The author presents a one-arm survey-based study about a mindfulness-based mobile app (7Mind). Users who participated in a mindfulness-based stress management course answered questionnaires about their mental state before starting and after completing the course. Mindfulness was measured by using the Freiburg Mindfulness Inventory. In total, 7117 subjects submitted questionnaires and of those, 829 participants filled out both questionnaires. The author used Wilcoxon-signed rank tests for evaluating pre-/post differences and found statistically significant differences in Mindfulness and perceived pain among other variables.

Importance and Originality:

Stress reduction is important in today’s world and mental health disorders are more and more common. Therefore, digital apps to reduce mental stress and to support healthy daily behaviors seem to be an efficient alternative to psychotherapy or weekly meditation/mindfulness courses with an instructor. However, there is little evidence about the impact on app-based online courses on stress management. Therefore, I consider this study as important.

Major Comments:

• The validity of the study is limited since no control group was enrolled and compared to. However, the author mentions that in the discussion. Still, the author could be a bit more cautious with conclusions coming from this study.

• The author does not distinguish between statistical significance and content relevance. I would suggest not focusing on p-values and deriving conclusions from them, instead I would suggest comparing the distributions depicted in figure 4a and 4b. Since the sample was quite big, even tiny effects would have been statistically significant, without having relevance content wise. Hence, in my opinion a Bonferroni-correction is not necessary.

Minor Comments:

• I see some “Errors” which should possibly be citations (e.g., mental disorders (Error), meditation applications (Error).

• “This study investigates whether digital mindfulness-based stress management training offered to the broad society as a health insurance-covered prevention program can affect it positively.” I would suggest the author states specifically what “affect positively” actually means in his study.

• “The study aimed to find out whether an App-based intervention can change the Freiburg Mindfulness Inventory (FMI) score significantly within a general population.” Please shortly state what the FMI measures.

• CRM system: Please give the whole name instead of the abbreviation.

• I am not sure if Figure 1 is informative, actually correct and therefore necessary. In addition, the caption is non-informative.

• “Participants with only a single time point (only one survey taken or inability to link the forms) were excluded from the confirmatory analysis.” I would not say that you analysis is confirmatory but exploratory.

• Furthermore, it would be interesting to evaluate the subgroup of participants, which only responded one of the questionnaires in terms of their outcomes.

• Instead of “Participants with only a single time point “ I would suggest writing “Participants which only answered one questionnaire…

• “The Null-Hypothesis was "There is no significant change between t0 and t1". Please consider reformulating your null hypothesis; the Wilcoxon signed-rank test tests whether there the two distributions of t0 and t1 are equal.

• “This provided the confidence intervals and p-values.” Which confidence intervals were provided/ confidence intervals for what?

• “Alpha was set to 0.05 for all tests using data once.” I would suggest to delete “… for all tests using data once.”

• Link missing: “(insert a link to the file)”

• “The 4-digit code for identification was based on only three questions, so it was possible that the same code was created more than two times (which led to the inability to match the data of this person in the datasets)”: How often did that happen?

• Link missing “(add address here)”

• Table 1: I would suggest using different terms for the columns instead of “T0” and “Merged t0t1”. Interested readers who scroll through the article should directly understand the content of the table without first reading the whole manuscript.

• Could the author please reverse the order of the columns of table 2? I would suggest the following order: N, Median pre, Median post, Median difference, 95% CI of the Median difference (in one column), if you wish a p-value

• I would suggest rounding the numbers to two decimal points only.

• “The 95 percent confidence interval lies between 7.500078 and 6.500035.” Please round the numbers and reverse the order of the confidence bounds.

• Please amend the caption of Table 2 (e.g., Overview of Median changes in FMI Scores, including sample sizes N, …,) as well as the captions of Figure 4a and b: Violin Plots of FMI scores at t0 / t1

• Where can I find the appendix, to find out about the secondary hypotheses?

• “All statistical tests turned out to be significant, and all differences pointed in the intended direction of improvement. This confirms that mindfulness-based stress management programs and meditation can also be taught by Apps.” I think that the wording “confirms” is too strong without a randomized experiment, please use something like “suggests/hints/indicates”.

• “The improvements of this prevention course need to be reevaluated after six and twelve months to examine their long-term effects.” Is this also part of your research?

• “However, the absence of a placebo group combined with the open-label design does not control the Hawthorne effect.” Please explain shortly what the Hawthorne effect is.

• It is a bit confusing to read about pain reduction in the discussion, when there is no paragraph about it in the methods and results section of the main document. I would suggest adding a methods and results section about the pain reduction in the main document. As an alternative, the author could put everything about pain reduction into the appendix.

• “The app intervention has shown a significant effect on the FMI score that pointed in the intended direction.” I recommend that the author does not use causal language here due to the mentioned limitations (dropouts, selective audience and the missing control group).