900x Reporting Rate for "blood clots"

Post #1193

Mar 06, 2025

NOTE: An error was found in this prior post on this topic, and this new post is the updated (corrected) version.

When comparing 10 Moderna lots to 8 years of flu shots in terms of the reporting rate for MedDRA code 10014523 (embolism and thrombosis) — aka “blood clots” — then you find a 900-fold difference. Here is the search page for Moderna, showing the 10 lots which got averaged out in terms of the mean reporting rate per million doses:

While 383 reports show up in the bottom 10 rows above, the actual sum of embolism/thrombosis reports for this group was 411, due to typographical variations covered here. The total dose count for these 10 lots was 12.5726 million doses. The average reporting rate for embolism/thrombosis was 33.1 reports per million doses.

Here is the search page for the 8 seasons of flu:

Counting started with the 8th month (August) of 2011, or 3rd row, and all flu seasons had clear demarcation (no reports came in for March-July, which covers the “inter-season” months between successive years of flu shot dosing). Counting ended with March 2019, so that all flu shots given in the last flu season (2018/19) were covered.

The dose count for all 8 seasons was just over a billion doses of flu shots, and the grand total of reports of embolism/thrombosis was 42 — which makes for a reporting rate of 0.04 reports per million doses, a value more than 900 times lower than the mean value for Moderna shots.

Sensitivity Analysis

Because of the convenience sample of Moderna lots, a sensitivity analysis would help to improve validity. Common methods of sensitivity analysis — such as “leave one out”-analyses (one selection is left out and new statistics are formed, repeating until all have been “left out”) — are not strong enough, so stronger methods were created.

One of the strongest methods is to compare an aggregate value to every single individual value in the other data set. Not only is the mean of one group shown to be higher than the mean of the other, but it is shown to be higher than every single individual value. This newly-created method is called Every Single Individual method.

Regarding the mean of flu seasons versus Every Single Individual Moderna Lot, here is a graph of individual lot ratios compared to flu:

Moderna lot 022 M 20 A (far left) had over 1100x the reporting rate for clots versus the average of 8 seasons of flu shots. Each individual lot was at least 800 times worse than flu shots. But to round out the sensitivity analysis, let’s reverse the analysis to make a contrast between the average of Moderna lots and “every single individual” flu season.

Regarding the mean of Moderna lots versus Every Single Individual flu season, here is a graph of individual lot ratios compared to Moderna lots:

Notice how, when reporting for flu shots was low (2014/15 flu season), the Moderna/flu ratio jumped. But most importantly — when flu shots were “at their worst” regarding leading to adverse event reports of clotting (2017/18 flu season) — then it was still true that the average Moderna lot was almost 500x worse than flu shots were, at their worst.

The evidence suggests that Moderna shots were several hundreds of times more likely to result in reports of clotting than flu shots. This finding was robust to one of the strongest methods of Sensitivity Analysis that has ever been created:

the Every Single Individual method.