Editor's note:
In 2009, new expert consensus guidelines outlined the major factors involved in medication nonadherence for patients with schizophrenia and bipolar disorder, as well as the recommended strategies for clinicians to assess and enhance adherence. Medscape spoke with the guidelines' lead author, Dawn Velligan, PhD, a researcher at the University of Texas Health Science Center, San Antonio, about data specifically related to adherence with antipsychotics and their implications for practice.

Medscape: In 2009, you and your colleagues published expert consensus guidelines^[1] related to adherence issues in serious mental illness -- bipolar disorder and schizophrenia. In the consensus survey for those contributors, persistent side effects or the fear of the side effects was identified as the top potential reason for nonadherence in patients with those illnesses. How does the expert consensus concept compare with what the literature says about risk factors for nonadherence?

Dr. Velligan: The literature is quite mixed^[2,3] regarding the extent to which side effects affect whether an individual will discontinue medication or take less medication. First of all, adherence is multidetermined, and side effects are just 1 potential influence on adherence behavior. In addition, the different studies in the literature might measure side effects differently. It might not be a specific side effect, but instead the cumulative impact of several different side effects, that is important for an individual in determining whether they want to take a particular medication or not.

Also, the subjective impact of the side effects might be very important to one patient, whereas another may be more focused on the relative risk-benefit ratio. People who are taking medication are going to tolerate more side effects if the medication works better, so it is a very complex issue. And the literature is fairly divided: Some studies show that, indeed, side effects do predict whether people take medication; others do not. But providers are very sure that side effects matter because they hear it from their patients. Patients tend to complain about things like movement side effects or dysphoria, sexual dysfunction, weight gain, and sedation.

Providers and patients really need to get together in a model of shared decision-making to choose the best treatment option for the specific individual. Some people may be much more bothered by sedation than other people, for example. Other variables that influence adherence include:

• Insight,^[2,4] if you think you do not need medication;
• Cognitive impairments^[5] in memory and distractibility -- people forget to take medication even though they are on board with the idea that it might be a good thing to do;
• Lack of a daily routine;^[6] and
• Problems with the treatment and reimbursement infrastructure^[6,7]

Medscape: So, in a conversation between a patient and a clinician, the patient may be complaining about side effects and he or she may also be nonadherent, but the 2 are not necessarily related.

Dr. Velligan: It depends. For the individual patient, it may be a part of the adherence picture, but it is definitely not the whole story.

Medscape: There is huge variability in side effects among antipsychotics, both the older and the newer. What data are available relating adherence specifically to such factors as weight gain, sedation, or extrapyramidal effects?

Dr. Velligan: There are a lot of studies^[8,9] looking at these issues in the literature, and some studies comparing older-generation and newer-generation drugs suggest that patients may be more bothered by movement side effects: extrapyramidal-type side effects. However, a lot depends on the context. Individuals who work may complain most about cognitive problems and sedation, and certainly, the health system cares a lot about the long-term effect of such things as glucose dysregulation and weight gain. Although the side effects are different for the different medications, whether or not they affect adherence is a very individual issue. If you are underweight and you could stand to gain a few pounds, weight gain is not a side effect that is going to bother you; so, if you have a study that has a lot of individuals of different weights in it, you might not find that it correlates with adherence. Different studies have many different variables that go into the equation; sometimes these things predict adherence and sometimes they do not. It is not as clear a picture as we would like.

Medscape: What do we know in general about adherence patterns between older and newer antipsychotics? You mentioned that there is a perception that adherence might be better with newer agents because of the lack of extrapyramidal effects. Is that actually the case?

Dr. Velligan: Overall, the studies suggest that there is very little difference between first- and second-generation antipsychotics in terms of how well people adhere. We had a lot of hope that the lower liability for movement side effects would improve adherence, but the data suggest otherwise. A few studies have shown slight differences that favor the second-generation antipsychotics, but there are also some methodologic problems with them. It does not look like there is much of a difference in adherence, which kind of makes sense to me. I have always dealt with other reasons for lack of adherence, such as poor ability to adhere to a medication regimen, a chaotic lifestyle, and disorganization.

Medscape: Adherence is not a mental health-specific issue.

Dr. Velligan: No, absolutely not.

Medscape: Moving away from side effects and talking about fundamental symptom control as another factor that you mentioned earlier, are there data on adherence with antipsychotics relative to persistent symptoms?

Dr. Velligan: Absolutely. One of the reasons that people discontinue antipsychotics is because they do not work well. If a medicine does not work, people are more likely to discontinue it; that's why you get studies, such as Clinical Antipsychotic Trials in Intervention Effectiveness (CATIE),^[10] in which 74% of the people eventually wanted to try something else. Sometimes it was the clinician who said this is not working well enough, and sometimes it was the patient who said it is not working well or it has side effects. We also know that even the most adherent individuals will experience persistent symptoms. [Because of safety and other concerns], physicians are often slow to move patients on to such medications as clozapine, which is the only medication that has repeatedly demonstrated that it improves treatment-resistant symptoms. Very few patients are on that drug, but it can be a good choice for those who have persistent symptoms and are not getting the relief that they want.

Some studies in the literature show that if you adhere better to medication, you are going to have lower levels of symptoms, but there are also studies that do not confirm that. The issue, to me, seems to be that prescribers recommend a dose of antipsychotic in the context of a person's life and in the context of the fact that they are partially adherent to medication at best. And so, over time, the doses of medication tend to increase, and the individual does not really need all of the medication that they are prescribed. If you have that situation, you can see that symptoms are not going to get better just because you make someone take 100% of their medication when they don't need that much. It's a very complex relationship.

Medscape: All of this points to a high degree of individualization in the patient's therapeutic plan and in the conversations that have to happen regularly to assess the patient's adherence, what has changed, and what is influencing it. What kind of questions do you recommend going through with the patient or with their family or caregivers to make sure that you are collecting all of the relevant information?

Dr. Velligan: A lot of the consumer work that is done by such people as Patricia Deegan^[11] suggests that one of the most important questions to ask is what the patient or the family wants out of the medication. The goals of the treating physician may not be at all the same as those of the patient or family. Somebody might want to take medication that is going to help them be more active in taking care of their child, and they might not mind so much if the voices are not completely gone, or if some of the other symptoms are still there. But if you sedate them to get rid of the other symptoms such that they cannot play with their child, they are not going to like the medication.

It is really important that the physicians ask what the person wants out of the medication and that there is some kind of shared decision-making that goes on. In order to do that, the client needs a lot of information. They need to know what the choices are and what the potential side effects are, and they need to be willing to try things for a while and see how they go. All of that takes time, and unfortunately, in underfunded public systems, there is not a lot of time to do that. You have to use nurse practitioners, physician assistants, and peer counselors to help get all the information across that is needed, so that the 15 minutes or 10 minutes with the prescriber can be used the best way possible to get the best outcome.

Medscape: Do you recommend bringing the family into the equation routinely?

Dr. Velligan: I do recommend that the family come in whenever possible, and I tell family members to go to visits with their relative. It is really the only way that you can know what is going on. A lot of times, especially if people are unassertive or if they have a lot of negative symptoms, they do not demand good treatment, and when you do not demand good treatment, you do not get it. So the family can be there to help. If I am a busy physician and somebody isn't on fire, I might say, "How have you been doing over the past 3 months?" They know I'm busy, so they're going to say, "Fine"; then I'm going to say, "Okay, take the same stuff for the next 3 months." Those kinds of visits happen all the time in community mental health, and having family present can stop that.

Medscape: Talk a little bit about some of the ways clinicians can bring up the questions of adherence without putting patients on the defense.

Dr. Velligan: I think it is really important to realize that they would not like to take any of this stuff either, and from that perspective, you can have an open dialogue about what patients think about the medicine, what they want to take it for, and how the 2 of you will know whether it is working or not working. You can get patients on board with adherence if you talk about it as a trial that we are going to do together to see whether this works. It helps to tell patients, "The only way I can tell is if you take it regularly during this period, and then we can see what side effects occur and whether or not it is helping your spoken goal to get better, and then we can reevaluate it in the next meeting."

That kind of approach, which is not punitive but collaborative, can really be helpful. Physicians can also preface the discussion by asking, "Everybody forgets to take their medicine; how many times do you think that has happened? Where do you keep your medicine? What happens when you run out? How do you get refills?" A lot of people will be adherent but then are too afraid to take the bus to go get their pills. Lots of things can interfere, and those barriers can be identified if clinicians ask about them.

Medscape: What are the key messages that we need to consider when thinking about adherence in our patients with serious mental illness?

Dr. Velligan: The key messages are the same for anybody with long-standing disorders: What do you want from treatment, and how can we work together to get you there? If you are respectful and you follow that kind of model with any disease, such as diabetes or heart disease, you are going to have a much better outcome than if you just say, "You have high blood pressure -- here, take this medication." That kind of approach doesn't work very well. I don't think that you need to be disease specific; you need to be person specific.

Supported by an independent educational grant from Bristol-Myers Squibb Company.