ONS week 14 figures out

The ONS figures for week 14 (England and Wales) are now out. Finally we can see a rise in deaths, many of which can be attributed to Covid-19 (although we don’t know how many have been caused by such things as the cancelled treatments). There are 6000 more than the five-year average (click to enlarge).

Does this change my mind on the lockdown? No. I predicted a rise over April as the virus hits, and then a fall back in summer (which the government and the relevant scientists also expect). Then probably a bad winter. As I recently showed on my blog, we’ve had two winters in a row which have been very low on excess winter deaths. This indicates that there are large numbers of vulnerable people at risk from viral infections. (Laurence Hodge alerted me to this.)

Normally a good winter with lower deaths is followed by a bad winter with higher deaths, so two good winters in a row could see a particularly bad winter next time. But destroying our economy is not the way to deal with a disease. We have a nasty virus going around, but that’s preferable to having a nasty virus going around and a crashed economy.

Also, it has to now be doubted whether the lockdown is making much difference, esp. in non-urban areas. Why make a bad situation worse?

As for the argument that a lockdown is needed to allow the NHS to cope — which, as it happens, was the main argument originally advanced for the lockdown — it now turns out that the NHS currently has 40% of it acute care beds unoccupied.

Some extra deaths could be caused by the lockdown, and the many cancellations of existing treatments. We don’t know at this stage. But note that there were only 500 extra respiratory deaths.

Note also that we now have strong confirmation that young people aren’t affected. The week 14 increase is mainly in elderly people. The death numbers up to the mid-40s, though, are completely normal, and then it gradually starts to increase as we go up the age ranges. 57 more deaths for 45-49 age range than last month, a difference which probably isn’t statistically significant. About 350 more deaths than previous weeks for the 65-69 age range. 600 more for 70-74. 800 for 75-79. 1000 for 80-84. 1000 more for 85-89, 1000 for 90+. So the great majority of the extra deaths are in people over 70.

Do I think this is worse than flu? It probably is. Is it a once-in-a-century disaster? I still don’t think that, but the economic crash could be.

Update: We may well see a rush of deaths over April, though, as there are now reports that Covid-19 has got into thousands of care homes in the UK. If true, many of the elderly people who are in a bad way in them, but who managed to survive over the last two winters, are now at high risk. Possibly many of the excess winter deaths that we saw managed to avoid with this recent mild winter are going to be compressed into the next month. But we still need to look at things from a more long-term view.

Share this article on social media:

28 thoughts on “ONS week 14 figures out

  1. Hi Hector,

    The ONS says that the respiratory deaths catch some Covid – and vice versa, so neither is a subset of the other.

    If we have a COVID death which is not respiratory, is that implying a death ‘with’ rather than ‘of’?

    Or are the COVID deaths in the respiratory the COVID deaths with another respiratory disease?

    Or are they completely useless categories?

  2. And another interesting thing, the % of each age group dying dosn’t seem to have changed much – suggesting CORVID is tracking normal death rates. the 85+ category has hung around 40% of the total number of deaths each week. If anything it’s come down slightly.

  3. Thanks for that Hector.
    Purely for “fun”, I’ve just gone through today’s Worldometer figures and worked out the percentages of deaths from the total numbers of cases.
    Usual riders of course that a) the numbers given by Worldometer are correct and b) my calculations are also correct. Someone with more of a mathematical bent than I have might like to check please?

    USA 4.02%
    Spain 10.46%
    Italy 12.8%
    China 4.06%
    France 10.9%
    Germany 2.45%
    UK 12.78%
    Ireland 3.42%
    Sweden 8.39%
    Japan 1.87%

    Make what you will of that and I acknowledge varying demographics and geography etc, but UK percentage suggests to me that, as I’ve thought all along, our numbers are being “manipulated”, especially when comparisons with others on the list are made. And, didn’t someone once say, “Comparisons are odious”?

  4. “now reports that Covid-19 has got into thousands of care homes in the UK.”

    So the people we already knew were vulnerable are the most exposed. How much of that is down to the panic instilled in the whole population causes them to be thinking about themselves and not the priority area? How much health care was missed because of isolation?

  5. It seems almost too trivial a point to mention, but if we’d never heard of C19 and something else had caused the economy to grind to a halt, preventing people from going to work, preventing relatives from visiting the elderly and so on, would we eventually expect to see it reflected in the national death figures? I think the answer is obviously yes. It wouldn’t be apparent on the first day, and maybe not the first week. But after a few weeks I would expect to see people being tipped over the edge – the effects are often going to be nonlinear. Who knows what four weeks of artificially-imposed loneliness does to a person’s zest for life? The effect of an enforced change of diet? A lack of exercise? A lack of sunlight? Loss of routine? Boredom, drinking, comfort eating. It’s probably not going to be positive or neutral in terms of longevity.

  6. It does defy belief that is can be 10x deadlier one place than another.

    NZ has lost a few more lives, all in their 90s. They died alone and will be denied funerals.

  7. Surely that’s Spiegelhalter’s point: the Wuhan virus kills people in much the same proportions as other causes of death. His inference is that, statistically speaking, everyone has doubled their risk of death in the next twelvemonth – you’ve got the old risk plus the Wuhan risk. If you’re a fit thirty-year old the extra risk is, in absolute terms, negligible; if you are an ill eighty-year old it is large. Naturally his analysis can be no better than the data he’s using.

    https://medium.com/wintoncentre/how-much-normal-risk-does-covid-represent-4539118e1196

  8. Very likely to have a significant impact. Social contact is critical for the well being and even more so for those in rest hoes. People with close families live significantly longer.

  9. When I saw the ONS deaths figures for week 14, the first thing that struck me was that there were 6302 total deaths more than the 5-year average, but only 3475 deaths with COVID-19 that week – so 2827 non-COVID-19 deaths more that week than the 5-year average.

    This got me thinking, so I downloaded the ONS data into a spreadsheet, and plotted graphs of non-COVID-19 deaths for 2020 vs 2015-19 average deaths, and non-respiratory deaths for 2020 vs average non-respiratory deaths for 2015-19.

    I also calcuated 12% of the COVID-19 deaths each week this year, as this study found in Italy “only 12 per cent of death certificates have shown a direct causality from coronavirus“ https://web.archive.org/web/20200324214448/https://www.telegraph.co.uk/global-health/science-and-disease/have-many-coronavirus-patients-died-italy/ and then plotted a graph of total deaths less 12% of the deaths with COVID-19 for 2020 vs 2015-19 average deaths.

    The graphs can be downloaded, here is the non-respiratory deaths graph https://www.dropbox.com/s/k26dja2o1g4ow44/2020-04-14%20Non-respiratory%20deaths%20graph.jpg?dl=0 here the non-COVID-19 graph https://www.dropbox.com/s/ckgl8z90vqr3vnc/2020-04-14%20Non-COVID-19%20deaths%20graph.jpg?dl=0 and here the non-12%-COVID-19 graph https://www.dropbox.com/s/sh4rhiuhex33grh/2020-04-14%20Non-12pc-COVID-19%20deaths%20graph.jpg?dl=0

    From the graphs, it can be seen that the 2020 non-respiratory deaths & non-COVID-19 deaths were similar to the averages for 2015-19, up until the lockdown was introduced on 24th March – and then the 2020 figures have risen significantly. For week 14 (week to 4th Apr 20), 2020 non-respiratory deaths were 5559 above the 2015-19 average, and 2020 not-with-COVID-19 deaths 2607 above the 2015-19 average.

    The lockdown may be saving people from dying of COVID-19, but it appears to be causing more people to die of other causes 🙁 Whether from the NHS being so focussed on COVID-19 that it ignores other ill people, elderly people deprived of their social life feeling there’s nothing left to live for, people with heart attacks etc being too scared to go to hospital/call an ambulance, suicide, despair etc no-one knows – but the lockdown appears to be having an immediate human cost, as well as an economic cost.

    If you would like more details on my calculations, or my spreadsheet, please contact me by e-mail and I’ll be happy to oblige!

  10. The ONS note Of the deaths registered in Week 14, 3,475 mentioned “novel coronavirus (COVID-19)” So if excess deaths are of the order of 6,000, it suggests that there is under-recording from the official statistics, as it is unlikely that the difference is made up from those not being treated for other conditions (surely something the press would have found by now?).

    However, while that will doubtless be the focus of much attention in the press, the real issue is comparison with previous severe flu epidemics. Unfortunately we will have to wait until we have a clear peak in the data to make the real comparison, but it is of the same order as the 16,237deaths for England and Wales recorded in the week ended 9 Jan 2015.

  11. …so 2827 non-COVID-19 deaths more that week than the 5-year average.

    And this is with deaths from influenza and pneumonia hardly rising at all; whilst road traffic and other regular activity are reduced to a tiny fraction of normal.

    This is significant.

  12. I’m in my 70s, pretty fit (for my age etc), do no work, have pension, no mortgage, no debt, savings adequate blah blah.I don’t go out in the evenings, preferring to get drunk at home, which saves me even more money.
    Before anyone accuses me of being smug, I’m going up the bloody wall, so I can only imagine what it must be like to be of working age, stuck at home, with kids off school, work looking increasingly iffy, mortgage and car payments to be made and the high spot of the day/week/month a visit to Tesco.

    Q1. What parameters are the government going to apply, in order to end this ridiculous situation?

    Q2. What will the government then do next Winter, when someone sneezes in China? Bang us all up again?

  13. I’m in my 70s, pretty fit (for my age etc), do no work, have pension, no mortgage, no debt, savings adequate blah blah.I don’t go out in the evenings, preferring to get drunk at home, which saves me even more money.
    Before anyone accuses me of being smug, I’m going up the bloody wall, so I can only imagine what it must be like to be of working age, stuck at home, with kids off school, work looking increasingly iffy, mortgage and car payments to be made and the high spot of the day/week/month a visit to Tesco.

    Q1. What parameters are the government going to apply, in order to end this ridiculous situation?

    Q2. What will the government then do next Winter, when someone next sneezes in China? Bang us all up again?

  14. COVID-19 (ICD-10 U07.1 and U07.2) is a respiratory disease (ICD-10 J00-J99). If COVID-19 detected or suspected then it is counted as COVID-19 but that does not mean it was the cause of death. The difference is ‘with’ not ‘of’.

    COVID-19 deaths in week 14 3,475, but Respiratory Deaths are 2,106, so at least 1,369 are ‘with’ not ‘of’.

    Week 14 2020 is very high for this late in the flu season, yet the weekly total of 16,387 is only 150 more than the peak of 16,237 seen in week 2 of 2015

  15. Covid now has its own special ICD categories, U07.1 and U07.2. However, there are also new protocols that enable Covid-19 deaths that are respiratory deaths to also be inserted into the old respiratory J00-J99 categories.

  16. And wasn’t the method of recording altered (to the government’s advantage?) in Week 12?
    It’s simply a hunch, but I still keep getting large smells of rodent with these numbers.
    I do feel these figures are being both manipulated by “authority” and reported as such by the MSM. No great conspiracy theory, merely done to keep the frightened populace under wraps by frightening them further, and to justify the incarceration.
    Seeing the people panicking and clearing the shelves because of what had been reported on the telly, they then panicked themselves and brought in the lockdown, which they now realise was a hasty decision, one that will cause an ENORMOUS amount of damage, of all sorts.
    What to do?
    Bump up the figures (week 12), put the PM in hospital for added realism, cross your fingers and now try to figure out what the hell we’re going to do next.
    My guess is the figures will start to come back down, just before they decide to let us out.

  17. Yes John your a handful years younger than me, and I have been saying much what you say from the start.

    It seems now that C-19 comes late in the season on top of a moderate flu/pneumonia season (circa 25k). This gives us a high overall death rate for early/mid April. It is confusing as I said on another thread (and Tom says above) that there appear to be overlapping categories where ONS say if more than one cause of death is listed (inc. C-19) it can be recorded twice. As I read it and I am more than happy to be put right, it means that a death can be recorded twice. If not they would know how many such deaths they dealing with and they don’t seem to.

  18. The all-death figures do not ever count one death twice.

    The Covid-19 category is an ‘overlapping’ category, and so some deaths will go into it and another category. But the all-deaths do not have any deaths double-counted.

  19. Death certificates may record more than one ’cause of death’ COVID-19 is notifiable so it is on the death certificate if tested or suspected. But it may be entirely incidental, you could be fatally injured in a RTA but if tested positive for COVID-19 then that’s with COVID-19 but with not necessarily of. Getting these details right is why it takes time to produce reliable statistics.

  20. @Tom

    Re: “With” not “From”

    afaik NHS & ONS are being clear it’s “with”

    Every BBC R4 News etc I heard today says “Died From Coronavirus” – not much accuracy, truth and impartiality in BBC News, yes News not opinion

  21. Work and Pensions Secretary Therese Coffey appears on the show to discuss the latest developments with the coronavirus including the increasing number of NHS staff deaths.
    Piers Morgan has a hysterical rant and of course says “Died From Coronavirus”
    https://www.youtube.com/watch?v=DZTq2DHKMHo

    “Care Home Deaths Not in Stats Yet” – same as previous years

    Care Home PPE – unless state run, not Gov responsibility. Why didn’t they and NHS Trusts order more PPE in Dec Jan, Feb and early March?

    Piers and Minister ignorant on VAT too. UK Gov can only change VAT status if EU gives permission. Therese Coffey should have said “I don’t know, ask Sunak” instead of digging down rabbit hole and looking stupid.

    OTish, but interesting:

    Berlin Metro – three languages: German, English and… French? Nope; Dutch? Nope…
    https://youtu.be/u3KsFU4MqJQ?t=91

    wow – comments permitted

  22. And remember: these percentages do not distinguish between ‘of’ and ‘with’, and are determined by calculating against known infections. Given few countries have tested more than a small percentage of the population (one study estimates as low as 6% of actual infections have been confirmed in Western nations), the actual percentages will be lower. Possibly much lower.

Comments are closed.