Yes, this stood out. The same judgement applied to a selection of cases that is not hand-picked (first N cases, except for having to choose a value of N) sounds like a good basis.

Maybe something good comes out of this, e.g. saving patients by treating them with anticoagulants preventively or something like that.