The parallel world of numbers

Posted On 10/04/2022

In science, the measurement of properties of observed objects plays a central role. In this process, we expect reliable results that are as precise as possible, the subsequent analysis of which should enable us, among other things, to explain phenomena, identify correlations or even make forecasts for future developments. One of the standard quality criteria required in this context, besides validity and reliability, is objectivity[1]: a measurement should be independent of the person who performs it, or who determines the conditions under which the measurement is performed (no, we are not talking about quantum physics here).

The detection of COVID-19 cases through PCR tests is such a measurement, so the criteria mentioned here should also be observed. The practical implementation of contact tracing over the last few months has been changed several times, among other things due to the emergence of the Omicron variant. This circumstance allows us to draw informative conclusions about objectivity in this context, as we will see in the following.

Since 6 December 2021, the Ministry of Health has published the incidence rates (per 100,000 persons) of vaccinated and unvaccinated persons in its daily COVID-19 reports[2]. We have visualized this data for the period 06/12/2021 to 07/04/2022 in the following graph.

As no reports are produced over the weekend, no data is available here for the weekdays Friday and Saturday. On the COVID-19 website, these data are visualized continuously for all days, but they have already been provided with a moving average, which is a disadvantage for our purposes.

It is noticeable that the ratio of the two incidence rates changes continuously. In January, for example, the rate of the unvaccinated (blue) is much higher than that of the vaccinated, while this behaviour reverses in the 2nd half of March.

We would like to find an explanation for this and therefore look at the vaccine effectiveness resulting from these rates[3]. This can be estimated with the formula:

$E = 1 - \frac{I_V}{I_U}$

where $I_V$ and $I_U$ are the incidence rates (tested positive per 100,000) of the vaccinated and unvaccinated, respectively.

We are aware that the figure calculated in this way can only provide a very rough estimate of this size. However, this is not relevant for the further explanations, as the main point is to put the incidence rates into a relationship. If one only looks at the quotient of the two rates (as it is also given in the daily reports), one comes to the same conclusion in this context. However, the vaccine effectiveness also allows an interpretation, which is shown in the following table.

Vaccine effectiveness	Meaning
100%	No one who has been vaccinated falls ill.
above 0 %	The incidence rate of the vaccinated is lower than that of the unvaccinated: vaccination has a positive effect
0%	no effect
below 0%	The incidence rate of the vaccinated is higher than that of the unvaccinated: vaccination has a negative effect

The vaccine effectiveness calculated from the incidence rates is shown in the following graph.

Here, the fluctuations mentioned at the beginning become even more apparent. Since vaccine effectiveness cannot change for such a short time, this phenomenon must have another reason.

With the advent of the Omicron variant, contact tracing reached its performance limits. In the period we looked at, its modus operandi changed twice:

From 31 December, only the unvaccinated were given a quarantine order in case of a “high-risk” contact, which had to be terminated after 7 days by a negative PCR test[4]. Up to this point, such orders were still issued to everyone, and it can be assumed that this change only took effect with some delay. From that moment, vaccinated (and recovered) people no longer contributed to the statistics in the “contact tracing” category with positive PCR tests.
Finally, on 11 February, quarantine was abolished across the board, regardless of vaccination status[5].

(A new version of the Covid-19 law[6] also fell into this period (11 January). However, this had no influence on contact tracing).

In the following chart, the dates of these changes are now plotted, and we have also grouped the points according to their values for a better overview.

A first discontinuity around Christmas 2021 is probably due to the outbreak of the Omicron wave: here, the vaccine protection most likely proved less effective compared to the new variant, resulting in reduced vaccine effectiveness.

Obviously, the further erratic changes are largely due to the way contact tracing is carried out. Between early January and mid-February, only the unvaccinated continue to be tested with a PCR test at the end of quarantine. This leads to a higher number of cases and incidence rate in this group, and correspondingly then to better vaccine effectiveness.

However, this has far-reaching consequences for the validity of the data published by the Ministry of Health.

Let’s ask ourselves the basic question: what should be measured at all? The answer in this context is of course the total number of Covid-19 cases according to the WHO definition, i.e., all persons with a positive PCR test[7]. However, a part is deliberately ignored here, namely all vaccinated persons who would potentially have tested positive after quarantine. We thus not only receive incorrect values for the number of cases of vaccinated persons in this period, but also for all other variables which are derived from this:

Incidence rate of the vaccinated
Total number of positive tests
Positivity rate of the tests
Effective reproduction number
etc.

Despite this far-reaching intervention in the tracking of positive cases, and the resulting falsifying influence on almost all statistical variables, all figures continued to be published without any correction or comment. This is a practice that can only be described as highly unserious and unscientific.

The fog is lifting

With the amendment on 11 February, at least all persons are now treated equally regarding quarantine regulations, no matter their vaccination status, which for the first time allows a more or less real comparison.

Moreover, the now very low number of vaccinations eliminates another “confounding factor”: the positive cases due to vaccination reactions after the first dose, which are attributed to the unvaccinated[8].

Since the end of February, the incidence rate of the vaccinated has repeatedly been higher than that of the unvaccinated, so the vaccine effectiveness has been correspondingly negative. If a linear regression is carried out, a clear downward trend (approx. -1% per day) emerges with a useful correlation (R2 = 0.57).

We would like to emphasize once again that this alone is only of limited significance for the actual epidemiological situation. However, a corresponding development increasingly appears in the statistics of other countries, and in some cases has been doing so for some time.

For example, the RKI recently stated in its weekly report of 31/03/2021[9] that for the age group 18 to 59 years, vaccination hardly brings any advantages for symptomatic cases, and rather an opposite tendency can be observed.

In the UK, higher incidence rates among the vaccinated than the non-vaccinated have been measured for weeks[10]. However, this information has now not been reported since the 14th week:

From 1 April 2022, the UK Government ended provision of free universal COVID-19 testing for the general public in England, as set out in the plan for living with COVID-19. Such changes in testing policies affect the ability to robustly monitor COVID-19 cases by vaccination status, therefore, from the week 14 report onwards this section of the report will no longer be published.

It remains to be seen until when our Ministry of Health will publish data on the incidence rates of vaccinated and unvaccinated people.

UPDATE 11.04.2022 20h00:

It already happened, this data will no longer be published from today on (daily report of the Ministry of Health from 11.04.2022): “Given the current state of the pandemic and the stable situation in hospitals with low numbers of COVID-19 patients in intensive care and normal care, the daily incidence rates of vaccinated and unvaccinated persons will no longer be published from now on.”

For information: the incidence rates are on average even higher at the moment than during the entire year 2021 (Screenshot: https://covid19.public.lu/fr/graph.html ):

The parallel world of politics

A higher incidence rate among the vaccinated compared to the unvaccinated cannot be explained solely by a decline in the effect of the vaccination. There is therefore a need for clarification here, especially because an increased number of clinical pictures are observed in vaccinated persons, which suggest a general weakening of the immune system, such as shingles.

The first scientific publications have already appeared on this subject. In our latitudes, it is above all the alternative practitioner Florian Schilling who is doing pioneering work in this field. We report on this topic in a recently published article[11].

However, our political leaders seem to be unaware of all these developments, or they are deliberately ignoring them. Either way, they continue on their chosen course (compulsory vaccination).

The legitimate question, however, is how long we can continue to look the other way.