People who work in his lab/hospital? Tbh the participation of a single data faker in a study throws the whole thing into question. Surely not all of these people can be ignorant of his history? So they are ignoring it instead. Why would this be? Perhaps because you are dependent on the faker in some way?

If there is a weak link in the causal chain it throws the whole thing into doubt. An author with history of faked data combined with exceptionally weak and low powered study design yields weak evidence. I barely count this as evidence in either direction about whether this therapy is effective.