Evaluating Treatment Response in Major Depressive Disorder

According to Mark Zimmerman, MD, professor of psychiatry and human behavior at Brown University, the FDA indication for antidepressants does not require improvement in functioning, quality of life, ability to cope with stress, or general well-being. However, recent research has shown that patients consider non-symptom factors just as important in evaluating treatment response. So how should clinicians gauge treatment response? 

Dr Zimmerman and his colleagues created the remission from depression questionnaire to help with this discrepancy. In this podcast excerpt, Dr Zimmerman examined the conflicting research, the creation of the questionnaire, and how clinicians can use it in their practice to better evaluate treatment response. 

Read the Transcript: 

The paradigm in treatment research for major depressive disorder for the past several decades has been to evaluate outcome in terms of symptom improvement. The FDA, as you noted in asking me, requires only that symptoms improve without regard to any other fundamental factors that are important to consider when evaluating the efficacy and effectiveness of treatment of depression. Factors such as one's coping ability, one's functioning, one's quality of life, and positive mental health. There are several lines of research to suggest that these are the factors that are of greatest interest to patients. About 20 years ago, we were very interested in the whole idea of how do you define remission from depression? About 30 years ago, there was a consensus conference in which they came up with a cutoff score on the Hamilton Depression Rating Scale, indicating that this was representative of remission, meaning individuals who scored below this factor were considered to be in remission. And they also called for research to be done to determine whether that cutoff point was valid.

We were interested in the concept of remission and decided to ask patients what did they consider to be the most important factors to define remission? So we had over 500 patients in our practice complete a questionnaire on which they rated 16 different factors as to how important they thought that factor was in determining whether or not they or someone else was in remission from depression and also circle the item that they thought was the best indicator of remission from depression. And what we found was that symptom resolution came in fourth on one measure, sixth on another. By that I mean the frequency by which each of the factors was rated as being important in determining remission. Symptom resolution was number 6.

As for the most important factor in determining remission, symptom resolution was number 4. Rather, what patients care most about is improvement in functioning, functioning normally, a sense of well-being, improved coping ability. There was a study across 8 countries of more than 2000 patients who were taking antidepressants, and they asked individuals what were the factors that they considered most important in determining whether or not medication was helpful for them or treatment was helpful for them? And what patients were indicating was it's stresses in their life, dealing with those stressors, coping with those stressors, various issues in their life. Symptom improvement was important, but much less frequently so than these other factors.

A similar study was done of individuals who were receiving cognitive behavior therapy. What was particularly interesting in that study, because that was a study in which both depressed patients without comorbid anxiety disorders and patients with anxiety disorders without comorbid depression were assessed. For the depressed patients receiving therapy, they were most interested in achieving personal goals, resolving personal issues, dealing with various stressors. For the anxiety disorder patients, the number 1 concern was resolution of symptoms, improving their anxiety or fear. So with that in mind, we've been interested in when patients present for treatment, evaluating multiple factors and asking patients who are in treatment a simple question, Do you think you're in remission from depression? And then we associated that. We examined what predicted that? Symptom severity? Functioning? Quality of life? Well, in fact, all 3 predict it. And when you control for 1, all 3 remain significant in predicting it.

So with these multiple lines of research, we were interested in developing a different type of outcome measure, and we developed the remission from depression questionnaire that assessed these multiple factors in evaluating treatment outcome for depression. Of course, symptoms of depression but also other types of symptoms such as symptoms of anxiety and pain, as well as coping ability, functioning ability, quality of life, life satisfaction, positive mental health. So we developed this multidimensional questionnaire. The first study that we did on this questionnaire was to have patients fill it out as well as fill out a symptom scale: the QIDS, the quick inventory of depressive symptoms, and then a third questionnaire in which they made an evaluation of the 2 questionnaires. And we asked them which was more burdensome, which took longer to complete, as well as asking them, "Which one do you think better represented the concerns you had in getting treated for depression? Which of these scales better reflected the goals that you had in your treatment for depression? Which of these scales would you prefer to complete when being treated for depression?" So in terms of burden, both scales were judged to be equal. But in terms of preference, patients clearly indicated that the multidimensional scale was preferred and better represented what was going on with respect to their treatment of depression than strictly a symptom scale. So we were reassured by that finding.

Then we went on to look at the validity and reliability of the scale. Once again, we had patients complete both the RDQ and the QIDS. We also rated close to 300 patients on the Hamilton. These are patients in our ongoing practice. And we also asked patients "do you think you're in remission or not?"

What we found was that when you correlated the symptom scale with remission status as measured by the Hamilton, both scales were equally correlated. When you controlled for the other scale, there was no difference between the scales. However, when you asked patients "do you think you're in remission?" both scales were associated with that variable. But when you controlled for the factors on the RDQ, the symptom scale was no longer associated with patient sense of remission, whereas when you controlled for symptoms, the RDQ was still significantly associated with patients' self-assessment of remission, thereby suggesting that patient's understanding of their own sense of whether or not they're in remission from depression goes beyond simply symptoms.

We replicated that finding in a longitudinal study. We had patients complete the scales at baseline and then 4 months later and found both a symptom scale as well as the RDQ resulted in large effect sizes. And once again, with respect to predicting patients' self-assessment of remission status, the multifactorial measure was associated after controlling for symptoms, but the symptom measure was not associated with patients' self-determination or self-evaluation of remission status after controlling for the multifactorial scale. Again, concluding that a multifactorial approach towards defining outcome better reflected patient's perception of their status.

One last study related to this. This is a study that has been reviewed, requires minor revision. We're about to resubmit for publication. This was a study in a partial hospital program rather than outpatients in which we give the patients to scale at the beginning of treatment and at the end of treatment, a study of over 500 depressed patients in our partial hospital program. And also at the end of treatment we have patients rate globally, "How much better are you feeling now? Are you any better? If you are better, how much better are you feeling?"

What we found was that all of the subscales of the RDQ are associated with patients' global assessment of improvement. But when you do a regression analysis controlling for the association between the different variables, coping ability and general well-being predicted global assessment of outcome but not depressive symptoms.

Again suggesting that what patients are most interested in is not necessarily symptom resolution. Of course, they want symptom resolution, but it's not the most important factor that depressed patients consider when evaluating the effectiveness of treatment. So from our perspective, the FDA's focus on symptom resolution as the 1 and only factor to consider in determining whether a treatment should be approved for the treatment of depression is too narrow. I think other factors need to also be considered. So in the same way that the FDA requires medications for hypercholesterolemia or hypertension to not only lower cholesterol levels in blood pressure, but also reduce future morbidity and mortality, I think a medication that only improves depressive symptoms but does not improve functioning quality of life, and not just medication, although the FDA only approves medications, it is something that patients would be less interested in.

Mark Zimmerman, MD, is a professor of psychiatry and human behavior at Brown University and director of the Partial Hospital Program and Outpatient Practice at Rhode Island Hospital. Dr Zimmerman received his undergraduate degree from Columbia University and his medical degree from Chicago Medical School. He completed his postgraduate training at the Medical College of Pennsylvania and held an academic appointment as an assistant professor at the same time he was a resident in psychiatry.