So in a big collaboration like the MTA with six now seven sites, a principal investigator at each site, a co-investigator at each site who had to be in a different discipline, so if the investigator were an M.D., the co-I had to be a Ph.D. or vice-versa to ensure multi-disciplinarity, project coordinators during the trial. Head therapists at each site, both psychological and pharmacotherapists.

During the active treatment phases of this study in the 1990s there were over 15 hours of conference calls a week across the sites. Three hours for the principal investigators, a couple hours for the coordinators, a couple hours for the pharmacotherapists, a couple hours for the psychological therapists. So that as the clinical cases emerged, and not everybody followed protocol exactly, there were the times that we had to deal with emergent situations and fidelity issues regularly.

Now that we're in follow-up phases the conference calls have backed off to a more reasonable number. Yet, administratively there needs to be, there needed to be and needs to be a chair of the steering committee. And that's usually rotated every year, so there's no czar, but so that each site gets the chance of leading and there are bylaws that say many decisions are made by majority rule, the hope is for consensus and very importantly an interesting rule that emerged in the development of the MTA was let's say a decision had been made about a certain treatment or assessment or procedure. What if six months later a couple of the investigators say you know what that wasn't such a great decision.

A very important rule suggested to us by Helena Kraemer who is our statistician, a wonderful consultant, was to overturn a preexisting rule, no simple majority pertains, you need of the seven sites a five-to-two rule, a 70 percent majority, so that you don't keep undoing yourselves every other month.

And things like that have made a big difference in order to maintain consistency in protocol. Yet, enough flexibility that if things needed to be changed they could be changed.