Dr No had high hopes that, given recent data published by ONS, PHE and NHS England, on deaths by covid vaccination status and vaccination rates, he would be able to get something of an answer to the vexed question of whether the covid vaccines increase or decrease overall mortality. Fat chance. He quickly found himself wandering on a tundra of statistical quicksand. Key data was missing, and the numbers that were available simply didn’t add up. He spent far more time than he should have done peering up ONS drainpipes and down PHE rabbit holes getting nowhere. It was yet another landscape of cavernous numbers meaningless to man. Nearing desperation, he was about to give up, when it occurred to him that he could make a simple adjustment, and get an answer. This post describes the calculations, including the adjustment — which is not without it problems — and comes to two important conclusions: firstly that covid vaccines may well increase mortality in the short term, and secondly, being unvaccinated does not appear to carry increased mortality risk. 

In contrast, recent ONS and PHE reports have continued on their jolly tractor production is up and vaccinee covid deaths are down drum roll. Using opaque calculations, ONS tells us — complete with irritating animated chart — that “age standardised mortality rates for those fully vaccinated are consistently lower”. PHE’s latest vaccine surveillance report  is equally bullish, telling us that the “rate of death within 28 days or within 60 days of a positive COVID-19 test…is substantially greater in unvaccinated individuals compared to fully vaccinated individuals”. Tractor production is indeed up, and all tractors are pulling their weight, it seems. Or are they? There are at least two major problems here. The first is that both ONS and PHE report only covid deaths, and covid deaths in turn rely chiefly on what Dr No calls the sewing machine test, the PCR test that finds broken needles in haystacks, and then declares the haystack to be a working sewing machine. The second is that, by focusing on only covid deaths, you have no idea of what is happening to non-covid deaths, and so will miss, for instance, any vaccine related cardiac deaths. Tractor production may be up, but here and there, wheels are starting to fall off.

There is a simple, easy fix to both of these problems: instead of counting just covid deaths, count all deaths. The importance of, and benefits from, counting all cause deaths has been discussed many times on this blog: the robustness of the data: a death is a death is a death. All fuss about is it, or isn’t it, a covid deaths is at a stroke removed. But where to find all cause mortality data by vaccination status? It turns out that, although the published ONS “Deaths involving COVID-19 by vaccination status” report only, and unsurprisingly, covers covid deaths, the underlying data also includes non-covid deaths by vaccination status. Bingo! By adding the two together, we have all cause mortality, by vaccination status. The wheels might have come off PHE’s and ONS’s tractors, but Dr No’s tractor now has all four wheels firmly in place, and ready to spin. With NHS England data on vaccination status by age band, ONS data on deaths by vaccination status, and a reference dataset of mortality by age, everything needed to calculate standardised mortality ratios (SMRs) by vaccination status is now in place.

Only it turns out it isn’t. The SMR is a simple enough concept: you take the mortality by age band from a reference population — Dr No used all cause mortality for England in 2019, as the most recent pre-pandemic data, downloaded from here, recalculated to match NHS England’s age bands — and apply those rates to the same age bands in your population of interest — in this case, the unvaccinated, the partially vaccinated, having one dose, and so perhaps at most risk of vaccine related harm, and the fully vaccinated, to derive an expected number of deaths, had the population of interest experienced the same mortality as the reference population. By comparing the expected number with the observed number, we get the standardised mortality ratio, or SMR, as the observed number of deaths divided by the expected number, multiplied conventionally by 100, so that an SMR greater than 100 means more deaths than expected, and an SMR less than 100 means less deaths than expected. The problem is the ONS deaths by vaccination data is seriously incomplete. There aren’t just a few missing deaths, there are hundreds. The observed numbers are absurdly low, and so the SMRs — O/E x 100 — are all absurdly low.

Taking week 26, the week ending 2 July 2021, which is the most recent week for which data is available, ONS’s deaths by vaccination status reports a total of 6,956 deaths for all categories: covid and non-covid deaths, across all categories of vaccination status. This is 1,271 less deaths than 8,227 recorded for week 26 2021 in ONS’s own weekly death reports, and 1,739 less than the 2015-2019 five year average of 8,695 deaths. This means we need to adjust the deaths by vaccine status observed number to bring it in line with the actual number of recorded deaths. Although it involves a large and possibly unwarranted assumption — that all categories of vaccination status are equally affected — the simplest way to do this is to increase each count by the same factor so that the total number of observed deaths used to calculate the SMRs matches the total number of deaths in the weekly reports, multiplying each category by a factor of 1.18272 (6,956 x 1.18272 = 8,227). We can now recalculate the SMRs, using these adjusted observed numbers. The results are shown in Table 1.

WordPress Responsive Table

Table 1: SMRs by vaccination status. See text for data sources and assumptions. Data for all one dose categories aggregated into one category to match vaccination uptake data categories (unvaccinated, one dose, two doses). Unvaccinated by age band numbers were calculated by subtracting any dose numbers from the total population numbers for each age band. Note that the confidence intervals are tight, and far away from 100, the value which indicates no difference in mortality. Full spreadsheet used by Dr No to calculate the SMRs can be downloaded here

These results do not give much comfort — unless you happen to be unvaccinated, and plan to remain that way. Being fully vaccinated does appear to reduce mortality, some comfort maybe if you are already fully vaccinated, but the two stand out SMRs are for the unvaccinated, who have the lowest mortality by a long shot (no pun intended…), and the partially vaccinated, who have a strikingly high mortality. Perhaps Dr No’s friend, aged in his seventies and admitted to hospital in the spring with a confirmed heart attack three weeks after receiving his first jab did have a vaccine adverse reaction after all. Happily Dr No’s friend survived, but what these SMRs tell us is that he might not have been so lucky.

Dr No has to end this post with three important caveats, as he has presented SMRs which appear to show significantly raised mortality after a first dose of covid vaccine, a period when, it might be supposed, any acute adverse effects are likely to appear, a finding which is far from trivial, and greatly at odds with the official narrative. He has also reported an SMR which suggests remaining unvaccinated is not associated with higher mortality, again at odds with the official narrative. First and foremost, on pain of repetition, he applied the observed uplift adjustment equally to each vaccination status category, assuming equal effect across the board, an assumption that may or may not be justified; but for the present time unavoidable, as there is no way of testing the assumption, or other way of making the adjustment.

Secondly, strictly speaking, SMRs should not be used to compare different populations (unvaccinated, partially vaccinated and fully vaccinated) because the populations may differ in their degree of underlying risk — perhaps the recently vaccinated are somehow at an inherent non-vaccine related higher risk of dying — even if it is hard to see why this might apply in the summer of 2021, or how this could bring about such huge differences in the SMRs. Thirdly, we must ask the Mark One Eyeball question: do the results look plausible? On the face of it, perhaps not, but against that there is always the epidemiological version of Sherlock Holmes’s dictum: when you have eliminated the fluff, whatever remains, however improbable, must be the truth. Dr No has checked and rechecked the data, logic and the calculations, and while he may have made a mistake, he cannot see one. Readers are encouraged to inspect the worksheet used to calculate the SMRs, available here, and post any observations in the comments below.

With these caveats in mind, the SMRs presented in this post still don’t prove anything, but they do call into question the official narrative on vaccine safety, a narrative which can at best be said to be based on opaque calculation smoke and missing data mirrors. In contrast, the SMRs presented in this post are based on open data sources and transparent calculations, and appear to be valid, at least in general extent. But the real truth lies in ONS publishing, in full, the underlying age stratified all cause mortality by vaccine status data, which it has so far declined to do. Dr No fears that the nearer this data gets to confirming Dr No’s SMRs are indeed close to the truth, the less likely we are to see it published.

Comments

    • dr-no Reply

      Thanks for the link (tidied to keep it on the page, hope that is OK). Dr No had come across the underlying Fenton and Neil post, but not the one you linked to, which adds further analysis.

      Fenton and Neil takes a different approach, and comes to much the same conclusions (ONS narrative is suspect; we need to see the underlying data), but Dr No wanted to include the NHS England vaccination status by age band data as a sort of denominator input, and so took the approach he did. He is also not entirely sure one of the things Fenton and Neil did (adding covid and non-covid ASMRs) is necessarily valid, as ASMRs are not ‘real’ numbers.

      There is much the same problem with the PHE’s weekly vaccine surveillance reports, which is where Dr No started this enquiry. The most recent one (link in the post), for example, breezily claims mortalities of 49.5 and 156.0 per 100,000 for the fully and unvaccinated 80+ year olds respectively, but these numbers don’t all add up. Based on numbers vaccinated (from NHS England) the 49.5 figure does appear to be correct, but the 156.0 appears to come from thin air (198 deaths among 23,850 unvaccinated gives a rate of 83.07/100,000).

  1. steve Reply

    Dear Dr No

    I did mention awhile ago that I would not have this particular vaccine. My wife chose to have it, as did my 20-year-old son. My 17-year-old daughter decided to be as difficult as her 65-year-old father and not have it.

    This small piece of citizen science delivered the following outcomes (so far).

    We have all caught Covidness.

    My wife, double jabbed (AstraZeneca) has been very ill. Fatigue, Gastrointestinal problems, loss of taste and small, Covid tongue, shivers and fever and very sensitive skin.

    Yours truly, feels like he has a bad cold, and nowhere near as bad as his wife.

    His son had it for a few days and is now better.

    His daughter, not sure really, as just picked it up.

    Noticed on the BBC a hospital visit in which people who were double vaccinated were in intensive care, there was a person unvaccinated, and they were interviewed.

    The lady who was double jabbed said she was thankful or likely would have died. The unvaccinated said she had not bothered with the vaccine, but wish she had.

    A hospital consultant was interviewed and told people to get the vaccine, or they will be worse off.

    This appalling event was, I suppose, to say, we told you so.

    From my citizen scientist perspective, I think it raised more questions than it answered.

    Why does a 65-year-old vaccine refusenik get off scot-free and a younger double jabbed female suffer?

    Why do the BBC, the patients, the consultant assume having these vaccines are always good?

    Why not look at this from another angle, for example;

    Is it possible a vaccine may increase the chance of getting ill as and when people catch the wild virus? I’ve no idea whether Pathogenic Priming/ADE is happening or not, but shouldn’t it at least be looked at?

    There are obviously dozens of other questions.

    We are, by the way, very fit and are used to long walks, cycling and or tennis. Low BMI and no co-morbidities

    Take Vitamin D3, K2 (MY7), Magnesium, Zinc, Vitamic C and a new herb called Andrographis Paniculata (which was being considered as a possible treatment in Thailand and China).

  2. dearieme Reply

    Though reluctant I took my double dose of Pfizer, mainly to assuage my wife’s worries. Currently we are both of a mind to reject a booster jab. Wait-and-see seems to be a reasonable policy – but is it?

    • dr-no Reply

      W+S (wait and see) is a very well established time honoured and eminently sensible medical approach. 21 year old male with one day history of central abdo pain, nausea and low grade fever. Abdo – NAD (nothing abnormal detected). ?Early appendicitis. Plan: W+S.

  3. Frank Omega Reply

    Thanks Dr. No and Steve. I hope Steve’s family stays healthy going forward. I watch the U.K. and Israel data pretty closely from here in the States because the propaganda is even worse here I think. They have been trying to claim that the hospitals and deaths are 99% unvaccinated here, which I know is BS and does not line up with the U.K. and Israel data at all. Thanks Steve for the tip about Andrographis Paniculata. I am not sure if I can get it here, but I will look into it. I assume you know about Quercetin, Melatonin and I have read some good things recently about Nigella Sativa. Even antihistamines can help. Have a pleasant night.

  4. John B Reply

    ‘.. that “age standardised mortality rates for those fully vaccinated are consistently lower”.

    Correlation really is a devil. How can it be known that mortality of those fully vaccinated would be any different if not vaccinated, without confirmation by external data? The vaccinated population is largely below age 65 and therefore higher since the under 65 population is greater than the over 65 population. In the under 65 group incidence of Covid – pre-vaccine era – was already low, serious Covid even lower, and fatality almost absent. Over 90% of deaths are above age 70, median age 84, 95% of whom with comorbidity. A large number of people are excluded from being vaccinated because they have medical conditions which reduce effectiveness of their immune system, or pregnancy or with allergies.

    They are not looking at two groups which are identical. It’s like noting age standardised rates for banana consumption are higher in group A than group B: group A being chimps; group B being cats.

    It says ‘age standardised’ but I wonder how exactly they have done this when fatalities from CoVid are skewed significantly to the older age group, in very low in the under 50s? That would of course explain the vaccinated fatality (under 65) being lower. And how have they dealt with confounding factors – clearly not if the only benchmark is the Holy PCR (PBUH) and dying ‘with’ not ‘from’ is counted.

    As an aside: I note that when World ‘leaders’ appear before the media, both/all wear masks – uniformly black – despite both having been vaccinated. Do vaccinations work or not? If they work why the masks; if masks are needed and worn so ostentatiously then clearly confidence in their effectiveness is low. (Perhaps the media is riddled with a new killer ‘variant’. )

    Vaccinated people can transmit the virus – we are told – so why the vaccination passes, why are people with those allowed into places where the non-vaccinated who are in aggregate no more likely to transmit the virus because of natural immunity are not? Vaccinated people being ‘less likely’ to transmit doesn’t cut it: in a crowd it just takes one.

  5. Tom Welsh Reply

    “Being fully vaccinated does appear to reduce mortality, some comfort maybe if you are already fully vaccinated…”

    Dr No, I think you could have made this clearer. As far as I can see from your figures, being fully “vaccinated” appears to reduce mortality ***compared to being single-“vaccinated”*** – but to increase it compared to remaining unvaccinated.

    It seems to me that unless “vaccination” reduces mortality quite definitely, there is no advantage to it. If, on the other hand, it appears significantly to increase mortality…

    • Tom Welsh Reply

      Is it conceivable that anyone who is going to die as a result of being “vaccinated” will do so after the first shot – leaving a smaller vulnerable population after the second?

  6. Frank Reply

    Hello John B. I completely agree that the vaccine passport system is absurd since vaccinated persons get infected and transmit the virus at a rate not dissimilar to unvaccinated persons. I do not think the vaccine passport system is about public health, but about social control. They are laying the ground work for a system of social control to remain long after COVID is no longer a serious threat. Enjoy your day and thank you.

  7. Peter Hickey Reply

    Does Dr No think the rather high SMR figures (many > 90 per 100,000) for deaths after 21 days in the vaccinated group in table 5 of the ONS data set supports his findings ? Unless I’m mis reading this data it seems to suggest that over the long term, the SMR for the unvaxxed is indeed significantly lower than the vaxxed.

  8. dr-no Reply

    Thank you all for your comments.

    John B – your point about the different age structures, and the implications therefrom are important, so much so that Dr No almost included a chart showing the differences, and yes, direct standardisation (as used by ONS) does start to fall apart when the number of events (deaths) is very low, which is why it is a pity there is no explanation of methodology.

    Age standardisation should take care of comorbidities, that’s largely why you do it, because older people are more likely become ill and die, but it can’t take account of (and so standardise for) unknown differences that also affect mortality. Take the classic case of comparing hospitals using SMRs. Hospital A has a much worse SMR than hospital B, and the Can’t Quite Cope goons (the Care Quality Commission) decide hospital B needs to go into special measures, obvs, because hospital B is failing, its care is rotten etc. But the care in hospital B is reality every bit as good as that in hospital A, it is just that hospital B is in a very deprived area, and hospital A is in a very affluent area, and so, age for age, hospital B’s patients are invariably sicker than hospital A’s patients.

    This question – different underlying risk – is very pertinent to the business of comparing SMRs between different vaccine status groups. We have to ask: is there any reason to suppose that, apart from age differences, the three groups, unvaccinated, 1 dose, and fully vaccinated, differ in their underlying risk? Dr No can’t think of an obvious one, all the more so given the vaccines were rolled out across the board for each group as they became eligible. Why might someone who has only had one jab have a higher non-age related underlying risk of death than someone unvaccinated or fully vaccinated? It is worth noting this higher mortality is also very visible in ONS’s non-covid ASMRs, see chart below. Is there anything, apart, from the jab, given that age is taken care of, that can explain that strikingly worse mortality? If nothing comes to mind, then – Holmes’s dictum – we have to entertain the possibility that the vaccine is to blame.

    Tom – you are right, but the caveat about comparing populations applies. It is important to remember what indirect standardisation (which SMRs use) does: applies rates from a reference population (in this case, England 2019 rates) to the age structure in the populations of interest (different years, different hospitals, different vaccine status etc). For each population of interest, we compare the observed number of deaths to the expected number. This means that strictly speaking, the only fully valid comparison is between the population of interest and the reference population, and you can’t compare the populations, even if sometimes we do! Direct standardisation avoids this problem, but requires age stratified deaths, which ONS have not seen fit to release, so we are forced into using indirect standardisation/SMRs.

    Peter – yes, on the face of it they do, by and large, but frankly their ASMRs (SMR conventionally means standardised mortality ratio, and so indirect standardisation, ASMRs are age standardised mortality rates, sometimes DSRs, directly standardised rates, and these use direct standardisation) are all over the place. The curious thing is ONS published the non-covid deaths ASMRs in the dataset, but made no use of them in the report. One can’t possibly imagine why. The plot over the 26 weeks from ONS’s Table 5 looks like this:

    With the caveat that this is only non-covid deaths, even if in the second half of the period, covid deaths make relatively little contribution, compare numbers (counts) between Table 4 and 5 in the report (eg in week 26, 111 deaths incolving covid cf 6,845 non-covid deaths), we can see (a) the rise, and sustained high ASMR in the 1st dose >21 day group (yellow line, Dr No’s Vaccinated 1st dose SMR combines this lot with the red line, and this lot has the worst SMR) and (b) a steady rise in ASMR for the fully vaccinated, which appears to be converging with the unvaccinated. Make of this what you will…

    The other curious thing (in addition to not publishing the raw age stratified numbers) is why break the data down into covid and non-covid data, and then only use the covid deaths data in the published report? In any sort of study looking at the effect of an intervention, it is a given that we need to look at all cause mortality, so as not to miss the ‘unknown unknowns’. Lets say drug X treats aggressive prostate cancer, and does it very well, so prostate cancer mortality falls in those treated with drug X. Get that press release out today! Unfortunately, an unknown unknown at the time was that drug X also drills holes in your stomach, and you bleed to death. You only get to find that out if you look at all cause mortality. Yet ONS’s report does the equivalent of reporting only prostate cancer mortality. To be fair, some non-covid mortality data is in the underlying data, but how many MSM hacks and ticks bother to look at the underlying data? Of course it could never be the case, but it almost looks as if ONS want to bury the embarrassing all cause mortality, and instead focus on the tractor production is up message.

  9. Peter Hickey Reply

    Thanks for update Dr No. Looking at the raw data for “2nd dose” can’t help thinking something is definitely wrong here. Firstly the death rate per 100,000 seems unrealistically low (even though it is slowly rising), as seen in Dr No’s graph. Secondly, the total number of 2nd dose deaths in table 1 of the supporting document doesn’t match the numbers in the raw data viz:- Deaths involving Covid Table 1 610 – raw data 578 Non Covid deaths Table 1 68,733 – raw data 60,859. Furthermore, as already alluded to by Dr No, the total all cause mortality figures for the same period from the weekly ONS mortality stats are 304,768 while the total deaths in table 1 are only 265,982. Maybe I’m missing something but I have written to ONS to seek an explanation as to why these overall death totals are so different.

  10. Helen McArdle Reply

    This feels like one of those puzzles which you spend ages on only to find out there was a misprint somewhere, and it was never solvable.

    I haven’t got any answers, only observations:
    1. Like you said in your post and Peter Hickey observed, the numbers don’t seem to add up. The dataset note 8 says that there can be a delay in death registration and links to a page explaining why (violent deaths, suicides, post mortems etc). However, when I added up all the non-Covid deaths in table 5 (184398) and the Covid deaths in table 4 (43956) I get 228354. Whereas in the other table you link to (also ONS data) there are 285929 deaths (not sure I have added these correctly, as Peter quotes 304768). So by my count 57575 deaths are missing, by Peters 38786 . I can’t accept that there are that many delayed registrations, given everything is electronically done these days, and I assume the weeks 1-26 tally in both tables. And this is a different number again from table 1 (265982), which gives the non-Covid deaths as 214701 and Covid deaths as 51281, so more than I counted by adding up all the columns in the Table 4 and 5 excel spreadsheets.

    2. Regarding John B’s observation, I thought the same – I wondered how you can age standardise a population if the populations aren’t static, but evolve over the time period you are measuring. If you look at travellingtabby.com for the Scottish data, apparently 100% of the over 60s are now double vaccinated. This is nonsense of course as I have patients over 60 in Scotland who haven’t been vaccinated. But I guess they are in such low numbers that they don’t count. In week 1 only 2.6% of the Scottish population (113459) had had 1 dose and only 36 people had had 2 doses in Scotland. And they were pretty much all in care homes. (England were clearly much faster than us as they had apparently done 267629 double doses by week 1 according to the ONS table). By July 2nd, 70% of the population had had 1 dose and 50% were double jabbed, and it was mostly the lower risk people unvaccinated. So the denominator in each of the UV/SV/DV groups changed over time, so that by some point there were only a tiny number of very old and very vulnerable in the unvaccinated group. Older people at risk of any death were mostly in the double vaccinated group by then.

    3. I also therefore agree that the double vaccinated Non-Covid death rate is implausibly low compared to the others and from your Table 5 graph suspiciously flat (of course we know why it starts really low, as there were very very few double jabbed in the UK by week 1, but why does it stay so low?).

    4. In the definitions table they separate 2nd dose into under 21 days and over 21 days but in tables 4 and 5 they just have 2nd dose. In table 1 the Covid deaths after 2nd dose is 640 (182 21d) but in table 4 the covid deaths after 2nd vaccination add up to 578. In fact all the numbers in table 1 are different to the numbers of deaths in table 4. Weirdly the difference is 5863 higher covid deaths for unvaccinated, 531 for SV 21d and only 62 for DV. It almost looks like they shoved a load of covid deaths in the unvaccinated group for want of anywhere better to put them ;-)

    4. The total population. I had expected that over time the total denominator population of UV, SV and DV would remain fairly constant if as the older people died, younger people turned 18 and were eligible for the vaccines. However we hadn’t started vaccinating 18 year olds so it looks like they weren’t added to the denominator population. Thus the total population in week 1 was 39359819, but by week 26 it had dwindled to 39245327 (114492 less). Apart from a boost to the population denominator in weeks 9-11 it was a gradual decline in numbers. And yet there were 228354 deaths from the original denominator population, so it does look like they did add some, because the numbers don’t balance. The new denominator for the unvaccinated would be younger proportionally than the denominator in week 1 of Jan. Would this have an impact on the figures?

    5. In my experience those that take 1 vaccine but not 2 are sometimes those who had unpleasant reactions to the first vaccine- often younger people who aren’t too scared of Covid so decide on balance not to take another (older and other highly motivated people have taken the second vaccine even where the first jab caused serious reactions). Also those who got Covid after the first vaccine sometimes didn’t bother with a second. And those who died before their second vaccine.

    Thanks for another great post. I think they just release this data to mess with our heads, and some ONS statistician somewhere is gleefully sitting on a pile of lost and hidden numbers.

  11. dr-no Reply

    Peter and Helen – thank you, pertinent observations all. It is almost as if the ONS producers of the data are unaware that down the corridor someone else is producing routine weekly death counts with very different totals. One of the best ways of verifying data is by checking with alternate sources, but that seems to beyond ONS’s ken. It will be very interesting to hear ONS’s response (if indeed there is one).

  12. Helen McArdle Reply

    One other possibility is that the system for measuring vaccination status may be wrong. In Scotland, many of my colleagues have reported that their 2nd vaccine has been recorded as a 1st vaccine. Their ‘vaccine passport’ is invalid, as it records them as only having had one vaccine.

    This might quite a big system glitch: https://www.heraldscotland.com/news/19606119.covid-missing-vaccine-passport-leaves-fully-jabbed-falkirk-chemist-400-pocket/ s

    ‘…a call handler on the Covid Certificate helpline told him they were now referring thousands of cases a week on to the specialist ‘resolutions team’ set up to iron out glitches in the system. The Scottish Government refused to disclose how many cases have been logged with the resolutions team to date, saying it does not “currently publish those statistics”.’

    I believe it all has something to do with the new vaccine IT recording system, which is separate from the GP record. If something similar has happened in England then the whole thing is a mess, and would create a glut of people recorded as single vaccine>21 days who are actually double vaccinated.

  13. Helen McArdle Reply

    That Scottish Herald reporter is no relation BTW.

    Quote from another article in August,
    ‘A retired IT director from south London, claimed he made more than 100 attempts over five months to correct his record to show he was fully vaccinated after realising that his first jab, from February, was missing. “Two weeks ago, I decided I had had enough, so I took drastic action. I went and had a third vaccination,” he told the Telegraph’. 

    Looks like the vaccine recording system is a mess in all over the UK.

  14. steve Reply

    Incompetence runs far and wide and whilst this may not be directly related to the original thread, this is what I had to say to Track and Trace a few moments ago. Multiple cock-ups, multiple dangers…

    Date: 29th Sept 2021

    The poor track and trace person kept asking me the same questions that I had already uploaded on-line about my daughter.

    She mentioned she didn’t have access to all the data!

    Well okay, but now we have shared very personal information multiple times and there are now multiple places this data could be extracted or compromised.

    To say the data is secure is almost meaningless because errors happen, and the government is as guilty as anyone in the private sector at letting out data which it shouldn’t.

    By the way, I’ve spent my life around IT and know that bad stuff happens despite the security.

    Thanks and regards

  15. Helen McArdle Reply

    Public Health Scotland have produced their equivalent stats: https://publichealthscotland.scot/media/9330/21-09-22-covid19-publication_report.pdf

    But see appendix 9:
    Vaccination status for all individuals who test positive for COVID-19 by PCR is extracted from the data used to produce the PHS vaccine uptake/daily dashboard. Vaccine records include the number of doses and date of vaccination. Individuals are listed as unvaccinated if there is no vaccination record linked to their unique CHI identifier at the time of analysis. Vaccination status is taken at date of specimen for COVID-19 cases, acute hospital admissions, or death and assigned to number of doses according to the case definitions described below.
    COVID-19 vaccination status is defined as per the following:
    Unvaccinated: An individual that has had no doses of COVID-19 vaccine and has tested positive for COVID-19 by PCR or has had one dose of COVID-19 vaccine and has tested positive less than or equal to 21 days after their 1st dose of COVID-19 vaccine.
    50
    • Dose 1: An individual that has had one dose of COVID-19 vaccine and has tested positive for COVID-19 by PCR more than 21 days after their 1st dose of COVID-19 vaccine or less than or equal to 14 days after their second dose of COVID-19 vaccine.
    • Dose 2: An individual that has had two doses of COVID-19 vaccine and has tested positive for COVID-19 by PCR more than 14 days after their 2nd dose of COVID-19 vaccine.

    Imagine a scenario whereby Covid vaccines temporarily increase risk of Covid infection and death, let’s say within 21 days of the first vaccine (not a completely implausible immunological phenomenon). Using the above categorisation, these cases would be officially reported as ‘Unvaccinated’ and we would never spot it.

    If we add this definition of vaccination status to the vaccine IT recording glitch, is it possible that all these numbers are just numbers and wishful thinking?

    • Peter Hickey Reply

      Your scenario is indeed very plausible Helen. I believe it is widely acknowledged that the immune response is dimished during the two weeks following vaccination and it is therefore very worrying that harms caused by the vaccine during the first weeks will be misattributed as unvaccinated. I note the definitions are from Public Health Scotland and these appear to conform to the globally decreed vaccination status definitions. It’s not entirely clear whether Public Health England are using same criteria but according to the definition statement in the data set it seems not in the data we are looking at, ie

      Vaccination status is determined on the date of death occurrence if a death has occurred, and on the last day of each week if not. Possible values are

      unvaccinated

      vaccinated with 1 dose only, date of death/last day of week is less than 21 days after vaccination

      vaccinated with 1 dose only, date of death/last day of week is at least 21 days after vaccination 

      vaccinated with 2 doses, date of death/last day of week is less than 21 days after second vaccination

      vaccinated with 2 doses, date of death/last day of week is at least 21 days after second vaccination 

      In any case it is clear that obfuscation abounds, making it extremely difficult to see clear evidence of what many suspect, ie that these vaccines are indeed harmful. I guess the question is this subterfuge or just plain incompetence.

  16. Steve Prior Reply

    “I guess the question is this subterfuge or just plain incompetence”.

    Having seen lots of video channels taken down and the notices which YouTube etc have sent to channel owners, it is clear that anything outside the mainstream narrative is not allowed.

    For example, no one is allowed to suggest Vitamin D3 as a possible mitigator. Last time I checked, vitamin D (a secosteroid) is created by all humans to varying degrees. Sun hits the skin, cholesterol in the skin starts the process, followed by some nifty work of the Liver and Kidneys.

    The focus of this and other Governments is all about total control.

    They say follow the science, but what they really mean is follow our science and our scientists. They are the only ones getting air time, anyone with an opposing view is left in the margins.

    I spoke to my doctor before refusing this jab. He was all for it and said he had had it. But he couldn’t answer my questions, which I believe should be part of our informed consent.

    My suspicion is the vast majority of people didn’t give informed consent, but simple read a leaflet and believed they gave informed consent.

    I get the sense there are doctors here who are somehow thinking more. So, why didn’t my doctor think a bit harder instead of jumping on the jolly old bandwagon!

    My answer to the question “”I guess the question is this subterfuge or just plain incompetence”, it’s mostly subterfuge with a huge smattering of the underlying structure which drives behaviour.

  17. dr-no Reply

    The vaccine status definitions do sort of make sense if you are only interested in covid events once presumed immune, though it is not clear in the ONS English data whether the Second Dose columns in Tables 4 and 5 means all second doses (definitions 4 and 5 in the list of definitions quoted by Peter above) – presumably they do. The English 1st dose deaths data is is a little clearer, divided as it is into under and over 21 days post vaccination, with the former showing much the same mortality as the unvaccinated, while the latter show a steep rise and then sustained high level from around week 13. Very curious: why should this be so?

    Correlation is not causation of course, but Dr No wondered if it might be something to do with the differential roll out by vaccine brand. One might imagine weekly vaccine doses by brand was readily available, but it is not. Instead, it is buried in the Yellow Card reports archived in the Wayback Machine here. Dr No has gone through these reports and pulled out the figures (a very tedious and boring but worthwhile process) and has plotted cumulative weekly 1st doses by brand (left Y axis) and ASMRs for non-covid deaths >21 day after the 1st dose (right Y axis):

    Correlation is not causation, but it is nonetheless striking that the rise in AstraZeneca Vaccine use is mirrored a few weeks later by a rise in ASMRs.

    Subterfuge vs incompetence? Like the ever wonderful Prudence Kitten on Dr K’s blog, Dr No by nature generally favours a version of Hanlon’s Razor (“Never ascribe to malice that which is adequately explained by incompetence”), but there have been times lately when even the sharpest of razors starts to dull on the sheer volume and density of the obfuscation routinely on display…

    Edit 11:52 30 Sep 21: text and chart title edited to make it clearer the weekly vaccine counts are cumulative

  18. Peter Hickey Reply

    Update:- ONS have responded to my query regarding the apparent anomalies in the deaths by vaccine status publication.

    From their reply, it appears the figures as published in table 1 are derived from data contained within Table 13 of the monthly mortality analysis data:- (see abbreviated link)

    http://bit.ly/3lri9Yx

    This data is based on the date of death occurence, rather than date of registration, which is the criteria for the weekly ONS death registrations report.

    Additionally, the top level weekly data in the registrations report is for England and Wales, while the Table 13 data in the monthly mortality report is for England only.

    So in a nutshell, we see:-

    265,982 deaths in table 1 of the deaths by vaccination status report

    270,138 in table 13 of the monthly mortality stats.

    286,373 in the weekly ONS death registrations report for the same period.

    I guess that as all of these stats are a bit of a moving target, ie new registrations coming in all the time, probably with dates of occurence outside of the time period in question, the above differences could be possible.

    With regards to the difference between the summation of the different categories in Table 1 and the corresponding totals in tables 4 and 5 of the associated data set, it seems that in order to calculate age standardised mortality rates, ONS use a subset of the England population which is based on the number of people over 10 years of age, who are still alive and were included in the 2011 Census. This works out to about 79% of the current population and this difference appears to be spread fairly evenly across the different vaccine status, supporting Dr No’s original assumption.

    So although nothing here takes away from Dr No’s conclusions, it certainly helped my understanding of how ONS arrived at the figures.

    I am still awaiting an answer to one further question to ONS which I summarise as below:-

    So to clarify, my question is does NHS England use the same criteria as Scotland to determine if a person is designated as unvaccinated, or is the actual date of the vaccination used as the vaccine status marker

    Finally I have come across the following NHS vaccination archive for those interested in further research:- http://bit.ly/2WWXja5

  19. Peter Hickey Reply

    Further to my last, may I suggest plotting the daily deaths by date of occurence, as derived from table 13 in the ONS monthly mortality stats:- http://bit.ly/3lri9Yx

    This chart provides a rather different perspective to the weekly deaths registration chart that certainly I and maybe many of us, are used to seeing.

    Look at the date the second wave starts to begin and then look at the age cohort of those most impacted by this second wave.

    I have done this and a certain correlation is quite marked. I would be very interested to see if anyone else notices the same pattern or am I barking up the wrong tree ?

    • dr-no Reply

      Peter – it’s a bit more spikey, weekly vs daily counts smooths things a bit, but nothing major jumps out, either for all cause deaths or alleged covid deaths. Table 13 doesn’t have deaths by age, but using the weekly figures (which do), again nothing stark jumps out (rank order of the age bands pretty constant). Could you perhaps elaborate on the marked correlation you saw? If you have somewhere online that can host images, you could always upload a plot and link to it from here.

      • Peter Hickey Reply

        Thanks Dr No – It’s the date of the commencement of the second wave, which shows quite clearly on daily chart. This appears to strongly coincide with the start of the first phase of the vax roll out. The end of this wave occurs around the time that the vast majority of 65 + had been vaxxed. When looking at the age group of those most affected by the second wave it appears to be the 65+ that are most affected. The plot for the 65 and under seems pretty flat over the same period. I appreciate correlation is not causation and that the older ages are more vulnerable to respiratory and other illness but it’s just something that struck me as quite a coincidence. It does also seem to tie up with the report from the undertaker in Milton Keynes. I strongly believe all is not as is made out to be and keep looking for a signal in the data that can account for all the anecdotal reports I keep hearing about re the adverse effects of the jabs. However it’s like looking for a needle in a hay stack, with the signal being the needle and the numerous diverse sources of data and unrevealed data being the haystack. I’m no data expert and have no wish to draw conclusions that aren’t warranted hence appreciate your opinion on my thoughts. I can post the charts via links later if need be.

        • dr-no Reply

          Peter – ‘I strongly believe all is not as is made out to be’ as do many with open sceptical minds (open sceptical is not an oxymoron, they are natural bedfellows, even necessary bedfellows). It’s not helped by the occasions when ONS appears to be morphing into the ONO (Office for National Obfuscation), or PHE becoming the UK Health Security Agency (a very different name, and so presumably a very different remit and perhaps powers too). But we have to be very careful not to over-speculate, and present the over-speculation as conclusions. That said, speculation is fine (even necessary, if we are to generate hypotheses), but it must always labelled as such. Alternatively, at one level down, we can just present some data and say ‘that’s curious’ (and so implicitly invited the reader to speculate). But we have to remember none of this proves anything, though in the fullness of time observations and speculations may lead to something more substantial.

          Dr No’s plot of the daily deaths from Table 13 does show the second wave taking off in early Dec 2020, just as vaccinations started, but only just over three quarters of a million people got their first dose in December 2020, and of those, two thirds, just over half a million, were over 80 years of age. The second wave then peaks around 19th January. At the same time, the number of people vaccinated rose to 3.5 million, of which 2 million were aged 80 and over. The second wave does coincide with the start of vaccination, but by mid January 2020, the second wave started to decline, just as vaccine rates started to increase – in other words, the opposite was happening: decreasing deaths coinciding with increased vaccination rates.

          We’re still stuck with our old buddy correlation is not causation, and it really does apply here, because there are just so many possible confounders. Without a breakdown of deaths by age and vaccination status it is impossible to further tease out what went on (the ONS deaths by vaccination status data (in)conveniently only starts in January 2021). For Dr No, the most curious/striking plot remains the one in this comment which shows a steep rise in ASMR in the first dose > 21 days group soon after the rapid roll out of the AZ vaccine. As ever, correlation is not causation, but the rise is striking, and the curious will naturally want to know more about what is going on.

  20. dr-no Reply

    Peter – thanks for the feedback.

    We don’t know whether ONS deliberately use obfuckstration, or are just boringly illiterate, but, however they got there, their explanations of methodology are always as clear as mud.

    To calculate an ASMR (age standardised mortality rate) using the direct method, which is what ONS have done, you need two things:

    (1) a standard population, with numbers/proportions of individuals in each age band. The choice of this standard population is somewhat arbitrary. In theory we could go Roman and use P. Sulpicius Quirinus’s census of Syria and Judea in circa 6AD (as Clive James would have said, P. Sulpicius Quirinus will always be with us. Quirinius perennius), but more conventionally one of the standard populations set up for the purpose is used. ONS used the 2013 European Standard Population, a standard population very commonly used in epidemiology.

    (2) age specific mortality rates for the same age bands from your population of interest. You then apply these rates to the standard population, age band by age band, and so get your ASMR.

    This creates problems, especially when numbers of deaths are small, and perhaps zero, in certain age bands. Furthermore, to know the rate, you need to know both the number of deaths in each age band (the numerator) and the number of people at risk (the denominator) in that age band. It seems this part of the data comes from the oddly named Public health Data Asset (why Data Assest rather than Data Set?) because it allows linkage via an individual’s NHS number with vaccine status data from NIMS. From ONS’s methodology section we have:

    “Age-standardised mortality rates are calculated for vaccination status groups using the Public Health Data Asset (PHDA) dataset. The PHDA is a linked dataset combining the 2011 Census, the General Practice Extraction Service (GPES) data for pandemic planning and research, and the Hospital Episode Statistics (HES). We linked vaccination data from the National Immunisation Management Service (NIMS) to the PHDA based on NHS number, and linked data on positive coronavirus (COVID-19) Polymerase Chain Reaction (PCR) tests from Test and Trace to the PHDA, also based on NHS number.”

    So that is how they got their denominators, stratified by, among other things, age and vaccination status. Never mind that the ‘asset’ is based on a census that is 10 years old — not quite Quirinus perennius, agreed, but you get the drift, or that the GPES ‘data for pandemic planning and research’ is none other than the hugely controversial GPDPR, better known as the NHS Data Grab that was in the news over the summer, and from which millions have opted out, and which does not even appear to be up and running yet (no “specific start date“) or that the use of hospital episode based Hopelessly Erratic Statistics is always an amber if not a red flag. Let’s not even get started on Not In My Surgery part of the vaccine rollout, and just accept they have some denominators. What is even less clear is how they got their numerators, ie number of deaths, in each of the bands. In fact, it is as clear as mud. We have to assume they used some form of linkage – NHS number, perhaps? More opacity. Even the formula they use looks impenetrable (though it probably isn’t, at least to those with Bletchley Park code breaking skills):

    Why on earth use ‘mid dot’ symbol, rather than the infinitely more familiar “x” to represent multiplication?

    One way of getting round these problems (of needing to know mortality rates across all age bands and categories, sometimes with very small numbers) is to use indirect standardisation. Traditionally, direct standardisation is seen as the purer approach, the Escoffier recipe, with indirect standardisation more in Jamie Oliver style, but it is still a valid methodology, which only needs the total number of deaths in each vaccine category, which removes the band specific small numbers problem, and the proportion of the population vaccinated at the time of interest, which is available from the NHS vaccine archive.

    The other important thing to remember is that the ONS report focuses exclusively on ‘deaths involving covid’. All cause mortality doesn’t even get a mention. By focusing exclusively on covid deaths, ONS automatically makes its data and so its analysis, messy, because there is no way of knowing how many of the alleged ‘deaths involving covid’ were actually deaths caused by covid, and this brings with it significant risks of bias. Using indirect standardisation on all cause mortality circumvents these problems, which is why Dr No used it.

    The bottom line still remains that we don’t know (a) exactly how ONS calculated its ASMRs and (b) how reliable its data is, given the missing deaths. If the missing deaths are not randomly distributed. there is ample opportunity for bias, and the only reasonable conclusion is that ONS’s findings fall a long way short of being reliable estimates. Dr No’s SMR calculations avoid some of these problems, but is still at risk from the random missing data assumption. The only satisfactory solution remains unchanged: ONS needs to publish the raw data.

    PS wrote the above yesterday, before seeing your latest comment – will have a look at the daily deaths shortly.

  21. Peter Hickey Reply

    Many thanks for the reply Dr No. I absolutely agree that we must be very careful not to fall down the trap of “confirmation bias” and that correlation does not equate to causation. I did try to clarify this in my previous post viz:-

    “I appreciate correlation is not causation”

    ” have no wish to draw conclusions that aren’t warranted”

    However, while correlation does not equate to causation, it can certainly act as a red flag that warrants further investigation.

    The onset date of the second wave and it’s coincidence with the commencement of the vaccine roll out is indeed quite striking. What further attracted my attention was that the majority of the over 65s were vaccinated during the period of this second wave. This can be seen quite clearly as the markedly steep rise in that age group within the following chart, which is an extract from the UKHSA weekly influenza and Covid 19 report:-

    http://bit.ly/3DFYaf8

    Furthermore, when analysing the age breakdown of deaths during the period of this second wave it is interesting to note that this second wave stands out very clearly for the 65+ cohort but is virtually undetectable in the below 65 cohort. There is indication of this in the following chart, which is derived from the ONS weekly death registrations:-

    http://bit.ly/2YMRNal

    A further point of interest when analysing the second wave by age group on the weekly date of registration chart is that the end date of the 2nd wave for the over 80s appears to occur approximately 4 weeks earlier than it does for the 65-80 cohort. This also happens to coincide with the fact that the over 80s were nearly all vaccinated well in advance of the 65-80 cohort.

    Dr No does indeed make a very valid observation, ie that no other wave occurs following the further rollout of vaccines to the below 64 age group. However, it must be noted that it is after this further roll out, which is much slower and far more gradual than the first rollout, when we start to see the curious steep increase in SMR that Dr No has already pointed out.

    Dr No also mentions that the ONS Vaccine status report only commences from the beginning of January 2021, whilst the vaccine roll out program commenced during the week ending Dec 11th 2020. From the daily mortality report I estimate nearly 44,000 deaths occurred during the period from Dec 11th to Jan 2nd, of which approximately 6000 might be considered excess deaths when compared to the 5 year average for the same period. How many of these were in the vaccinated as opposed to the unvaccinated we might ask ?

    I hope Dr No does not see me as drawing any concrete conclusions from my observations, merely that I do indeed find them curious coincidences and worthy of further debate and consideration. I can’t help but think that open publication of the simple crude death rate by age cohort against vaccine status (actual date of vaccination rather than 14 or 21 days after this date), for all those that have unfortunately died, would go a long way to answering our questions. I’m certain ONS must have access to this data.

  22. dr-no Reply

    Peter – ‘I hope Dr No does not see me as drawing any concrete conclusions from my observations, merely that I do indeed find them curious coincidences and worthy of further debate and consideration.’ Absolutely, indeed Dr No used a similar approach in his latest post (plotting country level covid vaccination rates against covid ‘cases’ and deaths). The need for caution is mainly for readers new to the topic, and/or perhaps without a scientific background, who might unwillingly but understandably fall into the post hoc trap. It is also to pre-empt any critics who say we are over-interpreting the data. These charts generate questions that may or may not be answerable, not answers that can’t be questioned.

    You are also right to suggest that by far the best way to attempt to answer these questions is to look at age and vaccinated status stratified overall all cause deaths data for the entire period. In a way breaking down the deaths into covid and non-covid is a distraction, because we have no way of knowing how many of the ‘deaths involving covid’ were truly covid deaths, meaning any conclusions based on this breakdown are at best tentative.

Leave a Comment

Your email address will not be published. Required fields are marked *