What needs to happen is that every patient's record in a clinical trial/study, whether they formally consented or not, be hashed. And that hash identifier provided to the patient so they can assess their own entry.
Only in that way will data fraud in large trials be inhibited.