Non-Pharmacological Interventions to Improve Chronic Disease Risk Factors and Sleep in Shift Workers: A Systematic Review and Meta-Analysis

Round 1

Reviewer 1 Report

The article entitled “Non-pharmacological interventions to improve chronic disease risk factors and sleep in shift workers: A systematic review and meta-analysis” is extensively indexed and analyzed using suitable methods and resources. This analysis paper tends to be quite well written and very detailed in the methodology and literature surveyed. Overall, observations and recommendations are very important and should be taken into account for future studies. As detailed in the report, many objective sleep effects, subjective sleep duration, sleep quality, and some chronic disease risk factors improved with adaptive shift frequency and schedule adjustments, extended recovery periods, personalized behavioral strategies, and other interventions. Although some limitations exist with the literature used, which limited the scope of the recommendations relevant to the outcome of the research. What I would expect at this stage, if possible (?), is to emphasize in more detail the general points relating to the shortcomings of the studies included in the review and, probably, to suggest more general suggestions required to improve the outcome of this form of study.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear authors,

Non-pharmacological interventions to improve chronic disease risk factors and sleep in shift workers: A systematic review and meta-analysis

Thanks for this systematic review and meta-analysis. In the following sections, you will find several comments/suggestions to improve the quality of the manuscript.

Specific comments:

Lines 16-17: Please move qualitative synthesis after systematic review (before meta-analysis).

Line 21: Could the authors please clarify and provide comments regarding the following sentence? "Thirty-nine studies were included across 8 random effects meta-analyses" Do you refer to eight subclasses? Use "random-effects".

Lines 21-29: Authors should summarize findings on most important variables with quantitative data (e.g., Pooled Effect size: confidence interval of the effect
size; heterogeneity) and provide a short and remarkable conclusion.

Line 30: Authors should add more keywords accordingly. Please consider MeSH terms. Revise "interventions" and "shift work" since these are not MeSH terms.

Line 52: Replace "abnormal blood lipids" with "dyslipidemia". Please replace also in Table 1.

Line 63: Use a semicolon before however (... relationship; however, ...).

Line 99: A registration number should be listed and several sections (e.g, eligibility criteria, information sources, search, data collection process,
among others) should be described if you followed up PRISMA statement. If authors do not modify to follow this, please remove the statements related to PRISMA.

Line 109: Please change the statement "The PRISMA flowchart (Figure 1) and total search results are inclusive of updated searches" to the RESULTS section / "Study selection" subsection. Please follow the PRISMA statement accordingly.

Line 113: I understand the inclusion of AIHW's description in the introduction section to support the rationale, but why are the authors limiting to local impact (i.e., Australia)? You might remove AIHW since this corresponds to a global phenomenon.

Line 125: Please clarify what you defined as "valid measures".

Line 135: Modify to (M.E.C. and A.C.R.). Same hereinafter.

Line 138: Use a semicolon before however. Please correct to "PRISMA".

Line 155: Why zero have two meanings? Was this a typo? Please clarify if not.

Line 159-160: How was this agreement measured? Please report it.

Line 166-167: If you considered the Cochrane Handbook for Systematic Review of Interventions, please change "A priori sub-analysis" to "Pre-specified subgroup analysis".

Line 169-177: Good job.

Line 191: Please describe methodology only. Reporting "three studies [54, 60, 61]..." should be in the RESULTS section / "Synthesis of results" subsection.
Same hereinafter.

Line 205: Ibidem.

Line 221: Please, can the authors report how the heterogeneity was assessed? By looking at the results section, it seems you implemented I²; however, why did not include another approach to overcome the relevant drawbacks of this one?

RESULTS

Line 226: Grammar editing.

Line 261: Given you did not consider a date range, please report the literature obsolescence (e.g., Price's index).

Line 278: One of the arguments for full-text exclusion was "Slept in lab" (Figure 1). Item (c) states that some selected studies were "not eligible for inclusion in meta-analysis (... laboratory sleep...)". Can the authors clarify this, please?

Table 3: Please amend the symbols of units according to the SI (e.g, the symbol of the hour is h, not hr).

Line 285-415: The section "Risk of bias across studies" is missing. Please include this section and use the graphical results of the risk of bias assessment.

Line 419: Why qualitative results if you include here results of the meta-analysis?

Figure 2-8. Please include in all forest plots the favoring effect accordingly (e.g., favored treatment vs favored control/placebo).

DISCUSSION

The discussion section is concise and generally well presented. I would suggest the authors could also add the relevant meta-analytic findings in the section and to discuss in terms of the risk of bias of the selected references.

CONCLUSION

The conclusion section gives a nice overview of the considerations and findings of this systematic review and meta-analysis. However, I think the authors need to have some discussion relating to the aforementioned results/discussion section to lead into these subsequent points and link the whole narrative of the paper throughout.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper Non-pharmacological interventions to improve chronic disease risk factors and sleep in shift workers: A systematic review and meta-analysis, by Meagan E Crowther at all is a complex analysis of the most relevant publication related to this topic. It covers all the literature, in a very careful overview, with solid statistics and relevant conclusion.

There are some issues that needs to be solved by the authors:

1. page 3, introduction, line 64: In additional to the chronic disease risk factors outlined above, sleep loss is associated with an increased risk of chronic diseases [32-39].

I think you can mention those diseases in the text: insulin resistance, obesity, diabetes, chronic kidney disease, stroke, etc..

2. page 4. line 124: Measures could be self-report or objective measures. 

How significant/relevant is self-reported vs. objective measurements?

3. page 6, line 225. There are some discrepancies in the text and in the figure 1: 8161 papers in the text vs 8465 records in the figure 1. Additional records identified 55 in the text vs. 54 in the table. 

Empty space connected to full text excluded ... in the figure 1. You just need to make it clearer. 

4. Many issues with numbers and references: page 8, line16. In Karlson et al., (26) is at 64 position on the references list. 

5. page 11, line 14: However, sleep education utilising individualised cognitive behavioural therapy resulted in improved subjective sleep quality [97]. The reference is related to healthy meals at work...

6. From references 79, it seems that there are a lot of wrong placements of numbers and related references. You need to do a close check again to make the right connection. Maybe it is an error when using a editing program. However, this is an important issue. 

7. Some other examples: in the text is 98 but it is most likely 97 in the refence list.  118-119 instead of 119,120. 

118 added to  list of papers about light blocking glasses...

page 9, 120 in the text, probably 119 in the reference list. 

line 558:  99 instead of 100.

line 586: 104 instead of 105. 

line 607: 117 instead of 118. 

line 612: there is no 130 reference in the list. 

Because is a nicely done but long paper, maybe you can summarize the conclusion in a shorter and clearer way, if you think it is possible. 

You have done a lot of work, indeed. A good one!

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

The authors present a sophisticated, mature, and important and very valuable summary on the associations between shift work and adverse health outcome. The authors further summarize possible interventions and their effect on health parameters, in general, and sleep dimensions, in specific. The interventions are – luckily – limited to non-pharmacological interventions.

Overall, the manuscript is very well written; it is easy to follow the flow of the text. The authors were able to describe complex processes in a smart and easy-to-understand fashion.

The authors may discuss the following issues: how to deal with the chicken-and-egg-problem? Or the other way around? Is it possible that for instance people with a vulnerability to cardiovascular diseases and overweight are involved in job positions with higher risks of shift work?

How come female fertility was not mentioned among psychophysiological consequences of shift work? In my opinion, there is sufficient evidence that adverse working conditions, including shift work, are associated with lower fertility rates among female workers (Kloss et al., 2015). Perhaps, the authors might briefly explain their decision not to include this health issue, which also impacts on female identity and family planning.

Reference

Kloss, J.D., Perlis, M.L., Zamzow, J.A., Culnan, E.J., Gracia, C.R., 2015. Sleep, sleep disturbance, and fertility in women. Sleep medicine reviews 22, 78-87.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear authors, 

Thanks for working on the manuscript revision. Very good job.

Few comments in the following lines: 

• Abstract: I disagree with the author's response "However, there is a 200-word limit on the abstract and therefore, it is not possible to include the statistics and all other crucial details within this word limit." Please reorganize what you consider necessary in the abstract to report the most important quantitative findings.
  FYI - I let three meta-analysis examples including this basic statistical information that are found in MDPI Journals (organized by relevance):
  https://www.mdpi.com/1010-660X/55/10/673
  https://www.mdpi.com/2077-0383/9/5/1553
  https://www.mdpi.com/1660-4601/18/2/828
• Methods: Please highlight that this review intended to follow PRISMA guidelines (the protocol was not registered). I miss the Tau² regarding heterogeneity, so take it into account for your future SR with MA. 

Author Response

Thank you

Reviewer 3 Report

Thank for providing answers to all questions.

Author Response

Thank you

Reviewer 4 Report

In my opinion, the authors addressed well the reviewers' concerns and suggestions. 

The manuscript is interesting, important to the field and also an excellent point of reference for individuals interested in the topics.

Author Response

Thank you.