How are the incidence values determined according to vaccination status?
To calculate the unvaccinated incidence, the corona case reports are broken down by vaccination status and compared with the population figures from the Digital Vaccination Rate Monitoring (DIM). The formula is similar to the “normal” 7-day occurrence:
Covid-19 infections without vaccination within 7 days / number of people without vaccination * 100,000
The calculation of the incidence in vaccinated persons is done accordingly. Only persons who were fully vaccinated at the time of infection count as “vaccinated”. This means that at least 14 days have passed since the end of her vaccination course (two doses of Biontech, Moderna or Astrazeneca or one dose of Johnson).
Infected persons for whom no information on vaccination status is available are counted as “unvaccinated” when calculating LGL. Cases where a vaccination series has started but not yet ended, or where the last vaccination was less than 14 days ago, are not attributed to any of the groups – they are excluded from the calculation.
LGL indicates the fifth calendar week as the starting point for the evaluations – this is the earliest time for examination of “fully vaccinated” according to the mentioned definition. As with all other incidence values, the reporting date is crucial for the timely allocation. This is the day when the local health department first becomes aware of a corona case – that is, when confirmation of the infection by a positive PCR test is received there
What are the weaknesses of the data?
LGL chose an easy-to-understand form for its evaluation: the incidence is only divided into “fully vaccinated” and “unvaccinated”. Josef Eberle, Professor of Virology at the Ludwig-Maximilians University in Munich, is in favor of a differentiated analysis by age, as vaccination rates in particular, but also the risk of a serious course and the risk of exposure differ markedly in different age groups.
Another point of criticism: Different test behaviors. In emails and comments, many BR24 users asked if the incidence of vaccinated was not only lower because they were tested less due to the applicable 3G rules. Asymptomatic infections, which should occur more frequently in vaccinated people, would remain undetected.
Different test behaviors do not necessarily lead to bias
LGL states that due to the current 3G rule, it can be assumed that unvaccinated people can be tested more frequently than vaccinated people. However, the increase in asymptomatic new infections, which may remain undetected as a result, is not as large as is often assumed. Analyzing the number of Covid-19 cases with information on vaccination status AND symptoms, it shows that 19 percent of the unvaccinated infected are asymptomatic – among the fully vaccinated infected 30 percent. According to these data, it can not be stated that there will generally be more asymptomatic infections in unvaccinated persons.
In addition, data from controlled trials according to LGL also indicate that vaccinated people are actually simply less likely to become infected than unvaccinated people. The clear difference between the occurrences should therefore not only be rejected as a result of different test behaviors.
Virologist Josef Eberle also comes to this conclusion: Due to the 3G rules, the 7-day incidence of the vaccinated may be slightly underestimated. Nevertheless, the difference in the incidence of unvaccinated is so clear that the figures make it possible to draw conclusions about the effectiveness of the vaccinations – not only in terms of the possible course of the disease, but also in terms of the infection itself.
Vaccination status information is not up to date
According to a spokeswoman for RKI, it is possible that vaccinated people will be tested less than unvaccinated people. As part of regular screenings, for example in clinics or nursing homes, tests should also be performed regardless of vaccination status.
Expert Josef Eberle also points out that these key figures are affected by the usual delay in reporting coronadata. Although they are only reported weekly, at the time of publication they are by no means really updated data.
According to LGL, the information on vaccination status is even particularly hard hit. For example, vaccination information may not be available once the first case is reported, but will only be collected as part of further case studies and then added later.
The chosen calculation method can also limit the meaningfulness of the values: Failure to take into account the started vaccination series could, according to RKI, lead to an underestimation of the vaccination frequency. On the other hand, the fact that LGL allocates all cases where vaccination status information is missing to the group of unvaccinated whose incidence may be significantly too high. According to LGL, a total of 67 percent of Bavarian case reports in the period from calendar week five to September 7 contained information on vaccination status.
Conclusion: Data show a general trend – significantly fewer vaccinated among those infected
The incidence values published by LGL according to vaccination status are therefore subject to some restrictions. The values must therefore always be seen in connection with the case definition, data collection, reporting delay and also different age groups.
However, experts agree that conclusions can still be drawn. “Despite the known limitations, the evaluation provides an opportunity to analyze general trends in the relationship between the fully vaccinated and the unvaccinated population,” says LGL.
In fact, fewer vaccinated people tend to be infected with coronavirus than unvaccinated people. However, the information is not “live data” that can describe the current infection process.