Predicting outcome of Internet-based treatment for depressive symptoms

Abstract

In this study we explored predictors and moderators of response to Internet-based cognitive behavioral therapy (CBT) and Internet-based problem-solving therapy (PST) for depressive symptoms. The sample consisted of 263 participants with moderate to severe depressive symptoms. Of those, 88 were randomized to CBT, 88 to PST and 87 to a waiting list control condition. Outcomes were improvement and clinically significant change in depressive symptoms after 8 weeks. Higher baseline depression and higher education predicted improvement, while higher education, less avoidance behavior and decreased rational problem-solving skills predicted clinically significant change across all groups. No variables were found that differentially predicted outcome between Internet-based CBT and Internet-based PST. More research is needed with sufficient power to investigate predictors and moderators of response to reveal for whom Internet-based therapy is best suited.

For whom are internet-based occupational mental health interventions effective? Moderators of internet-based problem-solving training outcome

“…The finding that higher pre-treatment depression was associated with a higher likelihood of improvement across both groups seems plausible and is consistent with other research (Ruwaard et al, 2009;Spek et al, 2008;Warmerdam et al, 2013). Greater severity leaves more room for improvement and thus significant change in reduction of depressive symptoms.…”

“…Consequently, these ranges are based on the mean scores of the current study and may not generalize to other studies. Lastly, definitely not all possible predicting/moderating variables were captured in the current study; for example, number of previous episodes, education level, and dysfunctional attitudes, or other work-related factors, such as low-effort reward imbalance and lack of control over decision at work, could have also predicted outcome or moderated the effects of IPST (Bockting et al, 2006;Donker et al, 2013;Spek et al, 2008;Warmerdam et al, 2013). Future studies should aim to consider these variables as well as long-term effects with respect to depression onset (van Zoonen et al, 2014).…”

“…To date, few studies have explored the participant characteristics that determine the effectiveness of Internet-based interventions for depressed individuals (Donker et al, 2013;Ebert et al, 2013;Warmerdam et al, 2013).…”

“…Hence, the results of exploratory studies would be able to function as general guides to make ideal decisions for future randomized clinical trials (RCT). The selection of predictors and moderators was based upon (a) evidence of former studies exploring predictors/moderators of outcome in face-to-face CT (Hamilton and Dobson, 2002), (b) evidence of former studies exploring predictors/moderators of outcome in Internet-based intervention studies (Button et al, 2012;de Graaf et al, 2010;Donker et al, 2013;Spek et al, 2008;Warmerdam et al, 2013), and (c) theoretical assumptions based on working-related or intervention characteristics. Thus, the final list of potential predictors and moderators explored in the current study included: Demographics: age, sex, weekly working hours, experience with therapy/training; Clinical Pre-treatment Characteristics: depression severity, emotional exhaustion; and Motivational Variables: general self-efficacy.…”

Therapeutic alliance and different treatment formats when delivering internet-based CBT for depression

“…This might have affected the results since CBT has been found to be more effective for university students (Cuijpers, Ebert, Acarturk, Andersson, & Cristea, 2016b) and specific studies on internet-based treatment have shown that individuals with a university degree receive better outcomes (Warmerdam, van Straten, Twisk, & Cuijpers, 2013). It could be a problem for dissemination into regular health care, since ICBT might only fit a small well-educated group of patients.…”