Blog

Fernandez, E., Salem, D., Swift, J. K., & Ramtahal, N. (2015, August 24). Meta-analysis of dropout from cognitive behavioral therapy: Magnitude, timing, and moderators. Journal of Consulting and Clinical Psychology. Advance online publication.

“Dropping out” refers to clients who discontinue therapy prematurely and against professional advice. In contrast, “refusing” refers to clients who do not start a therapy that is made available to them. Together, both dropping out and refusing are referred to as “attrition” from therapy. Attrition is a problem for clinicians because of loss of revenue and time, and a problem for clients because their mental health needs remain unmet. In a previous meta analysis that included 669 studies, Swift and Greenberg (2012) reported that the average drop out rate across all therapies was 19.7%. In this meta analysis, Fernandez and colleagues looked specifically at drop outs and refusers in cognitive behavioral therapy (CBT). The authors reviewed 115 studies that reported drop outs, 36 of which also reported on the number of participants who refused treatment before starting. The average percent of patients who refused CBT prior to starting treatment was 15.9%, and the average percent of patients who dropped out after starting CBT was 26.2%. So the total average attrition rate was 42.1%. Compared to any other disorder, patients with depression were significantly more likely to refuse CBT (21.6%) or to drop out (36.4%). It is possible that depressed patients have a harder time summoning the energy to participate in therapy, and experience lower hope, greater social withdrawal, and lower motivation once they initiate CBT. For those receiving e-therapies (e.g., internet, phone, and CD-based treatments), pre-treatment refusal rates were 10% to 15% higher than individual or group CBT, and drop outs from e-therapies were 10% higher compared to individual or group CBT. Those offered e-therapy might be ambivalent about its utility, the therapeutic alliance might be limited, and they might have a lower sense of engagement in the therapeutic process. Finally, a greater number of planned therapy sessions was related to lower attrition rates. Perhaps the promise of more sessions raised clients’ hopes of achieving better outcomes.

Practice Implications

These findings suggest that engaging and encouraging clients to participate in the therapy may have to start even before therapy begins. This may involve enhancing readiness by means of motivational interviewing, for example. Clients who are depressed are particularly likely to refuse treatment or drop out, and so clinicians must pay particular attention to the level of motivation and engagement of depressed clients. Although e-therapies are promising in that they may allow a therapist or agency to reach more people including those who live in remote areas, the attrition rate of e-therapies may be unacceptably high. Attrition may lead to demoralization and lowered expectations for treatment among these patients, which may negatively impact future treatment. Perhaps e-therapies should not be considered as a first-line treatment for those who can easily access individual or group therapy. Alternatively, the high attrition rates of e-therapies may be reduced by supplementing the intervention with some in-person therapy sessions to enhance engagement and a therapeutic alliance.

Weitz, E.S., Hollon, S.D., Twisk, J., van Straten, A., Huibers, M.J.H., David, D., …. Cuijpers, P. (2015). Baseline depression severity as moderator of depression outcomes between cognitive behavioral therapy vs pharmacotherapy: An individual patient data meta-analysis. JAMA Psychiatry. doi:10.1001/jamapsychiatry.2015.1516.

The American Psychiatric Association guidelines for the treatment of depression indicates that although psychotherapy is adequate for mild to moderate depression, anti-depressant medications are indicated for the treatment of severe depression in major depressive disorder. These recommendations are mainly based on the findings of the National Institute of Mental Health Treatment of Depression Collaborative Research Program that was published in the mid 1990s. Several authors since then have disputed this claim, but no meta-analyses have been done on the studies of head-to-head patient-level comparisons of psychotherapy vs antidepressant medications for the purpose of evaluating their relative efficacy for severity of depression. In this meta analysis, Weitz and colleagues look at medications vs psychotherapy for depression and then evaluate if initial severity of depressive symptoms helped to explain any differences. The authors looked at all studies that compared cognitive behavioral therapy (CBT) against antidepressant medications for depression. They focused on CBT because it was the most often studied of the psychotherapies in this context. A systematic review turned up 24 studies, and they were able to get original patient-level data from the authors of 16 of the 24 studies. This represented over 1,700 participants with major depression. These 16 studies were no different from the 8 studies that did not provide original data. Between 17% and 54% of the 1,700 depressed participants met criteria for severe depression at pre-treatment. There were no significant differences between antidepressant medications and CBT on clinically relevant outcomes in terms of “response” (i.e., improvement) or “remission” (i.e., symptom-free). In total, 63% of patients in the antidepressant medication condition and 58% of patients in the CBT condition responded to treatment, and 51% of patients in the antidepressant medication condition and 47% of patients in the CBT condition met criteria for remission. Most importantly, the effects of CBT and antidepressant medications on response to treatment or remission did not differ based on initial severity of depressive symptoms.

Practice Implications

Patients with severe depression were no more likely to require medication to get better than patients with less severe depression. This meta analysis that included the majority of studies that exist on the topic found no evidence to support the guidelines that severe depression should be treated with antidepressant medications over psychotherapy. The authors conclude that CBT may also be a first-line treatment for severe depression.

Tao, K. W., Owen, J., Pace, B. T., & Imel, Z. E. (2015). A meta-analysis of multicultural competencies and psychotherapy process and outcome. Journal of Counseling Psychology, 62(3), 337-350.

Cultural factors shape health-related beliefs, behaviors and values. For decades, many have argued that therapist multicultural competence shapes the therapy process and affects patient outcomes. Some therapists have poorer outcomes with patients of racial/ethnic minorities compared to White patients. Multicultural competence refers to the ability to work effectively across many groups including minority groups. In 2008, an American Psychological Association Task Force detailed recommendations for multicultural competencies. Multiculturally competent providers are those who: expand their knowledge of their client’s background, use culturally relevant interventions, and gain awareness of their own assumptions and the impact of these on their therapeutic work. In this meta analysis, Tao and colleagues aimed to assess the relationship between multicultural competence in therapists with therapy processes and client outcomes. They reviewed 18 studies that included over 1600 clients, the vast majority of whom identified as a racial/ethnic minority. Therapist multicultural competence was assessed by client self report. Therapist multicultural competence was highly correlated with therapy processes like: therapeutic alliance (r = .61), client satisfaction (r = .72), and session depth (r = .58). The association between therapist multicultural competence and client symptom outcomes were moderate in size but significant (r = .29). A separate analysis showed that the relationship between multicultural competence and therapy process variables (alliance, satisfaction, depth) were significantly larger that associations with client outcomes.

Practice Implications

Therapists’ abilities to integrate aspects of their client’s cultural narrative into their interventions significantly accounted for difference in outcomes. In other words, clients who perceived their therapist as more culturally sensitive had better outcomes. This was likely related to more positive therapeutic processes (i.e., alliance, satisfaction, session depth) between clients and therapist dyads, within which clients perceived the therapist as multiculturally sensitive. A provider’s ability to recognize how their own personal backgrounds influence their own and clients’ behaviors will result in better therapy processes and improved client outcomes.

Author email: k.tao@utah.edu

Clarke, K., Mayo-Wilson, E., Kenny, J., & Phillig, S. (2015). Can non-pharmacological interventions prevent relapse in adults who have recovered from depression? A systematic review and meta-analysis of randomised controlled trials. Clinical Psychology Review, 39, 58-70.

According to the American Psychiatric Association, the risk for relapse from depression can be as high as 60% for those who had one episode, 70% for those who had two episodes, and 90% for those who had three previous episodes. Intervening after recovery from an episode of depression might prevent relapse. A relapse is defined as any significant deterioration in depression following a period of clear improvement. We know that relapse after discontinuing antidepressant treatment is greater than relapse after discontinuing psychotherapy, likely because psychotherapy and not medications result in the patient acquiring new coping skills and strengths. Clarke and colleagues conducted a meta analysis of psychological interventions that were designed or adapted in order to reduce relapse after the acute phase depression. These include mindfulness based therapy (MBT) which helps individuals process experience without judgment by using mindfulness techniques; cognitive behavioral therapy (CBT) which helps to modify thoughts and behaviors key to depression; and interpersonal psychotherapy (IPT) which focuses on helping to deal with interpersonal and social role problems related to depression. Clarke and colleagues reviewed 29 studies that included 4216 participants who had at least one episode of depression, had recovered after treatment, and who received either MBT, CBT, or IPT to prevent relapse. These were compared to control conditions that included wait-lists, treatment as usual, or some other active intervention. Compared to all of the controls, MBT, CBT, and IPT reduced relapse rates from 21% to 25% among patients one year post acute treatment. The effects for CBT were maintained up to two years post treatment. There were no differences between psychotherapies and control conditions in drop out rates.

Practice Implications

Psychotherapies (e.g., MBT, CBT, and IPT) reduce relapse from depression by about 22%. up to one year post recovery. Practitioners should consider offering MBT, CBT, or IPT as a form of booster sessions to reduce the likelihood of relapse from a previous episode of depression. Such interventions are important given the increasing relapse rates for each subsequent episode of depression.

Bradley, R., Greene, J., Russ, E., Dutra, L., & Westen, D. (2005). A multidimensional meta-analysis of psychotherapy for PTSD. American Journal of Psychiatry, 162, 214–227.

The psychotherapy research literature on treatment of post traumatic stress disorder (PTSD) has focused on cognitive behavioral therapy (CBT, with exposure and/or cognitive restructuring) and eye movement desensitization and reprocessing (EMDR). Exposure therapy involves confronting memories of the trauma or cues related to the traumatic event. Other CBT skills include developing skills for anxiety management or challenging distorted cognitions. In EMDR the patient is asked to develop an image of the traumatic event while tracking a bilateral stimulus. Most studies demonstrate the effectiveness of CBT for PTSD in the short term. However, many studies have excluded patients with comorbid conditions. For example, patients with PTSD often also have significant other symptoms like depression, substance abuse, other anxiety disorders, and personality disorders. In this meta analysis, Bradley and colleagues were interested in documenting the overall efficacy of psychological treatments for PTSD. They also wanted to report on any evidence on the long term efficacy of treatments for PTSD, and on evidence of the effects of excluding patients with comorbid disorders. Bradley and colleagues included randomized controlled trials published between 1980 and 2003 (i.e., 26 studies representing 1,535 patients). Also, they looked at outcomes defined in a few ways: change in symptoms as documented by the effect size, proportion of patients no longer meeting diagnostic criteria for PTSD (but who may have residual symptoms), and proportion whose symptoms improved significantly. Across all treatments, the average pre to post effect size was large (d = 1.43), and comparisons to control conditions were also large (d = .83). The results suggested that psychotherapy produced substantial effects for PTSD. Differences between types of therapy (CBT, CBT with exposure, EMDR) were negligible. Fifty six percent of patients no longer met criteria for PTSD, and 65% showed improved symptoms. At follow ups, 62% no longer met diagnostic criteria for PTSD and 32% were deemed improved, but the number of studies with follow up data were small (k = 10) and so the results could be unreliable. Of those who started treatment, 78.9% completed the therapy. Of those who were assessed, 30% were excluded because of suicide risk, drug or alcohol abuse, or “other serious comorbidity”.

Practice Implications

Treatment guidelines from the International Society for Traumatic Stress Studies list a number of effective treatments for PTSD. The evidence for efficacy is strongest at post treatment, and more research is necessary to demonstrate efficacy in the longer term. There is currently little evidence that any one treatment approach is more effective than another, and some researchers are debating whether specific interventions like exposure is necessary to treat PTSD. Bradley and colleagues argue that we need more research on alternative treatments for PTSD and research on patients with multiple symptoms and comorbidities.

Author email: rbradl2@emory.edu

Mallinckrodt, B. & Jeong, J. (2015). Meta-analysis of client attachment to therapist: Associations with working alliance and pretherapy attachment. Psychotherapy, 52, 134-139.

Attachment theory has become one of the most important conceptualizations of affect regulation and interpersonal relationships. John Bowlby and others suggested that attachment behaviour is hard wired so that infants can gain proximity to caregivers which is necessary for infant survival. Repeated interactions with caregivers coupled with the variety of caregiver responses (i.e., available, unavailable, or inconsistently available caregiving) lead to children developing internal working models of attachment. These models become the basis for attachment styles in adulthood. Attachment security in adults is associated with the ability to give and receive caring and love, and to adaptively regulate emotions. Attachment avoidance is associated with a tendency to dismiss relationships as important, and to downregulate emotional experiences. Attachment anxiety is associated with a preoccupation with relationships, and to upregulated emotional experiences. In a previous meta analysis, client general attachment security was modestly but significantly associated with higher levels of therapeutic alliance (r = .17). In another meta analysis, higher client general attachment anxiety was associated with poorer client outcomes (r = -.22). In this meta analysis, Mallinckrodt and Jeong assessed whether client attachment to the therapist was associated with client general attachment style and with the therapeutic alliance with the therapist. They included 13 studies representing 1051 client-therapist dyads. Client pre-therapy general attachment avoidance and anxiety were negatively associated with client-therapist attachment security, and the effects were modest but significant (r = -.12, r = -.13). Client-therapist attachment security was positively associated with therapeutic alliance (r = .76) and client-therapist attachment avoidance was negatively associated with therapeutic alliance (r = -.63), and these effects were large.

Practice Implications

Client pre-therapy attachment styles appear to have an impact on their attachment to the therapist. A client pre-therapy attachment style characterized by preoccupation with relationships and an over-emphasis on emotions (i.e., attachment anxiety) will likely lead to similar behaviors and preoccupations in the relationship with the therapist. Mallinckrodt and Jeong suggest that this might be the basis for transference-related phenomenon that therapists and clients experience in the therapeutic relationship. That is, client attachment anxiety and avoidance likely interfere with developing a secure attachment to the therapist. Further, client attachment avoidance with regard to the therapist may result in lower therapeutic alliance, which is key to achieve improved patient outcomes. Despite these challenges, therapists who can facilitate a secure psychotherapy attachment experience for their clients are more likely to see improvements in their clients’ functioning.

Author email: bmallinc@utk.edu