News
Online CBT for Depression, GAD Is Effective for Most
Major Finding: Among patients described as "responders," 75% of those with depression and 80% with generalized anxiety disorder showed improvement...
SAN FRANCISCO – The search is on for ways to refine cognitive-behavioral therapy for generalized anxiety disorder in order to improve upon current relatively modest success rates.
Cognitive-behavioral therapy (CBT) is less effective for generalized anxiety disorder (GAD) than for the other anxiety disorders. Only about one-half of patients are improved post-treatment, and less than one-third reach recovery, noted Richard E. Zinbarg, PhD, at the annual conference of the Anxiety and Depression Association of America.
Dr. Richard E. Zinbarg
At a session on advances in treatment of GAD, investigators presented randomized clinical trials assessing a variety of specific strategies aimed at enhancing the effectiveness of CBT in evidence-based fashion. The trials included a study of the impact of having patients keep a worry outcome journal, an exploration of the potential deleterious effects of a phenomenon known as relaxation-induced anxiety, and a study of the effectiveness of emotion regulation therapy, a relatively recent form of psychotherapy that’s part of the so-called “third wave” of CBT.
Worry outcome journal
Lucas LaFreniere observed that while CBT has been broadly shown to be effective for GAD, the various forms of CBT are packages of components that often include psychoeducation, stimulus control, behavioral experiments, exposure, cognitive reframing, relaxation training, and other elements in various combinations and sequences. Almost none of these specific components has been evaluated formally to learn whether they are pulling their weight therapeutically and making a positive contribution to outcomes.
Mr. LaFreniere, a doctoral student in clinical psychology at Pennsylvania State University in Hershey, presented a randomized trial of one such component, worry outcome monitoring, which currently is incorporated in some but not all CBT programs for GAD. Mr. LaFreniere and his coinvestigators developed a version of worry outcome monitoring they dubbed the worry outcome journal, or WOJ, which he characterized as “a brief ecological momentary intervention for worry.”
Lucas LaFreniere
The WOJ works like this: At four random times per day, WOJ users receive a phone message to drop what they’re doing and record on a chart what they’re currently worrying about. They briefly note the date and time, the content of their worry, the distress it’s causing on a 1-7 scale, how much time they’re spending thinking about it, and their prediction as to the likelihood that this negative event actually will come to pass, which by the nature of their illness generally is unrealistically sky high early on in treatment. Later, they return to record whether the worrisome outcome occurred. The WOJ data are often reviewed in session.
The hypothesis was that the WOJ would reduce worry by aiding GAD patients in attending to their worries more thoroughly and objectively, recognizing in the moment the high cost of their worrying in terms of distress and cognitive interference, forming more realistic predictions about the future, and changing their conviction that excessive worrying is a worthwhile use of their time.
“One thing that particularly motivates me as a treatment researcher is the idea that those with GAD could be making themselves chronically miserable in an effort to protect themselves from future catastrophes that likely are not even going to happen. That’s a lot of human suffering that isn’t necessary. What we can do to help with that, we should do,” Mr. LaFreniere said.
On the other hand, this was a matter that cried out for a controlled trial because of the possibility that attempts to reduce worry might have unintended harmful consequences.
“Those with GAD have positive beliefs about worry. They believe it’s useful: it motivates, buffers emotional shifts, facilitates problem solving, and marks you as caring and conscientious – good personality traits,” he explained.
His study included 51 GAD patients randomized to 10 days using the WOJ or to a thought log control condition in which prompted by their cell phone four times daily, they recorded whatever everyday thought was on their mind at the moment. An example drawn from personal experience, Mr. LaFreniere said, might be “I love enchiladas!”
Outcome measures evaluated at baseline, again at 10 days upon conclusion of the intervention, and finally at 30 days of follow-up were the Penn State Worry Questionnaire, the GAD Questionnaire for DSM-IV, and the Meta-Cognitions Questionnaire subscales for positive beliefs about worry, uncontrollability of one’s thoughts, and negative beliefs about worry.
“The big reveal was that 91% of their worries did not come true,” he reported.
The primary outcome was reduction in worries as measured by the Penn State Questionnaire. The WOJ group showed a significant reduction, compared with controls, immediately post-treatment – which remained significant, albeit attenuated to a moderate effect size, at 30 days.
At day 10, 18 of 29 WOJ users no longer met diagnostic criteria for GAD, compared with 6 of 22 controls. By day 30, however, there was no significant between-group difference on this secondary endpoint.
The WOJ group showed a significantly greater reduction than controls on the secondary endpoint of uncontrollability of beliefs at both days 10 and 30.
“The WOJ may be a viable ecological momentary intervention for reducing worry in GAD. Therapist-free use of WOJ led to decreased worrying after only 10 days. It’s quite possible that longer practice may yield even stronger results. After all, for a normal CBT protocol, we’re looking at 8-20 weeks of treatment,” Mr. LaFreniere observed.
“I’d like to underscore that there was no harm done: The WOJ didn’t increase detrimental beliefs about worry,” he added. “We had a worry ourselves as researchers – disconfirmed by the trial – that patients may take the non-occurrence of their worries as some kind of proof that worry prevented those bad things from happening.”
Mr. LaFreniere is interested in studying the WOJ for worry reduction in non-GAD populations.
“Worry can be very high in other anxiety disorders, major depressive disorder, bipolar disorders, and in insomnia. The WOJ is highly cost-effective and easy to disseminate. It could very easily be made into a smartphone app,” he said.
Relaxation-induced anxiety
Relaxation training often is incorporated in treatment packages for GAD. Yet, it’s possible that one reason CBT is only modestly effective for GAD is because of relaxation-induced anxiety (RIA), an understudied phenomenon defined as a paradoxical increase in the physiological, behavioral, and cognitive aspects of anxiety when a person tries to relax.
“It has been theorized that individuals who are especially concerned with maintaining control over physical and psychological processes find relaxation vulnerable, unpleasant, and activating. Thus, discomfort with perceived lack of control during relaxed moments – an inability to let go – may result in unsought increase in anxiety during therapeutic attempts at relaxation,” according to Michelle G. Newman, PhD, professor of psychology at Penn State.
Dr. Michelle G. Newman
She presented a secondary analysis of a published randomized clinical trial she coauthored (J Consult Clin Psychol. 2002 Apr;70[2]:288-98) in which 41 participants with GAD were assigned to CBT with relaxation therapy using standard progressive muscle relaxation techniques or to self-control desensitization. Relaxation therapy and relaxation-induced anxiety ratings were recorded at each session. Outcomes were assessed post-treatment and at 6, 12, and 24 months of follow-up using the Penn State Worry Questionnaire, the State-Trait Anxiety Inventory, the Hamilton Anxiety Rating Scale, and the Clinician Severity Rating for GAD symptoms. In addition, immediately after each in-session relaxation practice, patients were asked to rate on a 9-point scale how much they noticed an increase in anxiety during the relaxation session.
All subjects improved significantly, but those with a lower peak RIA – defined as the highest level of RIA experienced in any of the 14 treatment sessions – had significantly fewer GAD symptoms at the end of therapy as well as at 2-year follow-up. Peak RIA was unrelated to baseline GAD symptom severity or change over time in anxiety symptoms. However, patients whose peak RIA occurred during the last several treatment sessions showed less improvement in GAD symptoms at the conclusion of treatment than those whose peak came earlier.
The clinical implications of these findings are that therapists who use progressive muscle relaxation in the treatment of GAD should assess RIA at the conclusion of every session, and if a patient reports moderate or higher RIA, the duration of the relaxation training portion of therapy should not be shortened until after several consecutive sessions of lower RIA have been reported, according to Dr. Newman.
Emotion regulation therapy
Megan E. Renna brought attendees up to speed on emotion regulation therapy (ERT), a third wave variant of CBT that incorporates principles from more traditional CBT, such as skills training and exposure, supplemented by teaching emotion regulation skills. Those skills include the development of present moment awareness and cultivation of compassion. Both are grounded in research on motivational and regulatory learning mechanisms related to threat vs. safety and reward versus loss.
As detailed in a recent review article for which she was first author (Front Psychol. 2017 Feb 6;8:98), ERT is a manualized, mechanism-targeted treatment for what she termed “distress disorders”; namely, GAD and major depressive disorder, which are highly comorbid, share key underlying temperamental features, and for whom adequate therapeutic success is all too often elusive. ERT appears to be particularly useful during the emerging adulthood years and across a broad range of ethnic and racial patients, according to Ms. Renna, a PhD student in clinical psychology at Hunter College in New York.
The efficacy of the original 20-session, individual therapy version of ERT was established in a study of 20 GAD patients, half of whom also had major depression (Depress Anxiety. 2015 Aug;32[8]:614-23). But Ms. Renna said ERT’s developers – Douglas S. Mennin, PhD, of Hunter College, and David M. Fresco, PhD, of Kent State (Ohio) University, are interested in determining the minimum effective therapeutic dose of ERT. They have conducted an open randomized trial of a 16-session version of ERT in which the results proved similar to those seen with 20 sessions. Now they’re carrying out a study of 8 vs. 16 sessions. The study is ongoing, but at first look, the results with 8 sessions of ERT appear similar to 16, Ms. Renna said.
None of the speakers reported having any financial conflicts of interest.