Review Reports - Better Mental Health and Sleep under Behavioral Restrictions Due to COVID-19 in Japanese University Students: A Cross-Sectional Survey

Round 1

Reviewer 1 Report

Dear Editor,

I would like to thank the authors of the manuscript ID No clockssleep-2399496. entitled “Better Mental Health and Sleep under Behavioral Restrictions 2 due to COVID-19 in Japanese University Students: A Cross- Sectional Survey” for presenting the results of their study on associations between behavioral restriction by COVID- 19 and sleep pattern and mental health in first-year Japanese university students.

The manuscript presents the results of a cross section study based on data from four hundred and twenty-two students, who were tested with the questionnaire that comprised socio-demographic characteristics, the Pittsburgh Sleep Quality Index, Japanese version of the Epworth Sleepiness Scale, and the Patient Health Questionnaire-9. In order to establish the presence of associations between behavioral restriction by COVID- 19 and sleep pattern and mental health in first-year Japanese university students, authors interviewed 193 students that were taking their classes online due to COVID 19 restrictions in year 2020. and compared the results with data from students acquired in pre COVID 19 period (2018-2019). Authors revealed that self-restraint due to COVID-19 was associated with better sleep and mental health. Authors used basic statistical analyses as well as multiple linear regression models to estimate the association between the PHQ-9, PSQI scores and behavioral restriction by COVID- 19.

After reading this article in details, my main impression is that the article is well written, in adherence to Journal’s standards, and addresses a very important issue of sleep and mental health issues in first year Japanese students, as well as possible relations between behavioral restrictions imposed during COVID 19 epidemics and sleep quality and quantity. This study showed that they had longer sleep duration, improved sleep quality, less daytime sleepiness, and improved mental health under home confinement compared to a regular lifestyle. Sleep disorders and consequential mental health issues are very prevalent in the adult working population especially in the current concept of 24/7 modern society and if left unrecognized or untreated can lead to serious consequences on students’ health and quality of sleep and life.

There are several issues for consideration:

1. Title: Adequate. No remarks

2. Abstract: Adequate. No remarks

3. Introduction: Adequate. Remarks:

a. Authors should focus more on the quality of sleep in students in Japan. There are several papers that could be of use in order to explain the specificities of sleep and study characteristics in Japan. Authors mentioned only that people in Japan have shorter sleep time than in other countries.

4. Results: Adequate. Remarks

a. 2.1. Characteristics of subjects and the results of questionnaire survey: One passage is redundant, since in the first passage authors gave the explanation for the whole sample, and in the second, practically same data, divided by sex. Authors should consider joining two tables (1 and 2) in one.

b. Due to specific format of the manuscript, methods come after Discussion, so the full names of questionnaires should be introduced before using PHQ-9, JESS, and PSQI.

c. There are two tables under Table 4 heading – first one is copy of Table 6. Please correct

5. Discussion: Adequate. Remarks:

a. Authors should include some comparison with the quality of sleep in students specifically in Japan before and during the pandemics, since there are several manuscripts available.

b. Authors have not taken in consideration (or have not presented) the effects of work or dietary habits on quality of sleep and mental health. In Results, authors mentioned only the difference in commuting, exercise, living situation, but not presence or absence of differences in academic demands, part-time jobs, free-time activities. For example, there is a strong possibility that lack of options to have a part time job during pandemics influenced the material safety and mental health of students. On the other hand, lack of need for secondary financial income through part time job could be the reason for better sleep quality and mental health. Similar conclusion goes for dietary habits. If the authors did not perform that analysis, please address it in Limitations

6. Materials and Methods: Not Adequate. Remarks:

a. Under the behavioral restrictions due to COVID-19…: Authors did not provide the definition for this term or for the term self-restraint due to COVID-19. What kind of restrictions were self or local government imposed and in what period?

b. How was the survey delivered to students? Through University network, Google, social networks, mobile app? What was the response rate? How were the questionnaires distributed to students before COVID-19?

c. Demographic and lifestyle variables…. The question items included name, age, sex, undergraduate course, extracurricular exercises, and part-time jobs… Authors have not taken in consideration (or have not presented) the effects of extracurricular exercises, and part-time jobs on quality of sleep and mental health. What was the reason for excluding this data from the results?

d. PSQI – What version of PSQI was used, English or Japanese? What cut off score was used?

e. JESS- Epworth sleepiness scale is primarily used as a measure for excessive daytime sleepiness irrelevant of the cause. See https://epworthsleepinessscale.com/about-the-ess/

f. Although used as an additional tool, it cannot be used as a primary screening tool for obstructive sleep apnea. Berlin questionnaire or STOP BANG were created specifically for OSA. What cut off score was used for JESS?

g. What cut off score was used for PHQ-9?

h. Statistical analysis…. Multiple linear regression models were used to estimate the association between the PHQ-9 scores and WHAT? before and during COVID-19.

i. Similarly, this analysis model was used to estimate the association between the PHQ-9 (YOU MEAN PSQI?) score and WHAT? before and during COVID-19…

j. Four models were used to assess these associations: model 1 was crude, model 2 was adjusted for age and BMI, model 3 was adjusted for model 2 covariates plus exercise habits and living situation, and model 4 was adjusted for model 3 covariates plus PHQ-9

score…This is not in concordance with data presented in Results and Tables 3-5 *Crude. **Adjusted for gender, age and BMI. ***Adjusted for gender, age, BMI, exercise habits and living situation. ****Adjusted for gender, age, BMI, exercise habits, living situation and PHQ9 score.

Similar situation for models for PSQI and data presented in Results and Tables 6-8.

7. References: Adequate, relevant. Missing references for sleep quality and mental health for students in Japan

8. Tables: Adequate. Corrections needed. There are two tables under Table 4 heading – first one is copy of Table 6. Please correct

none

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Introduction:

Page 1: Line 37-42: The tense should be past tense. Needs grammatical check.

Page 2: Line 46. Results were lower in PHQ-9,JESS and PSQI. Does it mean that sleep quality was lower during Covid-19.

Methods: The surveys used were Pittsburgh Sleep Quality Index, Epworth Sleepiness Scale and Patient Health Questionnaire. Can the authors briefly describe the surveys. What score indicates good sleeping v/s disturbed sleep.

Results:

Table 2: What does "above cutoff" of PHQ9 score/JESS score/PSQI-J mean? Does it mean the % of the surveyed population that reported poor sleep? Language can be made clearer.

Table 3-5: If I understand correctly, the outcome is PSQI score?

Model 1: Empty variable.

Model 2: Gender, Age and BMI

Model 3: Gender, Age, BMI, exercise habits and living situation

Model 4: Gender, Age, BMI, exercise habits, living situation and PHQ-9 scores.

Why does Table 4 have two tables under? Can the authors clearly label each table and say what it stands for?

Which is the best model out of all of them? Use R², Root Mean Squared Error (RMSE), AIC (Akaike Information criteria) and BIC (Bayesian Information criteria) to figure it out and provide the coefficients for every predictor variable for the best model.

Table 6-8: Similar concerns. Which is the best model for each Table? Use the accuracy methods mentioned above and identify the best model and provide the coefficients for every predictor variable in that model.

You can

Also for Table 3-8: Instead of separate regression models for men and women, just keep sex as a predictor in the model for the whole population.

The introduction could be made better. Any time you are describing early phases of the pandemic, past tense should be used.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Editor,

I would like to thank the authors of the revised manuscript ID No clockssleep-2399496. entitled “Better Mental Health and Sleep under Behavioral Restrictions 2 due to COVID-19 in Japanese University Students: A Cross- Sectional Survey” for presenting the results of their study on associations between behavioral restriction by COVID- 19 and sleep pattern and mental health in first-year Japanese university students.

There are several issues for consideration:

1. Title: Adequate. No remarks

2. Abstract: Adequate. No remarks

3. Introduction: Adequate. Remarks addressed

4. Results: Adequate. Remarks addressed

5. Discussion: Adequate. Remarks addressed

6. Materials and Methods: Adequate. Remarks addressed

7. References: Adequate. Remarks addressed

8. Tables: Adequate. Adequate. Remarks addressed

Author Response

Dear Editor: We thank the reviewers for the time and effort that they have invested in reviewing our manuscript. We appreciate their insightful comments, which have helped us significantly improve our paper. Please note that our changes to the manuscript. We believe we have responded perfectly to the reviewers' comments and suggestions.

Response to Reviewer 1 Comments.

Thank you for your kind and accurate peer review. Your comments and suggestions have certainly improved our paper. Thank you very greatly.

Reviewer 2 Report

Thanks for incorporating my suggestions. I have one question for you. Which was the best model based on your analysis? Based on the BIC values that you provided in your cover letter, looks like Model 4 (the most complex model) has the least error. Can you provide a table with the B values for self-restraint (for the best model) and other covariates?

Author Response

Response to Reviewer 2 Comments.

Comments and Suggestions for Authors

Response: Thank you for your accurate and meaningful peer review. As you indicated, we have added partial regression coefficient (B value) and SE in the table.

We believe we have answered your point perfectly.