the 14.3% figure suggested by the UK.

Note: While this discussion will not challenge the accuracy of the Taiwan or South Korean figures, at least, in terms of deaths, we might still ask whether the number of infections might still be subject to some under-estimation, which would then affect the rate of deaths relative to infections. It might also be recognised that the number of deaths in Taiwan is so small that it cannot be use as a meaningful statistical sample.

The figure in the third column in the table above represents the estimated annual all-cause deaths as a percentage of each population, where the UK figure of 637,057 is in the same ballpark as the previously used figure of 616,000 for UK deaths in 2018. While there is still some significant variance in the %-deaths for each country, we might assume this variance can be generally explained by the demographics between the populations. So, given the question mark against the number of infections being recorded – see latest Covid-19 statistics, we might alternatively consider the number of Covid-19 deaths relative to annual deaths and the size of the population. If we initially ignore the UK figure, we can see the %-death and %-population figures for Taiwan and South Korea are very low and might be used as a better measure of the real risk of death.

But, why is the UK death figure so high?

As has been previously discussed under the Propaganda and the Covid-19 Pandemic heading, a UK Column news report, dated 15-Apr-2020, at minutes 1:15-4:10, questioned the accuracy of UK deaths being attributed to the Covid-19 pandemic by citing statistics for week-14, where it was estimated there were 475 deaths compared to the normal baseline figure of 10,500, i.e. just 4.5%. However, it was also highlighted that there were also 5,665 unexplained deaths in week-14, possibly as a consequence of the Covid-19 pandemic being prioritised over all other health conditions. This perspective might also be considered in terms of a statement made by Professor Walter Ricciardi, scientific adviser to Italy’s minister of health.

“The way in which we code deaths in our country is very generous in the sense that all the people who die in hospitals with the coronavirus are deemed to be dying of the coronavirus. On re-evaluation by the National Institute of Health, only 12% of death certificates have shown a direct causality from coronavirus, while 88% of patients who have died have at least one pre-morbidity, many had two or three.”

If we were to assume that UK Covid-19 deaths have also been inflated by a similar factor, as in Italy, then the percentage of deaths relative to infections, i.e. 14.3%, might actually be more compatible with South Korea, while still accepting some demographic variance. However, given this questioning of the UK figures, it is believed that some further consideration of the wider issues surrounding all-cause mortality is necessary.

Note: Let us start with a hypothetical example. A fragile 85-year old man is walking along, when a strong gust of wind causes him to fall, such that he sustains a serious head injury. As a consequence, he is taken to hospital for what is perceived to be a life-threatening injury, where unfortunately he is exposed to an influenza virus. Due to his poor general health, the shock of the head injury and a virus infection, his weaken immune system succumbs to pneumonia, where respiratory complications lead to his subsequent death. So, the question raised by this hypothetical example is what was the cause of death?

While it might be suggested that this sequence of potential causal events was triggered by the gust of wind, we might reasonably assume that this would not be listed as the cause of death. Equally, it is unlikely that the serious head injury would be cited as the cause of death, nor the exposure to the influenza virus. For prior to the Covid-19 pandemic, there was no mandated requirement to list influenza as a contributory cause of death, even assuming that the presence of this viral infection was known, such that it would probably be recorded as a case of pneumonia. As such, this death would have been recorded as one of the 637,057 (0.94%) all-cause mortalities in the UK.

How might this change in light of the Covid-19 pandemic?

Let us assume the same hypothetical example with the exception that the old man was exposed to the Covid-19 virus, not an influenza virus. Under the current requirements, the presence of the Covid-19 virus, tested or simply suspected, would have been recorded on the death certificate and possibly used by UK statistics as a Covid-19 death. If so, we possibly need to better understand some statistics surrounding all-cause mortality, which we shall discuss in the general context of UK figures. Below, are two general graphs, where at the top, we see the profile of the UK population by age along with the percentage of deaths associated with each age group. Below is a very limited breakdown of the causes of death in the UK.

From the perspective of total deaths, 27% of all deaths are associated with people in the top 95-year old group, although the chart left highlights that these deaths represent a much higher percentage (45%) of that population as their numbers are understandably much lower than other age groups. In terms of UK statistics, the 95-year age group would account for about 172,005 of the 637,057 (0.94%) all-cause mortalities. However, the bigger issue around the assessment of risk associated with the two charts above is why the mainstream media and the government only appear to be interested in presenting the large number of accumulating deaths being attributed to the Covid-19 virus along with the number of daily deaths as an unprecedented crisis requiring unprecedented action, i.e. lockdown.

Note: As of 22-May-2020, the total number of UK deaths attributed to the Covid-19 pandemic was 36,042, while the seemingly unprecedented daily death rate has fallen from a reported peak of over 1000 in mid-April 2020 to about 290. However, what is rarely, if ever, mentioned is that UK annual rate of death (637,057), as shown in the initial table, equates to a daily death rate of 1,745, where 1,361 (78%) are related to medical conditions. As such, the media appears more intent on highlighting the 290 daily Covid-19 deaths, while ignoring the much larger 1,361 daily deaths associated with all-cause mortality, where many might also have been avoided with more appropriate medical advice.

So, might we attempt to put the risk of death from the Covid-19 virus into some better perspective?

The risk of all-cause mortality equates to 637,057 annual deaths, which is 0.94% of the UK 67 million population, as shown in the chart at the top on a logarithmic scale. Of course, as highlighted by the blue curve, this risk increases with age, which should be no real surprise to anybody, where a very basic breakdown of the cause of death is shown below.

Without reference to any Covid-19 deaths at this stage, the percentage of deaths under all ages below 50 is less than 1%, which means that 99% of the 637,057 annual deaths have to be associated with the age groups above 50. In fact, the percentage of annual deaths is still above 90% if we only consider the age groups above 70. Again, the fact that the risk of death increases disproportionately with age should not come as a surprise to anybody.

But what about the specific risk of death from the Covid-19 pandemic?

Let us proceed on a speculative assumption that the UK Covid-19 deaths fit the same age distribution as the annual UK deaths, as shown below on top. Today, 22-May-2020, the current UK Covid-19 deaths are reported to be 36,042, such that we might also speculate on an upper limit of UK Covid-19 deaths for the whole of 2020 to be 50,000. However, given the questioning of the accuracy of the current Covid-19 deaths, we might also consider a more conservative figure of 25,000 deaths, such that we can  see the upper and lower projections against age, as per the chart below.

Note: What we might realise from the previous charts is that even the higher Covid-19 projection of 50,000 deaths would only represent 7.85% of the annual number of deaths, i.e. 637,057, while the lower projection of 25,000 would only represent 3.93%.

The following charts have been sourced from the EuroMomo website, which monitors mortality in 24 European countries. As such, the vertical scale of weekly deaths is an estimate of a population of some 400+ million people. Therefore, we might attempt a general correlation of the figures in the charts below with the UK all-case mortality figures by assuming an average figure of 50,000 deaths per week or 2.6 million all-cause deaths per year for the combined European countries. As such, the figure of 2.6 million might be compared with the UK all-cause mortality figure of 637,057 per year, as used in this discussion, on this basis the UK numbers would approximate to about 20% of the figures shown in the following charts.

First, some clarification of the composite charts shown. The top shaded grey region provides a more direct comparison of all-cause mortality for ‘all-ages’ against the ‘15-64 age’ group on the same vertical scale. However, as per the EuroMomo website, the details of the ’15-64 age’ group is enlarged in the inset at the bottom of the chart. Based on the ‘normal range’ average, the charts suggest that the average weekly deaths for the ‘ all age’ group to be in the region of 50,000 per week compared to only 7,800 for the ‘15-64 age’ group. While these figures are higher than the UK estimates in this discussion, these estimates were only intended as approximations of the overall increase in all-cause mortality with age.

Note: Broadly, the ‘all-age’ charts suggest a variation of 20,000 in the weekly deaths between the summer and winter months, which we might attribute to degrees of vitamin-D deficiency, although this is only a speculative assumption at this stage – see video ‘ Vitamin D and Human Health’ for more details. These charts also suggest a variation of 10,000 in the peak weekly deaths in the years 2017 to 2020, which may be attributable to various viral infections, e.g. influenza, which are also known to be seasonal.

The 2020 spike in the weekly all-cause deaths suggests an increase from about 70,000 to 85,000 (20%) for the ‘all-age’ group and an increase from 9,000 to 10,000 (11%) for the ’15-65 age’ group, which we might assume has been caused by the Covid-19 pandemic. As such, we might assume that these percentages reflect the increase in all-cause mortality in 2020 resulting from the pandemic. However, it is difficult to correlate a 20% increase in all-cause mortality with the UK figure of 637,057, as used in this discussion, for it would suggest an increase of 127,411 deaths in 2020 directly attributable to the Covid-19 virus. As previously discussed, the projected increase in the UK all-cause mortality has been estimated to be in the range of 25,000 (3.9%) and 50,000 (7.8%), where 90% of these deaths would be in the +70 age groups. We might also consider whether some percentage of these additional all-cause mortality deaths in the +70 age groups were people that already had multiple health conditions and a weaken immune system. If so, it is possible the Covid-19 virus might be analogous to the ‘gust of wind’ in the earlier hypothetical example and that any additional deaths may be reflected in lower all-cause mortality figures next year.

Note: It is realised that many people will not like the discussion of all-cause mortality based primarily on statistics as each death is not just a number but a human life. However, the reality of the current situation, especially in terms of the social and economic consequences of lockdown, requires the impersonal perspective of statistics.

Of course, at this time, this discussion of statistical numbers is still speculative conjecture. However, concern has to be raised when the general public is constantly confronted with the alarming numbers of daily and aggregated deaths being attributed to the Covid-19 virus without any reference to the actual statistical increase in all-cause mortality. Based on this outline discussion, there is reasonable, but not necessarily conclusive, evidence that the overall UK Covid-19 deaths may have been considerably over-estimated, such that any suggestion of a 14.3% death rate relative to infections should be rejected in favour of a more realistic estimate. This suggestion might then support the idea that any excess Covid-19 deaths should only be compared against the typical estimate of all-cause mortality, which as a smaller percentage may support the case for an earlier relaxation of lockdown policies given the obvious negative impacts on society.

Note: The truth is that, irrespective of age, we all have a statistical risk of death, as suggested by the chart right. While people in the older age groups might not like to see their own risk presented in such a stark manner, it is possible that they have already come to understand the reality of this risk far better than younger generations.

Therefore, while a modern caring society should consider every practical way to mitigate the risk of unnecessary deaths, we might question whether it should put a disproportionate burden on the lives of younger generations. For one of the realisations we come to understand as we grow old is the importance of enjoying life when we are young. So, if society really wants to minimise the risk of all-cause mortality, maybe we should really be asking why current medical science continues to ignore the mounting evidence about poor diet being the main contributor to early and unnecessary deaths – see Prevention versus Cure for more details.

Note: While this discussion has tried to avoid straying into politics, it is clear that the Covid-19 pandemic now compasses not only politics, but also much wider social and economic issues. So while this discussion will refrain from direct comment on such issues, it will make reference to a video entitled Guilty Of Breathing, dated 24-May-2020. While it might be argued that this video does have political bias, the more important issue is whether it is generally factual in its statements.