140 Chapter 6 When we look at the two effectiveness studies more closely, some interesting observations can be made that may guide future research. With the inclusion criteria for the quasi-randomized study (a BMI ≥ 30 and an above average urge to eat in response to negative emotions), we expected to create a BED subgroup that would be more likely to profit from an emotion regulation focused intervention. Ultimately however, both the quasi-randomized and the non-randomized sample showed an average BMI ≥ 30 and high emotional eating. The only significant differences between the two samples were for BMI and BDI-II scores, such that both were slightly higher in the non-randomized study (BMI: non-randomized study M = 42.3, SD = 7.6; randomized study: M = 39.9, SD = 5.6; 95% CI = -4.10 to -.68; d = .251; BDI-II: non-randomized study M = 23.56, SD = 11.13; randomized study: M = 20.77, SD = 9.26; 95% CI = -5.49 - -.10; d = .344). Interestingly, differences in outcome between DBT-BED and CBT are smaller in the sample with more depressive symptoms and a somewhat higher BMI. This touches upon the idea that those with a relatively higher BMI and/or more depressive symptoms might be better off in DBT-BED. However, depressive symptoms did not moderate outcome (Chapter 5). And, although we did not test BMI as a moderator, it is unlikely that BMI would moderate treatment outcome because both groups have average BMI’s in the same high range. A notable difference between the two studies was the way assignments to the two treatments were made. In the randomized study, random chance decided, whereas in the non-randomized study (with less differences in outcome between CBT+ and DBT-BED) multiple factors were involved in group assignment, including the patients’ preference. This raises the question whether those in DBT-BED may have been more motivated/performed better because they actively chose to engage in this new treatment. With our data, we cannot answer this question. Yet, it does touch upon the possible relation between the patient receiving the preferred treatment, and treatment outcome. Intuitively, it makes sense that an active choice for a preferred treatment, or otherwise receiving the treatment of choice, could contribute positively to engagement and therefore possible better outcome. A recent meta-analysis (Windle et al., 2020) showed that receiving a preferred psychosocial mental health treatment did have positive associations with dropout rates and therapeutic alliance. However, there was no association with attendance, clinical and global outcomes, treatment satisfaction and remission. It is unclear whether these results, found mainly in people with anxiety, depression and alcohol or substance use disorders, also apply to individuals with eating disorders or with BED specifically. If so, matching treatment to patients’ preference may be a way to improve dropout rates and therapeutic alliance, but may not lead to better clinical outcomes. To date, research on the relation between the patients’ preference (e.g. shared decision making) and outcome in eating disorders is very limited (Jansingh et al., 2020).
RkJQdWJsaXNoZXIy MjY0ODMw