More Nivel worries

The revised Nivel report loosens up again. Of course, people are once again saying that vaccines have been shown to work, but with such bizarre figures and shaky methodology, nothing can be proven at all. Perhaps it could say something about the patient population of those 400 general practitioners, but even then they have to start over. But first something else that I don't hear anyone talking about.

The privacy trump card

In the Netherlands, it is not necessary to Explicit consent before you enter data Anonymized used for scientific research. This is because anonymized data can no longer be traced back to an individual, which means that it falls outside the scope of the General Data Protection Regulation (GDPR).

Even if you have objected to the use of your data, i.e. have given non-consent to the question 'may your data be used for scientific research', your data may still be used anonymously. If data can no longer be traced back to a person, that data does not fall under the GDPR nor under the EU privacy guidelines. And on top of that, there are also exceptions in science and statistics.

An exception has been made for the Covid vaccines. Due to privacy reasons, there is still no meaningful data. To combat this, a 'nonconsent' option has even been created. Almost no one knows why. It is exactly in line with not serializing the vaccination vials. As a result, it is no longer possible to trace how they got from the incubation tank to the arm of the endangered patient.

So, if something turns out to be wrong with the stuff later, it is impossible to find out how they were treated and for privacy reasons, the healthcare industry and government can keep the patient cards close to their chest.

If some cards do need to be laid (WOO requests), the data is first neutered with data obfuscation techniques.

While anonymizing while retaining information is not that difficult after all. The covid data is now deliberately impoverished by offering them unreadable for computers in low resolution, and by adjusting dates of death in such a way that they can no longer be linked to a vaccination date. So you still don't know anything.

However, anonymization can be done very easily without sabotaging a research goal.

• We do not examine newborns so the date of birth and time can be removed, age on January 1 is sufficient
• For everyone over 90, a randomization of plus/minus 1 year on the date of birth is applied.
• The date of vaccination and the date of death are crucial. On days with very few deaths, these together may be unique to 1 individual and therefore traceable. This link can be broken by shifting both dates by the same number of days, for example with a random number between 7 and -7. Or if necessary 14 and -14. An anonymization specialist probably knows more tricks.

None of that happens. Privacy is far too important to doctors. After all, if anyone should be trusted by patients, it is the doctor. Oh woe if that basis of trust disappears...

Vaccinated people could therefore indicate that their data should not be used, no one knows why. I wonder to what extent you can ban science from looking at certain facts. It would mean the death of science - assuming it is still alive. The Nivel report shows little evidence of this.

De privacy-move heeft tot gevolg gehad dat de non-consenters deel uitmaken van de 7% die niet in het vaccinatieregistratiesysteem CIMS staat. Er is dus niet een aantekening opgenomen "non-consent", nee, er is gewoon niets ingevuld. De prik ontbreekt, inclusief patiëntinfo. Overleden of niet? Onbekend. Iets fout gegaan bij het inlopen van de registratieachterstand? Zo'n percentage van 7% maakt het onderzoek al waardeloos, zeker met het oog op het onderzoeksdoel.

So much for data management. Once again.

But: Nivel to the rescue, they were able to limit the disaster!

The 'small percentage' (7%, as mentioned a disastrous data gap) of missing vaccinated people has been partially recovered. This probably included data from people who had not given permission to the GGD to make their data available; After all, doctors are very concerned about the privacy of their patients.

Anyway, through the general practitioners, personal confidants of those non-consent patients, their data could still be retrieved. In their practice records, they actually found patients who had reported their vaccination status to their GP and were missing from the CIMS register.

The patients knew that this could happen because there are posters and flyers in the waiting rooms of the affiliated Nivel doctors that provide information about this. That's tacit consent or something - yes, I'd like one of those. That information should be in every waiting room because anonymized data can of course always be used.

Non-consent and privacy concerns are played out very selectively. If vaccination readiness is at stake, they weigh very heavily. Vaccination promotion simply avoids it.

The part of the patients whose vaccination status is not reflected in the GP data could still be contacted to update their file. The doctors have that data. Then the non-consenters will come out on their own - or they will still participate because the GP asks them to. This could supplement or perhaps complete the data. After all, there is no problem with non-consent among the general practitioners themselves. But it's not worth it without WLZ data (see below). But nice for an internal bulletin among general practitioners. Doctor & Auto, Medical Contact?

• September 2022: Minister of Health, Welfare and Sport, Ernst Kuipers, requests VWS data from RIVM and CBS. They refuse 'for privacy reasons'.
• September 2022: Kuipers Data Privacy Impact Assessment (DPIA) to investigate this (source)
• February 2023: The Dutch Data Protection Authority sees no objection to the sharing of anonymized data (source) (Waarom ever dat 'assessment'... (source))
• April 2023: Kuipers has taken note of the data. New position: more openness about the corona approach endangers the functioning of the state. (source)

Representativeness

The study was conducted on the patient population of 400 general practitioners, totaling almost 1 million people. The vaccination rate there was 84%, which is slightly higher than the national vaccination rate of 82%. Relatively healthy people, after all, are vaccinated more often. That may also be true because the WLZ (nursing homes, for example) are not included, those residents are not registered with general practitioners.

But wait a minute: although there is only a very small part of the population in those nursing homes, almost half of the mortality comes from there! So you leave out a crucial driver of excess mortality into your research into the link between vaccination and excess mortality.

This research population is, in view of the subject, rather unfortunate in its choice.

Withdraw then, I would say. At least if you pretend to be able to claim something about the total population.

Determination of excess mortality

The fact that Nivel realised that they were analysing a special, non-representative group is also evident from the new mortality probabilities that they determined themselves, both for unvaccinated and vaccinated people. Apparently, the normal CBS mortality probabilities were inadequate.

In other words, the study did not use national averages. There is something to be said for that:

• The national average also includes WLZ (particularly nursing homes), so the expected NIVEL mortality rate should logically be considerably reduced.
• Furthermore, the vaccinated are slightly healthier than the average (because that includes the unvaccinated), so that risk of mortality could be reduced even further
• For the unvaccinated, the expectation must then be raised sharply to compensate for the lowered expectation in that large group of vaccinated people. Otherwise, we will not achieve total mortality. And that's what happened.

But slightly lower among vaccinated people...? Not really. The expected mortality increases considerably in each group! Almost one and a half times as many among those vaccinated aged 76-80! And among the unvaccinated they even expect 178% compared to the national mortality rate. Even the WLZ residents pale in comparison. We are on the eve of a huge wave of deaths! Fortunately, it concerns 2021. That wasn't too bad, also in general practices. Excess mortality has been limited to around 15% since mid-2021.

They estimated the mortality probabilities based on their own patients. Let's hope this doesn't say anything about the effectiveness of GP care. But how does that expectation work out? The shaded bars are the expectation, the solid bars are the mortality measured afterwards.

Bij 76-80 is de sterfte onder ongevaccineerden bijna 3,5x zo hoog als de verwachting. Met droge ogen constateert Nivel hieruit "een oversterfte van 233%". En dat over de periode van een jaar...! Dan zit er echt ergens een schroefje los. Dan heb je geen benul van wat cijfers zijn en wat HVE is.

So there are two more reasons to withdraw the study:

• With a mortality rate of 24.7%, this results in 22,158 deaths per year. Almost all 90,000 unvaccinated people over the age of 81 should have died by now. This is not reflected in the CBS figures.
• The unvaccinated people have therefore either been incorrectly estimated (and then I would immediately re-examine the expectations of the vaccinated people) or the deaths have been measured completely incorrectly. There is no reason (other than the Healthy Vaccinee Effect) why unvaccinated people die 3.5 times more often because others have been vaccinated.

If they had known this in advance, it could have nicely increased the willingness to vaccinate. If that's not a big stick, I don't know what is.

Er gaan dus duizenden extra niet-geprikten dood (3,5x zoveel als verwacht) omdat anderen zich hebben laten vaccineren. Dat "prikken doe je voor een ander" is in die zin toch wat betrekkelijker dan ons aanvankelijk werd voorgehouden. Elk medicijn met zo’n bijwerking op anderen zou onmiddellijk van de markt gehaald moeten worden.

Have your grandparents been a bit stubborn? Have they perhaps not had a shot? No!? Well then go and say goodbye quickly!

Found: old easter egg

I went looking for the difference in deaths in the non-WLZ population between not vaccinated and vaccinated. In week 25 (April) of 2021, mortality varied widely. In the left (or on mobile the top) graph, the ratio is 7.57 per 100K against 24.4 per 100K. That's a factor of 3.2. Unvaccinated people died 3.2 times as often as vaccinated people after two jabs. (Read at Herman Steigstra how that was possible.)

Then my eye fell on the graph that deals with those who had only had one jab. I even found a factor of 4.5 there! If we follow Nivel's reasoning a bit (although we didn't come up with our own expectation), 1 shot worked even better than two! The biggest difference in deaths, and this too shortly after the vaccinations have been administered. It doesn't get any better than that!

I am afraid that all these RIVM and CBS reports suffer from the same data problems and, above all: ignoring the HVE...

Edit: afterwards the term 'HVE' was added to the text in a few places. The phenomenon was described but not explicitly named, which is why this article did not appear on the list search for 'HVE'.