John M. Kane, MD: So a lot of interesting points. I think one of the studies that was probably most informative on that was Tyrone Cannon’s study of patients who were converting from the prodrome to actual psychosis, showing that they had brain changes that were unassociated with any antipsychotic drug treatment because they hadn’t been exposed to antipsychotic drug therapy. We’ve also seen some studies recently suggesting that with each relapse the time to response is reduced, and the group at the University of Toronto studied about 130 first-episode schizophrenia patients demonstrating that.

Jeffrey A. Lieberman, MD: You mean time to response is longer.

John M. Kane, MD: Time to response is longer.

T. Scott Stroup, MD, MPH: In the likely….

John M. Kane, MD: In the likelihood of response, right. So is that a pathway to treatment resistance, for example? If you experience multiple episodes you’re more likely to become treatment-resistant? We know that maybe 15% to 20% of patients are treatment-resistant from the get-go, but then there’s another cadre of patients who develop treatment resistance over time. So there are many concerns, obviously, about people experiencing multiple relapses.

T. Scott Stroup, MD, MPH: There is the other side. There is a group of people who have a psychotic episode, or who meet criteria for schizophrenia, who stop taking medications and don’t relapse. And that’s a small percentage.

Jeffrey A. Lieberman, MD: How many of those have you seen?

John M. Kane, MD: We don’t see those people. I think that’s part of the challenge. And there certainly have been recent studies that have raised those questions again—people experiencing dosage reduction and doing OK.

Jeffrey A. Lieberman, MD: John, you’re probably the pioneer of trying to evaluate how you can determine minimum effective dose. You did a number of studies trying to find the optimal dosage with minimal adverse effects while maintaining therapeutic effects. And you found, in the different ways that were experimentally pursued, differential levels of risk in terms of relapse or in terms of adverse effects. At this point, where do you come down, in terms of trying to balance those things?

John M. Kane, MD: I think we’re still left with the question as to whether there’s a small subgroup of people who may do OK without medication. Right now we don’t have any method to identify those patients. So if that represents 5% or 10% of the people that we see, we’re going to treat 100% with the treatments that we have. But I think people are still asking, “Well, what about this subgroup? Is there a way that we can identify them?” And some day, maybe with the right genomics and the right imaging stratification, we will be able to. But right now, I think we all agree that we want to promote the use of medication to prevent relapse.

And that’s a good segue into the next section. We want to talk about patient adherence and the role of adherence in relapse. We commented earlier that there are many things that influence people’s ability or willingness to take medication. Unfortunately, we don’t have good methods to monitor adherence. What we end up doing, most of the time, is asking the patient whether he or she is taking the medication. And patients may not be intentionally trying to mislead us. I think very often that they don’t know themselves how much medication they’ve missed.

So we may do things like pill counts, or use MEMS [medication event monitoring system] caps. But we’re not going to do frequent blood levels, and we have a hard time identifying those patients who were nonadherent. I think monitoring is a big problem. And one of the advantages of the low-acting formulations is that even if we were to say that they may not be superior in some respects, we know that we can identify nonadherence immediately when someone misses an injection. So that is an enormous advantage in and of itself.

T. Scott Stroup, MD, MPH: I want to make sure people know what MEMS caps are.

John M. Kane, MD: Oh sure.

T. Scott Stroup, MD, MPH: It’s a brand name of an electronic compliance monitoring system, and it’s the thing that lets you know when people open the pill bottles.

John M. Kane, MD: Right.

T. Scott Stroup, MD, MPH: If they’re not opening the pill bottles, you know that they’re not using them. If they are opening the pill bottles, you don’t know if they’re ingesting them.

John M. Kane, MD: Absolutely. And people use prescription refill rates. You can get data from large organizations telling you whether the prescription was filled. So, again, we have limited ways to monitor adherence.

What else could we be doing to improve adherence? Obviously there have been behavioral strategies. We’ve seen people doing group therapy to try to increase the motivation to take medication. That’s part of a therapeutic alliance that is trying to educate patients about medication. But at the end of the day, certainly in my hospital, it seems like half of the readmissions of people with psychosis are due to nonadherence. And so we still have a long way to go.

T. Scott Stroup, MD, MPH: I think that’s right. I think inpatient units are filled with people who have stopped their medications.

John M. Kane, MD: What about the introduction of the so-called atypical oral medications, Scott? Do you think that that helped, in terms of improving adherence?

T. Scott Stroup, MD, MPH: I think we hoped that it would, because the newer medications had different adverse effects, or had fewer of the older adverse effects. They didn’t have so many neurological adverse effects that seemed to be such a problem. And so, the hope was that they would work really well and not have adverse effects, and therefore, people would take them.

Unfortunately, it didn’t make much of a difference—the oral atypicals. They’re good medications. They’re efficacious. But as far as I know, they didn’t have any big advantage on adherence.

John M. Kane, MD: On adherence, yes. There were a few studies that suggested maybe some small advantages. And they certainly were an advance in terms of reducing some of the neurologic adverse effects, but then it turned out that there were metabolic adverse effects that were somewhat more common.