Jasp version 0

The Shapiro-Wilk test was used to assess the normal data distribution. Categorical variables were calculated using frequencies and proportions whereas continuous data has been estimated by means, standard deviations, and ranges.
One-way analysis of variance testing or the Kruskal-Wallis test was used to analyze differences among 3 groups. The χ² test was conducted for statistical analysis concerning categorical data when appropriate. Significant levels for multiple comparisons were adjusted using the Bonferroni procedure. Calculated P values were 2-tailed, a P-value <.05 was considered significant, and the range of confidence interval was 95% when appropriate. Statistical analysis was performed using JASP Team (2020). JASP (Version 0.14.1) (https://jasp-stats.org/).
An a priori calculation of sample size was done using g∗power 3.1.9.4 software (Heinrich-Heine-University, Dusseldorf, Germany). According to the analysis, at least 54 patients would be required (18 patients for each group), assuming a 2-tailed α value = 0.05 (sensitivity = 95%) and a β value = 0.95 (with a study power of 95%).

Changes between the first and second assessment in expected OLP effects were tested using the Wilcoxon-signed-rank-test due to non-normally distributed data. Considering sleep quality, bodily symptoms, mental well-being, and psychological distress, respectively, as outcome variables, mixed 2x5x2-ANOVAs were performed to assess the OLP-effect (within-factor “condition”) and the influence of time (within-factor “day”) as well as brand name (between-factor “brand name”). Since the order of the phases (placebo intake or control phase in week one) did not significantly influence the results, this factor was not included in the reported analyses. Holm-corrected post hoc-tests were applied were appropriate. Contrast analyses were calculated to test the hypothesis that OLP effects were larger with a brand name. As measures of effect size, η² (η² ≥ 0.01 small; η² ≥ 0.06 medium; η² ≥ 0.14 large) and Cohen’s d (d ≥ 0.30 small, d ≥ 0.50 medium, d ≥ 0.80 large) are specified. As explorative analyses, to identify potential predictors of the OLP effect, Pearson correlations between psychological factors and the outcome measures (i.e., the difference between the average score during placebo and control phase) were calculated (r ≥ |.10| small; r ≥ |.30| medium, r ≥ |.50| large). The alpha level was set to 5%. Analyses were calculated with JASP version 0.14.1 (JASP Team, 2020 ).

We used standard statistical methods for descriptive statistics. Categorical variables were presented as frequencies and continuous variables as mean (standard deviation, SD) or median (interquartile range, IQR), when appropriate. Normality was assessed through the Shapiro–Wilk test. Depending on the normality of the distribution, comparisons were made by Student's t-test or Mann–Whitney test for continuous variables, and by Pearson χ² test for categorical variables. The multivariate logistic regression was used to identify whether the FIB-4 score could be an independent predictor of poor 3-month outcome, and to establish the real prognostic value of demographic, clinical, and laboratory variables that reached statistical significance in the univariate analysis. To prevent biases, we did not include the variables already used for calculating the FIB-4 score in the logistic regression. An equivalent analysis was carried out for the secondary outcome, and is available in the Supplementary material. A two-tailed p-value of <0.05 was considered statistically significant for all tests. False discovery rate (FDR) correction was applied to deal with the multiple testing problem (results are expressed as adjusted p or _adjp-values). Analysis was performed using JASP Team (2020). JASP (version 0.14.1).

Statistical analysis was performed using JASP, Version 0.14.1 [43 ]. For descriptive analyses, mean values with standard deviation and percentages were built and presented as means (SD). The overall satisfaction with the course units was correlated with the different types of teaching formats (e.g., screencasts, lecture recordings, etc.) using point biserial correlation analysis with Pearson’s r [44 ]. A p-value of 0.05 was considered statistically significant. Free-text comments were evaluated and categorized by two different team members. Anchor examples were chosen separately and taken for each section after discussing them. Since the course structure differed between summer and winter semesters as described above, and as there was less time available to complete the “DigiPath” course in the winter semester, the results for the asynchronous course sections in the summer and winter semesters are presented separately. Since the student cohorts were predefined by the curriculum and could not be changed, a sample size calculation was not performed for this observational study.

Statistical analysis of clinical and psychological variables was conducted in R and JASP (JASP Team (2020). JASP (Version 0.14.1) (Computer software). ANCOVAs, t-tests and chi-square tests were used to compare Mclust subgroups on clinical and demographic variables (i.e., age). Results were corrected using Holm method. For the treatment trajectory’s analysis, we used repeated measures ANOVA with Mclust as a fixed factor and assessment time (t0, t1, t3, t6, t12) as the repeated measure to examine effects of time. Groups were compared at each time point using ANCOVAs. Sex and age were used as covariates in all of the analyses.