تاثیر خصیصه های بیماران از اثرات فوری درمان افسردگی در اثر فعال سازی رفتاری و داروهای ضد افسردگی طولانی مدت

Patients’ attributions of effects of treatment are important, as these can affect long-term outcome. Most studies so far focused on the influence of attributions to medication for anxiety and depression disorders. We investigated the effects of patients’ attributions made after acute treatment on the long-term outcome of antidepressant medication (ADM) and psychological treatment (behavioural activation, BA). Data are based on a randomized trial testing the effectiveness of BA vs. ADM for major depression (MDD) in Iran. Patients with MDD (N = 100) were randomized to BA (N = 50) or ADM (N = 50). Patients’ attributions were assessed at post-test (after completion of the treatments). Scores on an attribution questionnaire were factor analysed, and factor scores were retained as predictors of depressive symptoms at 1-year follow-up. Regression analysis was used to test whether attributions predicted depressive symptoms at 1-yr follow-up, controlling for symptom level, condition, and their interaction at post-test. Belief in coping efficacy was the only attribution factor significantly predicting 1-year HRSD scores, controlling for condition, post-test HRSD and their interaction. It also mediated the condition differences at follow-up. Credit to self was the single attribution factor that predicted BDI follow-up scores, controlling for condition, posttest BDI, and their interaction. It partially mediated the condition differences on the BDI at follow-up. Attribution to increased coping capacities and giving credit to self appear essential. In the long-term (at 1 year follow-up), the difference in outcome between BA and ADM (with BA being superior to ADM) is at least partially mediated by attributions.

Antidepressant medication (ADM) is a standard treatment for depressed patients in current psychiatric guidelines (American Psychiatric Association, 2000; (Frank, et al. 1990) and the most recent practice guideline for the initial treatment of patients with mild to moderate major depressive disorder (MDD) is antidepressant medication and depression-focused psychotherapy. For depressed patients with severe MDD with or without psychiatric features however, ADM is the first choice (American Psychiatric Association, 2010). The short-term effectiveness of ADM is well studied and comparable to that of CBT and IPT, although dropout from ADM is higher (Cuijpers et al., 2008). Much less is known about the long-term effectiveness of ADM, and how it compares to that of psychological treatment. A recent meta-analysis reported a trend towards superiority in relapse prevention of CBT compared to maintenance of ADM over 5 studies (OR = 1.62; p = .07). The superiority of CBT over ADM became significant after exclusion of one outlier, OR = 1.77, p < .05 (Cuijpers et al., 2013). The same meta-analysis reports clear evidence of superiority of CBT over ADM when ADM is discontinued after the acute treatment phase over eight studies, OR = 2.61, p < .001 (see also Imel et al., 2008). Thus, when patients stop taking antidepressant medication, those who recovered from their depressive episode are at a substantial risk for recurrence, whereas CBT appears to offer a better protection for future relapse. The superior effects of CBT over ADM in relapse prevention seem to hold for both the Beckian approaches (Hollon, et al. 2002); Hollon et al., 2005a and Hollon et al., 2005b; Dobson et al., 2008) and for Behavioural Activation (BA) (Dobson et al., 2008; Moradveisi et al., 2013a and Moradveisi et al., 2013b). The important question then arises: what explains the apparent superior long-term effects of psychological treatment over ADM? It has been argued before that where ADM only alleviates depression symptoms as long as the medication is used, patients in psychotherapy actually learn to get better and stay well (Paykel et al., 2006; Hollon et al., 2005a and Hollon et al., 2005b). More specifically, it was found that the skills that patients acquire in CT actually predict the prevention of relapse after treatment (Strunk et al., 2007). From a behavioural activation point of view, a likely reason of relapse after discontinuation medication is that patients did not change their coping skills. The lack of reinforcement, patterns of avoidance and rumination might still exist, although antidepressant medication might reduce temporarily their effects on mood. In contrast, those patients treated with behavioural activation have acquired healthy behavioural skills and new coping styles that might reduce relapse (Moradveisi et al., 2013a and Moradveisi et al., 2013b). Another explanation is that patients’ beliefs about why they recovered in therapy (attributions) impact the sustaining of gains. It has been postulated by Brewin and Antaki (1982) that patients who attribute gains to their own efforts are more likely to sustain those gains compared to those who attribute improvement to external causes such as a drug’s activity or a therapist’s charisma. A study by Basoglu et al. (1994) investigated attributions made by patients with panic disorder and agoraphobia who had participated in an RCT comparing 8 weeks of alprazolam or placebo (medication treatment) plus exposure or relaxation (psychological treatment; relaxation being the “psychological placebo”). At the end of 8 weeks of treatment, 40 patients who much/very much improved assessed how much they attributed their gains to medication or to their own efforts. At the treatment-free follow-up in week 43, those who at week 8 had attributed their gains to medication and felt less confident about coping without medication had more severe withdrawal symptoms and a higher loss of gains in comparison to those who at week 8 had attributed their gains to their own efforts during treatment. Another study by Biondi and Picardi (2003) that investigated panic disorder with agoraphobia reported similar results. They found that 60% of the patients with panic disorder who attributed improvement to medication in a combined medication-psychotherapy treatment relapsed, whilst those who attributed improvement to the self-reported no relapse. Although similar attributional processes have been hypothesized to play a role in the differential long-term effects of CBT vs. ADM in depression treatment, no study so far assessed this to the best of the present authors’ knowledge. Behavioural activation (BA) is a relatively new treatment for patients with major depressive disorder (MDD) (Jacobson et al., 1996). Recent studies have shown that BA is an effective treatment for depression that might even be more effective than cognitive therapy in severely depressed patients (Dimijian et al., 2006). To date, no study investigated the effects of attribution to medication and attribution to the self on treatment effects of BA in comparison to antidepressant medication (ADM) for participants with MDD. The data presented in this paper are drawn from a randomized controlled trial comparing BA and antidepressant medication (Sertraline) for patients with MDD, in which BA proved to be superior to ADM (Moradveisi et al., 2013a and Moradveisi et al., 2013b). The focus of this paper is on whether depressed patients’ attributions of treatment effects (i.e. to the medication or to the self), impact the long-term effects of treatment, assessed after approximately one year. If it is true that CBT has better long-term effects than ADM because of attribution of improvement to controllable factors in the self instead of to external factors such as medication, two predictions follow. (1) Attribution of treatment effects to the self will predict better long-term effects of treatment, even after controlling for the short-term effects. In contrast, attribution of treatment effects to medication will not be associated, or negatively associated, with long-term treatment effects. (2) Attribution of treatment effects to the self will mediate the long-term differences between BA and ADM that were observed in our trial. We tested the first prediction by assessing participants’ beliefs about factors explaining improvement after treatment, and testing their predictive power in explaining long-term depressive complaints, assessed at 49 weeks, whilst controlling for the level of these complaints as assessed immediately after treatment (week 13). The second prediction was tested by formal mediation tests, investigating whether attributions statistically mediated the difference between conditions in long-term effects, even when controlling for the short-term effects of treatment. Implicated in the attribution mediation hypothesis is that attributions that play a role in explaining the differences between BA and ADM on the long-term effects should differ significantly between conditions; we therefore also tested whether attributions differed between BA and ADM.