- Type:#article
- Year read:#read2019
- Bibtex: @andersson2015
- Bibliography: Andersson, E. et al. Testing the Mediating Effects of Obsessive Beliefs in Internet-Based Cognitive Behaviour Therapy for Obsessive-Compulsive Disorder: Results from a Randomized Controlled Trial. Clin. Psychol. Psychother. 22, 722–732 (2015).
Why and when I was reading this
I was doing a deep dive into the role of cognitions in OCD. This is Erik Andersson paper with some critique of Paul Salkovskis and Stanley Rachman CBT model.
Key takeaways
- They hypothesized that ICBT group would have greater reductions in obsessive beliefs after a cognitive intervention, and that it would predict greater symptom reduction.
- ICBT group actually had more obsessive beliefs! Increase, not decrease, of obsessive beliefs predicted better outcomes
- The authors discuss how this may result from better insight after the cognitive intervention, leading to higher obsessive beliefs and motivation to engage in ERP.
Using a cross-sectional mediation analysis, they found results in line with the original hypotheses: reduced obsessive beliefs from ICBT and association between lower obsessive beliefs and better treatment response.
Belief domains relevant to OCD
- Inflated sense of responsibility
- Overestimation of threat
- Perfectionism
- Intolerance of uncertainty
- Over-importance of thoughts
- Need to control thoughts
An earlier trial (Woody, Whittal, and McLean 2011) found that symptom change preceded and mediated change in obsessive beliefs, contrary to the cognitive hypothesis!
Page 1: Highlight annotation by Oskar Flygare on January 5th 2018, 1:59:51 pm:
Contrary to our expectations, the longitudinal mediation analysis indicated that (1) being randomized to ICBT actually increased the degree of obsessive beliefs after receiving the cognitive intervention at weeks 1– 3, and (2) increase in obsessive beliefs predicted better outcome later in treatment.
Page 1: Highlight annotation by Oskar Flygare on January 5th 2018, 2:00:09 pm:
We conclude that, although obsessive beliefs were signi fi cantly reduced at post-treatment for the ICBT group, early increase rather than decrease in obsessive beliefs predicted favourable outcome.
Page 6: Highlight annotation by Oskar Flygare on January 5th 2018, 2:12:43 pm:
The direct effect ( c-path) came out signifi cant (B= 1.44, t = 6.53, p < .001) meaning that randomiza- tion to ICBT predicted greater symptom reduction based on the weekly OCI-R measurements
Page 6: Highlight annotation by Oskar Flygare on January 5th 2018, 2:13:02 pm:
contrary to our expectations, results showed that being randomized to ICBT was associated with an increase on the OBQ-44 after module 3 (a-path; B= 6.98, t = 2.66, p < .01), while, in turn, increase on the OBQ-44 after module 3 was associated with subsequent decrease on OCI-R, i.e. better treatment outcome (b-path; B= .0098, t = 1.3205, p =.1902)
Page 6: Highlight annotation by Oskar Flygare on January 5th 2018, 2:13:22 pm:
early increase on OBQ-44 signifi cantly mediated a subsequent decrease in OCI-R.
Page 8: Highlight annotation by Oskar Flygare on January 5th 2018, 2:13:56 pm:
he OBQ-44 followed a similar path as in the main analysis, i.e. OBQ-44 increased slightly after completion of module 3 and then dropped substantially at post-treatment. Furthermore, this initial increase on the OBQ-44 correlated with better treatment outcome (r = .306, p < .05). Thus, although not experimen- tally controlled, the results from this replication were similar to the main analysis.
Page 8: Highlight annotation by Oskar Flygare on January 5th 2018, 2:17:52 pm:
Consequently, it still remains unclear whether the cognitive intervention in this study had a direct or indirect effect on OCD symptoms. By direct effect, we mean that the cognitive intervention itself lowered OCD symptoms independent of ERP frequency. By indirect effect, we mean that the cognitive interven- tion did not itself lower OCD symptoms, but instead made patients more motivated to do ERP, which, in turn, lowered OCD symptoms. Unfortunately, we have no systematic data on ERP adherence in this trial, and it is therefore not possible to explore this hypothesis further in the present study.
Page 8: Highlight annotation by Oskar Flygare on January 5th 2018, 2:14:46 pm:
Again, themediationmodelcameoutsignifi cant (indirect effect: est = .3747, 95% bias-corrected CI = [ .7488 – .1578]), but with reversed results compared to the main analysis, i.e. beingrandomizedtoICBTpredictedareductiononthe OBQ-44 (a-path; B= 2.86, t = 3.64, p < .001), and reduced score on the OBQ-44 predicted better treatment outcome (b-path; B=.13, t =6.00, p < .001). Means are displayed in Table 3.
Page 8: Highlight annotation by Oskar Flygare on January 5th 2018, 2:16:11 pm:
mixed results on the mediating role of obses- sive beliefs in therapy.
Page 10: Highlight annotation by Oskar Flygare on January 5th 2018, 2:27:26 pm:
In fact, in as much that obsessive beliefs play a role in the treatment of OCD, the present study suggests that the early mediating role is contrary to the expected from a cognitive view, i.e. that increase in these beliefs is related to better outcomes.
Erik Andersson comments on the cognitive paradigm
There are too many constructs and not all are meta cognititions. We should engage patients in a discussion of meta cognitions but not the others, because that may lead to reassurance.
- Inflated sense of responsibility
- Overestimation of threat
- Perfectionism
- Intolerance of uncertainty (M)
- Over-importance of thoughts (M)
- Need to control thoughts (M)