Article Text
PDF
Statistics from Altmetric.com
Question
Question: What are the predictors and moderators of response to Internet-delivered interpersonal psychotherapy (IPT) and cognitive behavioural therapy (CBT) for depression?
Patients: In total, 1843 adults aged 18 or older who spontaneously visited an automated Internet-delivered self-help program (e-couch) and who were not receiving treatment for depression from a mental health specialist.
Setting: Internet-based program, October 2009 to October 2010.
Intervention: A 4-week program on Internet-delivered CBT (n=610), Internet-delivered IPT (n=620) or active control (MoodGYM; n=613). The CBT intervention explained the rationale of CBT and covered identifying and tackling negative thoughts and undertaking behavioural activation. The IPT intervention consisted of four modules: grief, role disputes, role transition and interpersonal deficits. MoodGYM covered the identification of and behavioural methods to overcome dysfunctional thinking, assertiveness and self-esteem training.
Outcomes: Self-reported depression symptoms, assessed using the Center for Epidemiological Studies Depression scale (CES-D)
Patient follow-up: Thirty per cent of participants completed immediately post-test and 28% completed the 6-month follow-up. Drop-out rates were higher with MoodGYM (n=451; 35%) than with CBT (n=429; 33%) and IPT (n=414; 32%).
Methods
Design: Randomised controlled trial.
Allocation: Concealed.
Blinding: Not reported.
Follow-up period: Six months.
Main results
A significant reduction in depressive symptoms at postintervention and follow-up was seen for people assigned to either CBT or IPT who completed both follow-up assessments (completers). Both interventions were non-inferior to MoodGYM control based on both completer and intention-to-treat (ITT) analyses. A mixed-model repeated measures analysis was performed to examine predictors and moderators of ITT response. Sociodemographic factors (age, gender, marital status, education level), illness severity (depression symptoms, previous history of depression, disability, anxiety symptoms, quality of life, medication use), skills (mastery or perceived control, dysfunctional attitudes) and treatment preference (if expressed) were considered potential predictors and moderators. Females and participants with lower dysfunctional attitude scores had greater reductions in depression symptoms postintervention. Participants with lower mastery scores had greater reductions in depression symptoms at both postintervention and follow-up. At follow-up, older participants (age over 25 years) in the MoodGYM or CBT conditions had larger improvements in depression scores than those in the IPT condition, whereas younger participants (age 24 or less) in the IPT intervention had larger improvements than those in the MoodGYM or CBT conditions. Younger participants also had significantly greater postintervention improvements with IPT than with MoodGYM.
Conclusions
The effects of Internet-delivered mental health programs for people with depression symptoms may be modified by sociodemographic factors and clinical characteristics. Younger people may receive greater benefit from Internet-delivered IPT, whereas those over 25 may receive greater benefit from CBT.
Abstracted from
Commentary
Internet-delivered psychological treatments are increasingly supported in research and have been implemented in some parts of the world. Most studies have focused on cognitive behavioural therapy (CBT), although exceptions exist, for example, the original publication for the present study in which unguided, self-help interpersonal psychotherapy (IPT) was tested against two forms of CBT.1 This study recruited participants from the general population and found medium effect sizes for all treatments included. However, there was a substantial drop-out rate of 70% (1294/1843). While therapist-guided Internet treatments can be as effective as face-to-face therapy,2 automated, non-guided treatments can be helpful and effective, even if the effects tend to be smaller than in guided interventions. Research is needed to identify who will benefit from unguided treatments.
Few studies on Internet interventions have investigated predictors and moderators of treatment outcome. Hence, this study is an important contribution. The literature to date on predictors across different therapies and moderators specific to therapy form is scattered and inconsistent for face-to-face and Internet interventions. While this study had the potential to add much knowledge and used appropriate methods to handle data loss in the trial, the authors acknowledge a lack of statistical power, arising from small subgroups, as a limitation.
“Given the high drop-out rate, an interesting approach to the data would have been to focus on who adhered to the interventions as well as who benefitted from therapy at study completion. If gender does influence outcome and age is differently related to outcome for different therapy orientations, this could also be useful information.” It is important to note that moderators imply that the treatment works differently and probably via different mechanisms for different groups (in this case, age groups). An important challenge for future clinical and basic research is to better understand the mechanisms involved in generating and maintaining symptoms of depression.
While this work is important and informative, given the paucity of replicated findings in this area, the results should be interpreted with caution. The absence of more theory-driven tests of predictors and moderators perhaps reflects poor theory development within the field of Internet interventions and in analysis of what makes Internet-delivered therapy work.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
Footnotes
-
Sources of funding: Centre for Mental Health Research at The Australian National University, Canberra, Australia; the Faculty of Psychology and Education of the VU University, Amsterdam, the Netherlands; and Black Dog Institute, University of New South Wales, Sydney, Australia.
Footnotes
-
Competing interests None.