by Joye C. Anestis
***Thanks to a thoughtful thread on the ABCT listserv for inspiring this post and leading me to this article.***
An important and thoughtful question often posed by clients who are experiencing some symptom relief from treatment is "what is the likelihood that my mental illness will recur?" Just as, say, a cancer patient would want to the know the odds of relapse, so do those being treated for mental illness. The answer to this question can be complicated, as it not only depends on which mental illness you're dealing with but also, importantly, what treatment you received. There is ample evidence that relapse rates differ between treatments. The fact of treatment-influenced relapse rates is another important consideration when presenting therapies to clients...seems only logical to consider using the most effective treatment available with the most hopeful relapse data.
For today, I'm only going to focus on major depression. In 2007, Vittengl, Clark, Dunn, & Jarrett conducted a meta-analysis of studies examining the relapse-recurrence rates in cognitive-behavioral therapy (CBT) for major depression (if you are unfamiliar with meta-analysis, click here for a review). These authors were interested in answering 4 questions (the questions are delineated below). 28 total studies met their inclusion criteria (see the article for details on what these were) and were included in the review.
How common is relapse-recurrence in those who respond to acute-phase CBT (CBT applied during a major depressive episode with the goal reducing depressive symptoms and producing remission)?
- Importantly, an acute treatment is discontinued when goals are met after a certain number of sessions (e.g., 12)...so this data refers to short-term treatment. 13 studies contained data to answer this question. The mean proportion of relapse across these studies was 39% over a mean of 74 weeks. The relapse rate for individual studies varied greatly, suggesting that some other variable could be influencing these findings. The authors identified 7 possible moderators. For the studies that had higher relapse rates, the following factors moderated outcome: longer follow-up periods, use of relapse-recurrence rates from survival analysis instead of simple proportions, assessment of therapist competence, and use of major depression diagnostic criteria in relapse-recurrence definitions. Studies with lower relapse rates reported assessment of therapist adherence, left gaps in time in the follow-up assessment, and utilized a cutpoint in a depression measure in their relapse-recurrence definition. Unfortunately, actual interpretation of any of these moderators is difficult, as they are most likely highly inter-related.
Does acute-phase CBT reduce relapse-recurrence more than other acute-phase treatments?
- Acute-phase CBT vs. acute-phase pharmacotherapy (7 studies): Again, remember that an acute treatment is discontinued after a short period of time. CBT significantly reduced relapse-recurrence compared with pharmacotherapy. In fact, a patient treated with CBT has a 61% chance of a better outcome (no relapse) than one treated with pharmacotherapy. Over a mean of 68 weeks, the relapse rate of CBT 39% and the relapse of pharmacotherapy was 61%.
- Acute-phase CBT + pharmacotherapy vs. acute-phase pharmacotherapy alone (6 studies): Again, the inclusion of CBT is superior. Over a mean of 56 weeks, CBT + pharmacotherapy yielded relapse rates of 38%, whereas the relapse rates for pharmacotherapy alone were 65%. Patients who received the combination treatment had a 61% chance of a better outcome than those treated with pharmacotherapy alone.
- Acute-phase CBT + pharmacotherapy vs. acute-phase CBT alone (3 studies - due to the small number of studies, these findings should be viewed with caution): Adding medication to CBT did not reduce the relapse rate significantly when compared to CBT alone. Over a mean of 61 weeks, relapse-recurrence rates were 33% for CBT alone and 39% for the combination.
- Acute-phase CBT vs. other acute-phase depression-specific psychotherapies (4 studies - again, due to the small number of studies, these findings should be viewed with caution): Relapse rates did not differ significantly between CBT and other depression-specific psychotherapies, as a whole. Looking at the specific therapies, 2 of the 4 studies compared CBT to other behavioral interventions, one compared CBT to interpersonal psychotherapy, and the only study to show a significant difference (lower relapse rate for CBT) used a psychodynamic-interpersonal therapy.
Does continuation-phase CBT (CBT applied to prolong remission and reduce the likelihood of relapse-recurrence...essentially booster sessions of CBT) reduce relapse-recurrence more than nonactive control conditions?
- Relapse rates at the end of continuation CBT (4 studies): Continuation CBT significantly reduces relapse rate compared to non-active control conditions. Averaging over 41 weeks, continuation CBT had a relapse rate of 12% and the nonactive control condition had a relapse rate of 38%.
- Relapse rates after discontinuing continuation CBT (5 studies): Here again, continuation CBT is important, significantly reducing relapse rates compared to nonactive controls. Over an average of 153 weeks, the relapse rate of continuation CBT was 40%, compared to the 73% relapse rate for nonactive controls.
Does continuation-phase CBT reduce relapse-recurrence more than other active treatments?
- Relapse rates at the end of continuation treatment (5 studies): The difference was non-significant (p < .06); however, the relapse rate of CBT was 10%, compared to a 22% relapse rate for active controls, over a mean of 27 weeks.
- Relapse rates after discontinuing continuation treatment (8 studies): Here, continuation CBT reduced relapse significantly. Over a mean of 114 weeks, the relapse rate of CBT was 42% and the relapse rate for active controls was 61%.
So, relapse is common in any treatment for depression, but the rates are significantly lower for CBT than for medication...and providing booster sessions of CBT after acute treatment serves to further decrease the relapse rates. The authors interestingly place these findings in a public health perspective (I personally love it when researchers do this, because I find it really drives the point home)...so here's what these findings really mean:
- Based on current epidemiological data, approximately 35,000,000 people have major depression each year in the U.S. (Kessler et al., 2003)...
- And approximately 16% (5,600,000) of depressed folks receive adequate pharmacotherapy (Young et al., 2001)
- Furthermore, approximately 50% (2,800,000) of patients with depression respond to acute-phase treatment (Hollon et al., 2005) and
- Approximately 72% (2,016,000) of patients discontinue pharmacotherapy within 90 days (Olfson et al., 2006).
- Then the potential savings of treating these folks with acute phase CBT is about 448,000 relapses annually!
Considering the cost (both time and money) of long-term pharmacotherapy and the lack of desire in patients to take medication in the long-term, acute CBT with booster sessions seems the best way to combat relapse.