This study [0] has a very small sample size and they found a significant difference only in a few of the variables they measured. 11 fewer individuals who got t

This study [0] has a very small sample size and they found a significant difference only in a few of the variables they measured. 11 fewer individuals who got the booster in the opposite arm had CD8 T-cells 2 weeks afterwards than those who got it in the same arm. They say that whether participants got the booster in the same or opposite arm was randomly assigned, but the odds of of the final ratio of same-arm to opposite arm happening by chance are just 4%, and the further odds of getting the ratio for the subgroup that had their CD8 T-cells measured given the full study's participants is 4.4%, for a combined probability of just 0.1%. This would seem to suggest there was some confounding variable that caused participants who got the booster in the same arm to preferentially drop out of the study or otherwise be removed from the dataset. For the full study group, the same-arm participants were on average younger and more likely to be female, the breakdown for the sub group of interest is not given. Again, this suggests some confounding variable. While this warrants further study, I am unconvinced. [0] https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(23)00308-0/fulltext#%20 https://www.thelancet.com/journals/ebiom/article/PIIS2352-39...