I think this is a very interesting question and again one that has immediate relevance. The interesting question is whether it falls under our section 8 analysis. Having said that, I'd be happy to offer an opinion.
I actually think the clinician who is uninvolved with the research may be an appropriate party. Certainly, legally, I think they're an appropriate party, because the consent will be obtained in a clinical setting over which the clinician has legal responsibility, whether that is for research or clinical activities. It's important to note that the clinician should not be the researcher in that sense.
The other reason a clinician might be a good person to play this role is that they have other obligations toward the patient. They have fiduciary obligations that exist for the exact purpose of dealing with potential conflicts of interest.
One could argue that in other research settings, be it cancer research or other clinical research settings, the potential conflicts may be even more profound. I'm not saying there aren't issues in those contexts, but in those situations, we let the clinician get consent. Some would argue that they're the appropriate person to get consent because they know the patient well and they have these other legal obligations.