Ires Ghielen

Table 2. Effect sizes and heterogeneity measures for CBTs and MBTs in improving psychological distress in PD and MS patients, including subgroup analyses. N (studies) Hedges’g 95% CI I² 95% CI p-value NNT* Meta-analysis 1 All 12 0.51 0.22 – 0.80 66 27-80 <0.001 3.55 Excluding outliers @ 8 0.31 0.13 – 0.48 0 0-56 0.008 5.75 Outcome Depression 10 0.33 0.05 – 0.62 60 0-78 0.042 5.43 Anxiety 8 0.36 0.03 – 0.68 59 0-79 0.038 5.00 Psychological distress 5 0.79 0.32 – 1.25 66 0-85 0.015 2.36 Subgroup analyses Disease type MS 8 0.54 0.26 – 0.82 45 0-72 0.003 3.36 PD 4 0.37 -0.55 – 1.29 80 16-91 4.85 Control condition Waitlist 7 0.39 0.18 – 0.60 26 0-68 0.006 4.59 TAU 5 0.67 -0.16 – 1.49 82 49-91 2.75 Risk of Bias # High 8 0.57 0.14 – 0.99 71 22-84 0.003 3.18 Low 4 0.42 0.02 – 0.81 61 0-85 4.27 Meta-analysis 2 All 7 0.36 0.13 – 0.58 40 0-74 0.002 5.00 Treatment type CBTs 4 0.45 0.26 – 0.64 0 0-73 0.004 3.55 MBTs 3 0.06 -0.56 – 0.68 68 0-89 0.72 29.41 MS = Multiple Sclerosis; PD = Parkinson’s Disease; CBTs = Cognitive Behavioral Therapies; MBTs = Mindfulness Based Therapies; NNT = Number Needed to Treat; TAU = treatment as usual. *according to Kraemer & Kupfer [50]. @ outliers include Okai et al. [41], Kiropoulos et al. [35], Ghielen et al. [40]. # low risk of bias include studies scoring 1 according to the risk of bias assessment tool, developed by Cochrane [24], a score > 1 is considered high risk of bias. 6