Vaccination effectiveness: can it be something more?

The last remnants of vaccine effectiveness are being picked out in all kinds of ways. From shifting from vaccinated to unvaccinated (because they did not consent to registration of their data) to comparing subgroups of vaccinated instead of vaccinated versus unvaccinated.

Having confidence in a government when it comes to an emergency injection does not mean that that trust is there if that same government notes it somewhere. Administration is apparently more frightening for 7% of the population than an injection. That group has made it difficult to assess the vaccine effectiveness achieved. Last week, two rivm publications were published. The confusion is great and what emerges from it above all: the vaccines are a tragedy. Or not?

Vaccine effectiveness against hospitalizations

Below is an example graph of vaccine effectiveness as we would like to see it. We recognize the usual quarterly wave pattern of the disease that comes and goes with waves; a fascinating interplay between immunity and virulence. Especially with the elderly, of course. In times of high virulence, a vaccine has the most effect.

What is good to see: the last prick (purple) has the most effect. Then at 80+ we see the repeat jab and the booster. At 60-79 especially the booster, there fewer people have taken the repeat jab. In order to remain well protected, it is therefore important to get your vaccine in time.

This is how we get corona under control together.

So far so good.

Only: I changed the title of the chart, it's not vaccine effectiveness. It's the other way around: what's shown in this graph are the vaccination statuses of hospitalizations per day, in absolute numbers.

Vooral herhaalprikkers liggen in het ziekenhuis met corona. En nou ja, als het merendeel van de bevolking de herhaalprik heeft gehaald, is het niet verwonderlijk dat dat in het ziekenhuis engszins terug is te zien - tenzij die prikken echt supergoed zouden werken natuurlijk. Dan zouden mensen die netjes hun najaarsrondeprik hebben gehaald niet in het ziekenhuis terecht komen. In elk geval niet vaker dan de mensen die dat niet hebben gedaan.

In the media, attention is now mainly drawn to various calculation methods that could increase vaccine effectiveness by a few percent. Maurice wrote something about it yesterday. Below we review another report.

In de bevolking: 7% "non-consent"

In a study published last week¹RIVM in collaboration with Julius Health Center: Information bias of vaccine effectiveness estimation due to informed consent for national registration of COVID-19 vaccination: estimation and correction using a data augmentation model wordt betoogd dat het niet goed mogelijk is om de vaccin-effectiviteit te bepalen omdat niet iedereen toestemming heeft gegeven om zijn/haar identiteit bij de prikdata te registreren. In de ziekenhuizen blijft er daardoor onzekerheid over een restpercentage van mensen met vaccinatiestatus 'onbekend', een percentage dat nog door de tijd kan fluctueren ook.

Door met deze 'non-consenters' te schuiven berekenen de onderzoekers in de studie eerst een vaccinatiepercentage dat veel lager is dan was gemeld in eerdere rapporten. Vervolgens rekenen ze voor hoe je tot hogere percentages kan komen. Die hogere percentages liggen dan weer dichter in de buurt van wat er eerder al was gerapporteerd. Is dit nu kunst- en vliegwerk om opmerkelijke rapportages in de eerdere rapporten alsnog te onderbouwen? Hoe dan ook: er wordt gegoocheld met het non-consentpercentage in Nederland: 7%. En dan ook nog tijdens de Delta-golf, toen de vaccins zich goed konden bewijzen vanwege de virulentie van Delta, eind 2021 alweer. Maar hoe zit het nu, in 2023?

Nu in het ziekenhuis: 2% "non-consent"

Last week also saw the publication of a report by the RIVM COVID-19 epidemiology and surveillance team.²COVID-19 hospital and intensive care admissions by vaccination status in the Netherlands (admissions 03 October 2022 – 01 May 2023) Hierin wordt over het laatste half jaar verslag gedaan van de toestand in ziekenhuizen en op IC's, naar vaccinatiestatus.

The number of non-consenters in hospitals appears to be only 2%

If you include the non-consent group among the vaccinated, this can have opposite effects:

1. In population screening, the total number of vaccinated people is increased. The total number of Covid patients is what it is. The chance of ending up in hospital as a vaccinated person with Covid is therefore reduced.
2. Patients are sick. If the non-consenters in that group of patients move to the vaccinated, the total number of vaccinated sick people is therefore increased. That is actually more chance of admission = worse VE. The chance of admission for corona is increased.

An example: Hospital with 70 vaccinated, 15 unvaccinated, 15 unknown

1. Unknown excluded: 15/85 =18%of admissions are unvaccinated, 82% vaccinated
2. Unknown as vaccinated included: 15/100 =15%of admissions are unvaccinated, 85% vaccinated
3. Unknown as unvaccinated included: 30/100 =30%of admissions are unvaccinated, 70% vaccinated

It is clear that there is something to be gained here for vaccine effectiveness. When drawing up the balance sheet in hospitals, the highest VE is achieved if the non-consent group is counted among the unvaccinated.

It is the scraping together of percentages, but it does not alter the fact that this is not scientifically investigated down to the percent. Ultimately, we would like to know exactly how it works. That's not to say we don't know anything until then.

In reality, there are far fewer non-consenters and more unvaccinated than in this example. In real life, there are much smaller differences. So small that you might wonder if they are relevant in the discussion of vaccines (insofar as discussion is allowed.)

So what are the real differences?

We take a look at the vaccination rates in the Netherlands and then at the vaccination rates among the corona hospital admissions of the past six months, as reported by rivm. The blue bar shows the actual percentage of vaccinated patients among the patients.

De rode staaf is het 'referentie'percentage, zoals je dat zou verwachten op basis van het landelijke vaccinatiepercentage. Dat staat los van of prikken op naam zijn geregistreerd of niet. We weten van een deel van de prikken dus wel dat er mensen zijn gevaccineerd, alleen niet wie. Dat maakt voor de hoogte van de rode staaf niet uit, alleen voor de blauwe.

If we do not correct for non-consent in the hospital and keep it out everywhere, we see that the vaccinated are slightly underrepresented compared to the national percentage. In any case, that is not a negative effectiveness of the vaccines among corona patients.

Twice as many people die outside the hospital as in the hospital. If those are more likely to be vaccinated than unvaccinated, that could explain the difference. Vaccinated people who die suddenly - without a hospital - lower the blue bar. Unfortunately, the data used to verify this will not be released.

For example, if we count 2 percent more vaccinated people than are currently registered (this happens if we consider the non-consent group as vaccinated), then that difference disappears.

At 4%, it is already turning into the disadvantage of the vaccines. As a result, the VE becomes negative. Note: this is only to illustrate the tipping point. In the hospital, it's only 2%.

In short, the differences are minimal. That also reminds us of the facts: the vaccines are nowhere near dramatically reducing hospital admissions. In the hospital, with an effective vaccine, you would expect much smaller, for example, half smaller blue bars. Now that the vaccinated are becoming corona patients in the hospital in about the same proportion as in the population, what is the point of vaccination at all?

So how important is vaccination?

The graphs in the report itself confirm the picture.

In Grafiekje A zien we de aantallen patiënten per vaccinatiestatus. Het valt op dat de paarse lijnen veel hoger zijn dan de andere. Dat komt omdat er veel meer "Herhaalprik najaarsronde" geprikten zijn in de bevolking, bij de 60-plussers. Je kunt de opnames dus beter relatief bekijken, per vaccinatiestatus. Dat is in grafiek B en C gedaan.

Graph C, where the scale has been reduced, is the clearest. We also see here the minute differences that we saw in the bar graphs.

Bij een effectief vaccin zou je verwachten dat de rode lijn van niet-gevaccineerden overduidelijk ruim boven de andere lijnen zweeft met 3, 4, misschien wel 10 keer zoveel ziekenhuisopnames, vooral bij de oudere leeftijdsklassen waarvoor het vaccin was bedoeld. Of denk aan polio, waar die factor nog hoger is. Op zijn minst zou ik in de 'per 100.000' grafiek iets willen zien zoals de paarse lijn in grafiek A, waarmee dit artikel ook begon.

But what do we see?

De rode lijn van ongevaccineerden piept er bij 80+ hier en daar iets bovenuit maar verder verdwijnt die rode lijn achter de overige vaccinatiestatussen van wel-geprikten. Er moet wel stevig ge-ASMR'd en gepersonyeard worden om hier nog iets van effectiviteit uit te persen.

But also look at 40-59: the repeat jabs come to the hospital with corona much more often than the rest! Obvious.

Dit lijkt mij een rampzalig beeld. Artsen of zorgverantwoordelijken die dit rapporteren en vervolgens overgaan tot de orde van de dag zullen ooit toch nog wel eens ter verantwoording worden geroepen? Of moet dat dan gebeuren door het ministerie dat de opdrachten tot uitvoering heeft gegeven...?

The differences have not been significant since Omikron and can be called negligible in the last six months. And yet there has been a lot of vaccination in 2022. In 2022, approximately 8 million jabs were made.

Another graph from the report, where I highlighted the 90-somethings.³The 90+ bars were iconic, while 90+ only covers 1% of the population. That would make a well-identifiable 1% of the population seem like an excuse for jabbing the entire population. Probeer vooral te kijken naar de onzekerheidsmarges, aangegeven met de lichter gekleurde lijntjes. Pas boven de 70 jaar is er enig verschil te zien tussen Niet-gevaccineerd en Herhaalprik. Het 'healthy vaccinee'-effect zou hier ook nog een rol kunnen spelen. Als alleen ouderen die er gezondheidstechisch redelijk voor stonden de najaarsprik nog hebben gekregen, zou dat al een deel van het 'succes' van die prik kunnen verklaren. Nog los van het Fenton effect, where recently vaccinated people are considered unvaccinated.

At 70+, the difference is only beginning to emerge: lower bars and the top margins (2.6) fall under the lower margins in the other age groups.

By continuing to deal with the Delta wave, the impression is created that vaccines are effective for everyone. To the extent that they have ever been, they have not been for about three-quarters of the population: the 70-year-olds, for a year and a half. For those three-quarters, only potential side effects are de facto served.

Als we ooit de vaccinatiestatussen van alle patiënten onder ogen krijgen, dus niet alleen coronapatiënten, dan kunnen pas zinvolle risk/benefit analyses worden gemaakt, ook over vaccinaties rond de Deltagolf en daarvoor... ik houd mijn hart vast. Dat daar het nodige aan de hand is zagen we al in het article by Hans Verwaart. For a long time, there has been no efficacy for the vast majority.

The fuss about percentages more or less effectiveness of corona vaccinations is a mock battle in the rearguard. The effectiveness does not meet the minimum effectiveness requirements for vaccines.

The WHO itself states:

And don't start messing around about the difference between trials and real world, efficacy and effectiveness. They are flut products and it is extremely worrying that our government is allowing a power grab from an industry that does not even adhere to its own codes, an industrial cooperative with the WTO as its governance.

aside

2% non-consent in hospitals, which is 5% lower than the national percentage. How is that possible? Options:

1. In the hospital, two-thirds still reveal their vaccination status. My guess is that most will be happy to tell the doctor that they have done everything they can to prevent Covid. Most people are going to talk anyway when it comes down to it. (So: non-consent = actually all vaccinated.)
2. Unvaccinated people do not want to tell the doctor that it is their own fault that they got corona. (So: that 2% is unvaccinated but has registered non-consent.)
3. The vaccines work quite well because of a subgroup of 7% vaccinated non-consenters, less than a third end up in the hospital. That's what you might expect from a somewhat effective vaccine with a high vaccination rate.
4. Young people in particular have refused and they do not come to hospital with Covid. But ages of non-consenters are unknown because not registered so we don't know.

We do not know what happened to those 5%, whether they were not included or made their status known after admission. We don't know if they didn't jog either. We just assume they were vaccinated.

There are unvaccinated people who have taken the trouble to report their non-consent. They did not want to be identified as jab refusers. If 2% of non-consenters come from the unvaccinated 20% of the population, then 1 in 10 unvaccinated people should have taken that trouble. It seems unlikely to me. Perhaps Maurice could gauge that?

Footnotes

• 1
  RIVM in collaboration with Julius Health Center: Information bias of vaccine effectiveness estimation due to informed consent for national registration of COVID-19 vaccination: estimation and correction using a data augmentation model
• 2
  COVID-19 hospital and intensive care admissions by vaccination status in the Netherlands (admissions 03 October 2022 – 01 May 2023)
• 3
  The 90+ bars were iconic, while 90+ only covers 1% of the population. That would make a well-identifiable 1% of the population seem like an excuse for jabbing the entire population.