Ires Ghielen

131 BEWARE: Study protocol of the pilot RCT 8. Freezing of Gait (FOG) questionnaire This 6-item questionnaire is used to identify current problems with walking and symptoms of freezing [44]. For each item patients indicate which of the 5 statements is most applicable. 9. Visual Analogue Scales (VAS) Before and after each therapy session patients indicate how they are feeling at the actual moment, using 10 VAS. Patients put a cross on a line of 10 cm with extreme feelings at the edges, for example ‘relaxed’ or ‘tense’, ‘in control’ or ‘out of control’, and ‘happy’ or ‘sad’. Patient researchers Since the experimental treatment is a new approach, the opinion and experiences of the participating patients are considered very important. Therefore, two patient researchers of the Dutch Parkinson patient association (Parkinson Vereniging) contribute to the study by anonymously documenting the patients’ expectations and evaluation points for qualitative analysis. Prior to, and directly after the intervention period, a semi-structured group interview takes place in which the patients are asked questions related to the study and can share their experiences and suggestions for adaptations in future studies or in the implementation phase. This interview is with the patients (both participants and researchers) only. Statistical analysis Calculation of sample size We expect that the experimental group will show a 10% larger improvement on the GSES after treatment, compared to the TAU group. Based on data on the GSES in other chronic diseases [45], we expect to detect a 10% larger reduction in GSES score from 32 to 28.8 with an overall sd of 3.3. A minimum of 16 patients is required per arm of the trial. With that, including a dropout of 10%, we estimate that 36 PD patients (18 per arm) are needed to achieve a sufficient statistical power of 80% with two-tailed significance level set at p<0.05. Data analysis To promote data quality, double data entry will be used. Multiple imputation will be applied on missing data. ANOVA with repeated measures will be used for normally distributed outcomes at interval and ratio level. Friedman’s non-parametric repeated measures test will be applied for outcomes with non-interval or ratio scale level, or for non-normally distributed variables. Factors in the analysis will be 7