Behavioral Activation for Depression: A Randomized Controlled Trial

Depression is a leading cause of disability worldwide (WHO, 2020). Nationally, more than 10.000 new cases of depression are registered each year with a higher frequency in women (Sundhedsstyrelsen, 2015). With depression comes reduced quality of life for the affected individual and a great economic burden for society. For instance, collectively, people with depression have more than 600.000 more visits to their general practitioner (GP) than people without depression, and when employed, they have 1.7 million more sick days than people without depression (Sundhedsstyrelsen, 2015). Access to effective treatment is thus crucial. Behavioral activation (BA) is an efficacious psychological treatment for depression; its effect may exceed that of medication and is equal to that of cognitive behavioral therapy (CBT) (Cuijpers, van Straten, \& Warmerdam, 2007; Ekers et al., 2014). BA is therefore recommended as a first-line psychological treatment for mild to moderate depression by the National Institute for Health and Care Excellence (NICE CG90; 2009).

Depression is often described with reference to the disorder's cognitive characteristics, such as the depressogenic content of thoughts ( Beck, Rush, Shaw, \& Emery, 1979) and the tendency to engage rumination and self-criticism (e.g., Mennin \& Fresco, 2013) as well as behavioral characteristics such as lack of motivation, avoidance, and withdrawal (e.g., Lewinsohn, 1974). Although less commonly found in the description of depression, research has also documented bodily characteristics, including alterations to the motor system. Concerning the motor system, gait patterns associated with sadness and depression have been characterized by reduced walking speed, arm swing, and vertical head movements (Michalak et al., 2009). Successful treatment of depression thus may include - not only improvements in cognition and behavior - but also normalization of the motor system.

Although CBT and BA produce similar results in the treatment of depression, they are rooted in different theoretical models of depression. Traditional CBTs target the depressed individual's dysfunctional appraisals. They do so by inviting the client to restructure their thinking and expose themselves to situations that will help them in this endeavor. This is based on the assumption that the root of the problematic behavior in depression, first and foremost, is a result of dysfunctional thinking (e.g., "it doesn't matter anyway", "I will always be depressed") ( Beck et al., 1979). BA, on the other hand, is primarily rooted in a behavioral theory of depression. It builds on Lewinsohn's (1974) seminal work, holding that depression is the result of a loss or lack of response-contingent positive reinforcement, leaving the patient stuck in a maladaptive pattern of behavioral avoidance and withdrawal, preventing them from having rewarding interactions with their surroundings. In order to combat this avoidance and withdrawal behavior permeating depression, the patient in BA is invited to monitor their activities - mapping the dysfunctional patterns - and then to engage in activities that are important and joyful, if not at first, then over time (Lejuez, Hopko, Acierno, Daughters, \& Pagoto, 2011; Martell, Dimidjian, \& Herman-Dunn, 2010). Indeed, in the beginning, certain activities may not feel instantly rewarding, and the depressed person may only have a vague memory of a time when that activity was associated with joy. However, in BA, patients are invited to try to act 'outside-in', instead of 'inside-out', meaning that they refrain from acting according to their mood (low mood will likely lead to more avoidance and withdrawal behavior) and instead approach the planned activity regardless of their current mood (i.e., outside-in).

Previous research has shown that patient compliance with the activity schedule is causally associated with depression reduction (Ryba et al., 2014). Thus, theoretically and empirically, following through with the activity schedule presents as one of the most important mediators of treatment gains. However, patients may struggle with initiating the planned activities. In a sense, BA asks patients to do exactly the thing they feel like they cannot do, that is, get up and get going. Despite psychoeducation about the rationale of the treatment, and even when clients get initial experience with the fact that they feel better and more competent after a certain activity, they may still struggle with action initiation (Martell et al., 2010). Recent evidence points to motor manipulations of posture and movement as having the potential to be an important supplementary intervention within BA, enhancing the likelihood that clients follow through with the planned activity schedule. In a recent meta-analysis, we investigated the effect of motor manipulations (i.e., expansive/upright and contractive/slumped postures) on affective and behavioral outcomes. Across studies, we found that the manipulation of motor activity altered mood, emotions, and actions such that when individuals adopt neutral or expansive postures, they reported better mood, more positive emotions, and engaged in more approach behavior than when assuming contractive postures (Elkjær et al., 2020). These results could suggest that adopting neutral or expansive postures could be helpful for depressed individuals in initiating action.

Building on the solid findings from previous BA trials and the recent findings concerning the effects of motor manipulations, the present study is a randomized controlled trial that will evaluate two versions of BA, one version which is a standard BA program (Lejuez et al., 2011) and one which is a motor enhanced BA program (mBA). In both programs, the patient will receive all standard BA interventions, whereas these - in the mBA program - will be supplemented with interventions focused on noticing and manipulating posture and movement associated with action initiation. One hundred and seventeen individuals diagnosed with major depressive disorder (MDD) will be randomized to active treatment or a 10-week waitlist followed by active treatment. The active treatment is also based upon randomization, where individuals will receive either standard BA or mBA, and will be followed for 6 months post treatment. In the active treatment, individuals will receive 10 weekly video sessions, conducted using the secure and GDPR-approved video-conference program 'Zoom'.

Before and after the mBA program, patients will partake in an experiment. The effect of motor manipulations on responses to two different personally relevant emotional conflicts will be evaluated. One emotional conflicts will be one that is typical of those detected in depression, where an individual wishes they could or would enjoy approaching a situation but lack the motivation to do so (i.e., approach vs. withdrawal). The other will be one in which an individual wishes they could approach a situation but feel scared or nervous about doing so (approach vs. avoidance). Participants will be randomized to one of three experimental conditions: adopting expansive postures, contractive postures, or neutral postures. Effects will be evaluated concerning action tendencies, behavioral self-efficacy and experienced emotions. For the experiment taking place before treatment, patients will be compared to a group of matched healthy controls.

Hypotheses

It is hypothesized that 1) both BA and mBA will be more efficacious in reducing symptoms of depression than the waitlist, and 2) gains are expected to be maintained through the follow-up period.

It is also expected that 3) mBA with its extra treatment component focused on the motor system will be superior to BA of at least small magnitude (d=0.3; Bell et al., 2013). This added benefit of mBA is expected to be driven by compliance with the activity schedule as it is hypothesized 4) that the mBA condition can increase success with planned activities (i.e., quantity of completed activities and proportion of completed activities). 5) This success with planned activities is expected to mediate treatment gains in both BA and mBA.

Concerning the experiment, before treatment, 6) it is expected that depressed patients to show less approach behavior (i.e., less approach action tendencies, less behavioral self-efficacy, , and less positive emotions) in response to the emotional conflicts compared with healthy controls. Furthermore, 7) it is hypothesized that participants assuming neutral and expansive postures will respond with more approach behavior than participants adopting contractive postures. 8) Effects will be explored between groups (healthy vs clinical), thereby determining if the neutral and expansive motor displays are more or less facilitative of approach behavior, and if the contractive motor displays are more or less detrimental to such behavior, depending on clinical status. 9) Effects will also be explored comparing the two types of emotional conflicts.

Concerning the experiment following treatment, 10) it is expected that any differences between the healthy and clinical group detected at baseline will diminish or disappear, reflecting clinical improvement and normalization following treatment.

Collaborators:

Mai B. Mikkelsen, Gitte Tramm, Emme Elkjær, Kaj Sparle Christensen, Douglas S. Mennin, \& Johannes Michalak