This is also my take on this paper. It records an increase in reported side effects, which is not the same thing as an actual increase in side effects. Patients

This is also my take on this paper. It records an increase in reported side effects, which is not the same thing as an actual increase in side effects. Patients may have been experiencing these side effects anyway, but either didn’t report them, or didn’t associate them with the intervention. Knowing these side effects are associated with the intervention would likely be enough for Confirmation Bias to increase the reported instance.