There are two issues there. One is what you ideally want if you are going to pay for a drug. If you're a drug benefits manager in a province, you want to know whether it's worth it for you to pay for the drug. You essentially want to know whether that drug is going to reduce your costs--your hospitalization costs, your emergency-visit costs, your treatment costs.
Unfortunately, many of the phase three trials have end points that don't allow that to be determined. For example, this is the reason British Columbia is running its own trial for the Alzheimer's drugs: to determine whether or not they delay nursing home admission, because that's their big cost, right? Whether the person does one point better on Mini-Mental status isn't really that helpful, because that's not really the end point that you care about.
That's one issue, and Canada can't really take that step alone. We're 2% of the drug market, so do we want to take that step alone? I think we should take it internationally and say that in phase three pre-market standards, we want to have disease end points and not intermediate proxy end points. That means the trials are going to be longer. They're going to be more expensive, and--let's not kid ourselves--the cost of those trials will be downloaded onto the price of the drug, so I think there's not just a simple answer to your question.
The second issue is that at the moment you cannot control off-label prescribing. Once a drug is out in the market, you can't say, “Don't prescribe it for kids and don't prescribe it for people who are taking more than three other drugs”, which is the group on which it was tested. The group on which it was tested is not necessarily the group in which the drug is going to be used.
That can again be a pre-market requirement. You can say you want the target population on which you test that drug to be just like the target population in which you're going to use that drug. A granny who is on 16 drugs and who has three other health problems is going to be pretty unpredictable, as you can imagine, and you're going to need a very large sample of people on which to test that new drug.
The third thing you really care about is whether this drug is any better than the other drugs on the market. That's what Mary was talking about. If you want to know that, you make it a pre-regulatory requirement that the trials get compared to current comparators on the market for treating the same treatment indication. Again, can Canada go it alone? I doubt it. I think we are a small portion of the drug market and I think that there would have to be an international agreement to do so.
In the end, you're still going to have some unknowns--the rare events that are going to happen, the off-label prescribing that's going to happen--and there you're going to get some unexpected outcomes for which you have to have a post-market surveillance. There's no other way of doing it.
So either you get international collaboration on getting the right end points before notice of compliance or you actually put in a robust system after the fact to say that it's conditional upon your meeting the following requirements in the first two years the drug is on the market, and you've got that information system behind you to say you can actually measure it--not on the people that you select to be in your study, but on each and every person who's on that drug.