Overview:
In part 3 of our US disabilities analysis we observed that the rise in disability rates post 2/2021 correlates closely with the rollout of the vaccination schedule.
When looking at changes in disabilities on a wider time frame (since 2008) we observe that the disability rates rose or fell from month to month but tended to be relatively stable over time. However, as shown in part 1, the change in behaviour since early 2021 is clearly an abnormal occurrence with high level of statistical significance. It happens to be highly correlated to the cumulative Covid-19 vaccine rollout, but we cannot state that the correlation is statistically meaningful as it is based on a cumulative plot with obvious autocorrelation.
In this section we provide further evidence that the most likely cause of the rise in disabilities is the Covid-19 vaccines. For that purpose, we model the expected rise in disabilities due to the vaccination rollout in the general population. We do so by using the rates of Serious Adverse Events (SAEs) obtained by the safety analysis of the mRNA vaccine (Pfizer and Moderna) clinical trials, performed in the Vaccine journal paper we reviewed here.
In summary, the paper found a rate of 13.2 SAEs per 10,000 participants in the combined (Pfizer + Moderna) vaccine groups in excess of the placebo groups. They also found an increased rate of 12.5 per 10,000 SAEs of special interest in the vaccine groups versus the placebo groups. (SAEs are considered ‘of special interest’ if they are events known to be associated with vaccines in general, have been previously observed with the vaccine being studied, or could be associated with the vaccine based on animal models or theoretical models relating to Covid-19 pathology. The definition of an SAE of special interest is therefore narrower, but can be considered more likely to be related to the vaccine than when all SAEs are counted.)
The SAEs were measured from August 2020 to December 2020, before the unblinding of the clinical trials. The SAEs were recorded following dose 1 with a median follow up period of at least 2 months. For the Pfizer trial, SAEs stopped being recorded after 1 month after dose 2.
Estimating a time series of SAEs from the mRNA trials
We want to investigate the relationship between the rise in disabilities observed in the BLS data with an estimate of SAEs that would be observed over time, taking as a baseline the rate of excess SAEs in vaccinated individuals when compared to the placebo groups, in the mRNA vaccine clinical trials. We use the rate of 13.2 per 10,000 for SAEs observed during an estimated 2-month period of observation.
An event that leads to a permanent disability would meet the criteria for an SAE, and consequently it is not unreasonable to use the rate of SAEs as a basis for projecting the rise in disabilities from 2021.
Furthermore, the clinical trials were performed mostly in individuals aged from 16 to 69 (relatively young and healthy individuals). Consequently, we believe that the best subset of the population for which we should investigate the relationship between disabilities and SAEs, is either the Civilian Labor Force or Employed population aged 16-64.
In order to compute a time series of estimated SAE during the vaccine rollout, we use the rate of excess SAEs computed in paper and extrapolate them using the assumptions listed below.
We multiply the monthly rate of SAEs by the percentage of the population, aged 19-64, that received a Covid-19 vaccine (any dose or brand), based on CDC vaccination numbers. We assume that the rate of SAEs declines monotonically over time following vaccination, according to a certain unknown function. As an approximation, we assume that the SAE rate declines linearly over a given period of time (6 months) and thereafter stabilises at a residual value.
Assumptions for computing the SAE time series:
-> The initial monthly rate of excess SAEs of special interest corresponds to 12.5 per 10,000 (the rate computed in the paper) divided by the median observation period of 2 months. The standard error (SE) was 5.2.
-> The initial monthly rate of excess SAEs corresponds to 13.2 per 10,000 (the rate computed in the paper) divided by the median observation period of 2 months. The standard error (SE) was 8.2.
-> After 2 months the rate of SAEs falls to 2/3 of the original value and after 4 months it falls to 1/3 of the original rate. After 6 months we assume that the rate of SAEs falls to 1/10 of its original value.
-> The assumptions are based on the thought that different types of severe or serious events could occur over an extended period, but also that most events will be front-loaded. Additionally, we speculate that the milder the event the more it is likely to be due to a reaction at the injection site and serious events could manifest over a extended time. The terminal rate of 1/10th of the original value attempts to capture the possibility of longer-term effects. These assumptions constitute our best guesses, and both the starting rate of AEs or SAEs and the assumption of how they decay (drop off) over time, could be adjusted as more information becomes available.
-> Furthermore, it should be noted that the limitations of the paper in estimating the SAEs rates must also be considered, namely, that the SAEs may be under-reported, as explained by the authors. Additionally, we must consider that the computed rates have a large standard error, partly due to the authors' not having access to participant data. Finally, the rate of SAEs differed between the Pfizer and Moderna trials, and the combined rate computed in the paper might not reflect the number of each type of vaccine administered in the population groups of interest during 2021.
Increase in disabilities vs estimated rate of SAEs from mRNA vaccine trials
The following charts illustrate the relationship between the increase in the disability rate in the Civilian Labor Force (16-64) and the projected rate of SAEs, based on the Covid-19 vaccination administration rate for the 19-64 age group, while using the assumptions listed above. We compare both the SAEs of special interest, and the broader category of all SAEs, in separate analyses.
Serious adverse events of special interest.
The chart on the left shows the time series of the change in disability rate from 2/2021 to 12/2021 for the Civilian Labor Force (left scale), and also the time series of the projected rate of SAEs of special interest (right scale). The chart also shows the time series of SAEs of special interest considering the mean rate of SAEs minus 2 SE to the mean rate plus 2 SE.
The chart on the right shows the correlation between the rise in the disability rate since 2/2021 with the projected cumulative rate of SAEs of special interest. The regression R2 is 87.7% which is evidence for a strong relationship. We should also note that performing the correlation of cumulative time series is misleading, and the R2 should not be taken as an indication of establishing a statistically significant relationship as both time series have autocorrelation.
Nevertheless, the charts below strengthen the case for a causal relationship between the Covid-19 vaccines and disabilities , as the projected SAEs of special interest are based on the rates estimated from analysis of the mRNA vaccine clinical trials. Additionally, under the reasonable assumptions stated above, the time series are shown to be of the same order of magnitude with each excess SAE of special interest (of vaccinated individuals versus placebo) translating into 2.71=1/0.369 disabilities (as measured using BLS data). In other words, the rate of projected SAEs of special interest appears to under-estimate recorded disabilities by about 2.7.
We can observe that the rate of rise in disabilities is higher than the computed rate of rise in SAEs of special interest, which could be explained in several different ways, or by a combination of factors.
By the definition of an SAEs of special interest being more extreme than the rate of disabilities reported by the individuals surveyed by the BLS i.e. some disabilities reported by the individuals surveyed by the BLS may be caused by adverse events that are not deemed SAEs of special interest, due to the criteria used to define an SAE being overly restrictive.
The population of the Civilian Labor Force as a whole may be less healthy (and somehow more vulnerable to vaccine-related disabilities) than the vaccine trial populations, either due to the selection criteria for participation in the trial, or ‘self-selection’ bias.
Under-reporting of SAEs of special interest in the trial populations.
Other factors causing excess disabilities in the Civilian Labor Force in a concurrent timeframe to the vaccine rollout.
Serious adverse events (SAEs).
The chart on the left shows the time series of the change in disability rate from 2/2021 to 12/2021 for the Civilian Labor Force (left scale), and also the time series of projected rate of SAEs. The chart also shows the time series of SAEs considering the mean rate of SAEs minus 2 SE to the mean rate plus 2 SE.
The chart on the right shows the correlation between the rise in the disability rate since 2/2021 with the projected cumulative rate of SAEs. We should also note that performing the correlation of cumulative time series is misleading and the R2 should not be taken as an indication of establishing a statistically significant relationship as both time series have autocorrelation.
Under the assumptions stated above, the time series are shown to be of the same order of magnitude, with each excess SAEs (of vaccinated individuals versus placebo) translating into 2.56=1/0.39 disabilities (as measured using BLS data). In other words, the rate of projected SAEs appears to under-estimate recorded disabilities by about 2.6.
We can observe that the rate of rise in disabilities is higher than the computed rate of rise in SAEs, which could be explained by in several different ways, or by a combination of factors.
Some disabilities reported by the individuals surveyed by the BLS may be caused by adverse events that are not deemed SAEs, due to the criteria used to define an SAE being overly restrictive.
The population of the Civilian Labor Force as a whole may be less healthy (and somehow more vulnerable to vaccine-related disabilities) than the vaccine trial populations, either due to the selection criteria for participation in the trial, or ‘self-selection’ bias.
Under-reporting of SAEs in the trial populations.
Other factors causing excess disabilities in the Civilian Labor Force in a concurrent timeframe to the vaccine rollout.
We also can observe that the rate of disabilities lies close to the computed SAEs using the mean rate of SAEs plus 2 SE. Therefore, the actual measured disabilities fall close to the upper 95% confidence interval of SAEs from the clinical trials.
PHINANCE TECHNOLOGIES