Ires Ghielen

114 Chapter 6 This leads us into the discussion concerning the suitability of questionnaires to measure treatment effects. Since MBTs are focused on awareness and acceptance, and not aim to reduce symptoms, questionnaires that measure the prevalence or severity of symptoms are less appropriate. The studies by Bogosian et al. [32] and Simpson et al. [38], however, investigated mindfulness interventions and showed effect sizes of g = 0.55 and g = 1.04, respectively, in improving general mental health. In addition, when overall psychological distress was measured with general mental health questionnaires, a high effect size of g = 0.79, although with considerable heterogeneity, was found. The focus of an intervention, type of control condition, and the outcome measures used seem to be of importance in evaluating the effectiveness, and therefore need to be carefully considered when conducting an RCT. Overall, the included studies had low quality, only three out of seventeen studies reached good quality according to the risk of bias tool. The findings need to be carefully interpreted since risk of bias is present in most of the studies and might have influenced the treatment effects. Each study suffers from different types of bias, except for the performance bias which is always a risk due to the nature of these intervention studies. Limitations and implications First, the effect size is solely based on studies in patients with PD and MS, since there were no RCTs found in HD that investigated the effect of psychological interventions on psychological distress. Second, MS patients might be overrepresented in the meta-analysis since fifteen out of nineteen RCTs investigated MS populations, resulting in the effect size being driven mostly by MS populations, especially in the second meta-analysis in which only MS populations were included. Heterogeneity estimates were above 40% in most analyses, reflecting high heterogeneity within the meta-analyses, and most studies included small sample sizes, which resulted in low power. Finally, the overall quality of the studies was low and the quality of one study could not be assessed. It is therefore recommended to study psychological interventions in more detail and in larger patient samples in study designs with higher methodological quality. Especially in HD more research is needed, since no RCTs on the effects of psychological treatment were found in our literature search. It might also be interesting to investigate the addition of psychopharmacological therapies, besides