Work-Related Musculoskeletal Disorders of Dance Teachers in Germany: A Retrospective Cross-Sectional Study
, Rüdiger Reer 1, David A. Groneberg 2, Fabian Holzgreve 2 and Eileen M. Wanke 2,*Round 1
Reviewer 1 Report
Schmidt et al. utilize a retrospective cross-sectional survey to analyze work-related musculoskeletal disorders in dance teachers while accounting for sex and dance style. There are some limitations that are intrinsic to retrospective, exploratory investigations. The authors are very clear in pointing these out and qualifying their results based on the data collected. Overall, I think the study is well-structured and scientifically sound.
I recommend publication after the authors clarify the following minor comments.
1) "A final response rate could not be calculated due to the lack of information on the total population of German DTs" Please clarify why a reasonable estimate of the total population of German DTs is unavailable.
2) "N = 12 participants were excluded because important anthropometric information (e.g. age) and/or information on WMSD occurrence was missing." What were the mandatory questions on the survey once participation was agreed upon? I am assuming sex and dance style information was collected by all participants. Why not age?
3) "Based on the valid data (n = 209), the percentage of teachers in classical dance was 26.8%, 32.1% in modern and contemporary dance and 41.1% for the combination of all dance styles." Shouldn't the value for be 229?
Author Response
Comments and Suggestions for Authors
Schmidt et al. utilize a retrospective cross-sectional survey to analyze work-related musculoskeletal disorders in dance teachers while accounting for sex and dance style. There are some limitations that are intrinsic to retrospective, exploratory investigations. The authors are very clear in pointing these out and qualifying their results based on the data collected. Overall, I think the study is well-structured and scientifically sound.
I recommend publication after the authors clarify the following minor comments.
Thank you very much for the review and the kind feedback. Below you will find our responses to your comments.
1) "A final response rate could not be calculated due to the lack of information on the total population of German DTs" Please clarify why a reasonable estimate of the total population of German DTs is unavailable.
Answer: The professional title is not legally protected. In addition, there is no uniform and systematic recording and registration of this professional group. Compulsory membership in a professional association is not necessary.
2) "N = 12 participants were excluded because important anthropometric information (e.g. age) and/or information on WMSD occurrence was missing." What were the mandatory questions on the survey once participation was agreed upon? I am assuming sex and dance style information was collected by all participants. Why not age?
Answer: In the run-up to the survey, we thought a lot about the advantages and disadvantages of mandatory question items. We decided to give participants the option of skipping individual items in order to avoid complete abandonment of the survey and to increase the response rate.
3) "Based on the valid data (n = 209), the percentage of teachers in classical dance was 26.8%, 32.1% in modern and contemporary dance and 41.1% for the combination of all dance styles." Shouldn't the value for be 229?
Answer: Unfortunately, only 209 answered this item due to the survey strategy as mentioned before.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments to
Work-related musculoskeletal disorders of dance teachers in Germany: a retrospective cross-sectional study
The study aimed to assess the prevalence of work-related disorders and their causes in dance teachers. Overall, interesting results are presented. However, the writing needs to be checked on suggested cause-effect relationships, terms like influence, relation, affect, etc. need to be rewritten towards association indicating no clear cause and effect connection (considering that the design is cross-sectional). This also considers the indication of the work-relatedness of the symptoms as these are self-reported. Limitations of this prevalence study need to be discussed more thoroughly.
Abstract:
- Line 17: I suggest to change the prevalence to percentage
- Line 24: it is stated that “dance educators were neither significantly less nor more affected”, I suggest to rephrase this to something like “dance educators were not significantly differently affected for musculoskeletal disorders than the general working population…”
Introduction:
- Although a short and to-the-point introduction is preferable, I think this one lacks some in-depth arguments. It should be stated more clearly why dance teachers are specifically of interest to study (relevancy), what is known about the exposure to risk factors in this population that may cause musculoskeletal disorders, is this different from other physical occupations etc. Please present some numbers of other studies if available.
- I do miss a clear specific research question in the introduction.
- I suggest to replace ‘complaints’ by ‘symptoms’ in the entire paper.
- All results are presented separately for males and females, this should also be substantiated in the introduction.
Methods:
- Line 70: Already a specific definition of WRMD is given. However, it raises some questions. In what are disorders different from impairments (or what exactly are impairments)? What is considered chronic? Considering the impairments as chronic (>3 months continuously present?) may be confusing with the combination of a prevalence definition of a new symptom in the previous 12 months. The actual questions may be provided in an appendix.
- Arriving at ‘questionnaire design’ and ‘testing procedure’ this seems to overlap with ‘study design’ and ‘WRMD definition’. Following the questionnaire design, outcome assessment could be explicitly defined, i.e. the Nordic Questionnaire assesses initially the ‘trouble (ache, pain, discomfort)’ for musculoskeletal symptoms. As the definition of WRMD follows the assessment, it needs to be explicitly explained how the outcomes were assessed. In addition, the authors are advised to follow the STROBE statement and explanation to arrive at a sufficient level of information in the paper. In the current paper, too little information on exposure, outcome and confounder assessment is provided in the methods section.
- Line 100: it is stated that an incidence rate is calculated, but this should be prevalence rate considering that a cross-sectional design is used.
- In the statistical analysis it is indicated that Cramer’s V, phi and r are used as effect size measures. For V and phi it is not indicated what can be considered weak / moderate / strong effects. I would suggest to recalculate all effect sizes towards r.
Results:
- Line 135: I suggest to keep the prevalences at a percentage and not revert it to a proportion.
- For the figures, indicate in the captions how ‘relative’ was defined (relative to what?).
- Are large emphasis is given to sex differences in the results. If this is an important issue, this should be handled in the introduction and research question (why is it important and relevant to explore sex differences?).
- Line 197: I suggest not to use the heading ‘aetiology’ as this suggests cause and effect. Cause and effect cannot be identified in the present study considering that it is a cross-sectional study, the actual history (before the study) in exposure and outcome in the past has not been assessed and work-relatedness was assessed subjectively.
- Lines 205/208: write p<0.001 instead of p<0.000
Discussion:
- First sentence: it needs to be substantiated why the 12-month prevalence indicates an increased risk for WRMD in dance teachers. There is no direct comparison with other groups and it concerns cross-sectional data, thus it is questionable whether this can be concluded based on the results only. I suggest to start the discussion with the first conclusions based on the actual results. The next paragraph states no higher prevalence compared to the general population.
- Line 244: What are PE teachers? Ah, physical education?
- How valid is it that participants are able to determine whether their symptoms are work-related? What is the percentage of the number of work-related symptoms compared to the total number of musculoskeletal symptoms?
- The limitations mentioned are by itself limited, it is a cross-sectional study with subjective assessment of exposure, other risk factors, work-relatedness and symptoms. A lot of limitations can be mentioned, with the most important one that statements about cause and effect cannot be made.
- In general, I find the discussion of moderate quality. I suggest to strive for some larger paragraphs that discusses some important issues more in-depth. In the present version too many small paragraph are included that only point to some issues.
Author Response
Comments and Suggestions for Authors
Comments to
Work-related musculoskeletal disorders of dance teachers in Germany: a retrospective cross-sectional study
First of all, we would like to thank you for the many constructive and critical comments, which have also given us another perspective on our own work. In the following, we have processed all comments and taken them into account as best as possible for our revision. Thank you very much for your commitment.
The study aimed to assess the prevalence of work-related disorders and their causes in dance teachers. Overall, interesting results are presented. However, the writing needs to be checked on suggested cause-effect relationships, terms like influence, relation, affect, etc. need to be rewritten towards association indicating no clear cause and effect connection (considering that the design is cross-sectional). This also considers the indication of the work-relatedness of the symptoms as these are self-reported. Limitations of this prevalence study need to be discussed more thoroughly.
Answer: Thank you for this comment. We have tried to take this aspect more into account and to emphasize it within the work, especially in the limitations. It is definitely not our intention to show a cause-effect relationship, knowing that this study is exploratory and retrospective. We hope that this aspect will now be more clearly reflected.
The term ‘influence’ has now been avoided throughout the article to avoid suggesting a cause-effect relationship.
Abstract:
- Line 17: I suggest to change the prevalence to percentage
Answer: Prevalence was changed to percentage.
- Line 24: it is stated that “dance educators were neither significantly less nor more affected”, I suggest to rephrase this to something like “dance educators were not significantly differently affected for musculoskeletal disorders than the general working population…”
Answer: Has been adapted in line with the recommendation.
Introduction:
- Although a short and to-the-point introduction is preferable, I think this one lacks some in-depth arguments. It should be stated more clearly why dance teachers are specifically of interest to study (relevancy), […]
Answer: In a way we understand this argument, but on the other hand we find that the high physical demands from the literature and the goal to maintain health already suggest a need for at least a first exploratory study.
[…] what is known about the exposure to risk factors in this population that may cause musculoskeletal disorders, is this different from other physical occupations etc. Please present some numbers of other studies if available.
Answer: Statements about concrete figures on exposure to risk factors in this population are extremely limited or in a different context (e.g. acute injuries due to occupational accidents). That is the reason and the aim of this explorative cross-sectional study.
- I do miss a clear specific research question in the introduction.
Answer: A basic research question is stated in the introduction (line 45/46) and the relevance justified by the high physical demands and the maintenance of health of this occupational group.
- I suggest to replace ‘complaints’ by ‘symptoms’ in the entire paper.
Answer: Has been adapted. In line 247 the wording of the reference source was used.
- All results are presented separately for males and females, this should also be substantiated in the introduction.
Answer: We hope the consideration of the comparison between the sexes in the context of professional dance and a possible dance career in the past is now more plausible.
Methods:
- Line 70: Already a specific definition of WRMD is given. However, it raises some questions. In what are disorders different from impairments (or what exactly are impairments)? What is considered chronic? Considering the impairments as chronic (>3 months continuously present?) may be confusing with the combination of a prevalence definition of a new symptom in the previous 12 months. The actual questions may be provided in an appendix.
Answer: Thank you very much for this comment. It was clear to us that in the course of the extensive online survey and self-assessment, a clear definition or explanation of these terms to the participants was hardly possible. For practical reasons, we did not consider a more detailed explanation of the interpretation of the terms in the course of the survey to be feasible, and this is certainly a major difference compared to the recording of medically trained staff. Thus, the interpretation of the terms was left to the participants.
We agree that disorders and impairments can be confusing. Therefore, impairments were replaced with symptoms here as well.
- Arriving at ‘questionnaire design’ and ‘testing procedure’ this seems to overlap with ‘study design’ and ‘WRMD definition’. Following the questionnaire design, outcome assessment could be explicitly defined, i.e. the Nordic Questionnaire assesses initially the ‘trouble (ache, pain, discomfort)’ for musculoskeletal symptoms. As the definition of WRMD follows the assessment, it needs to be explicitly explained how the outcomes were assessed. In addition, the authors are advised to follow the STROBE statement and explanation to arrive at a sufficient level of information in the paper. In the current paper, too little information on exposure, outcome and confounder assessment is provided in the methods section.
Answer: Text passage was adapted. Information on main outcome parameters, exposure and possible confounders was added. We hope it is acceptable that an even deeper consideration of confounding variables has not been made due to the exploratory nature of the study.
- Line 100: it is stated that an incidence rate is calculated, but this should be prevalence rate considering that a cross-sectional design is used.
Answer: Has been adjusted throughout the paper.
- In the statistical analysis it is indicated that Cramer’s V, phi and r are used as effect size measures. For V and phi it is not indicated what can be considered weak / moderate / strong effects. I would suggest to recalculate all effect sizes towards r.
Answer: The passage was rephrased. We would like to keep the classification according to V, Phi and r in order to keep the relation to the respective test procedure (e.g. Chi-square or Mann-Whitney-U).
Results:
- Line 135: I suggest to keep the prevalences at a percentage and not revert it to a proportion.
Answer: The recommendation was followed. The prevalence values were adjusted.
- For the figures, indicate in the captions how ‘relative’ was defined (relative to what?).
Answer: Thank you for the comment. The information was supplemented for each figure.
- Are large emphasis is given to sex differences in the results. If this is an important issue, this should be handled in the introduction and research question (why is it important and relevant to explore sex differences?).
Answer: As mentioned above. Hopefully a little more plausible now.
- Line 197: I suggest not to use the heading ‘aetiology’ as this suggests cause and effect. Cause and effect cannot be identified in the present study considering that it is a cross-sectional study, the actual history (before the study) in exposure and outcome in the past has not been assessed and work-relatedness was assessed subjectively.
Answer: The headline has been changed into ‘Subjective cause attribution’.
- Lines 205/208: write p<0.001 instead of p<0.000
Answer: Has been adjusted.
Discussion:
- First sentence: it needs to be substantiated why the 12-month prevalence indicates an increased risk for WRMD in dance teachers. There is no direct comparison with other groups and it concerns cross-sectional data, thus it is questionable whether this can be concluded based on the results only. I suggest to start the discussion with the first conclusions based on the actual results. The next paragraph states no higher prevalence compared to the general population.
Answer: The comment is absolutely correct and the text passage has been adjusted.
- Line 244: What are PE teachers? Ah, physical education?
Answer: Has been adapted.
- How valid is it that participants are able to determine whether their symptoms are work-related? What is the percentage of the number of work-related symptoms compared to the total number of musculoskeletal symptoms?
Answer: The comment is absolutely correct and the authors are aware of the problem. And as you mention in the next point, the self-assessments are a limitation of the results in terms of their validity (We just asked for the work-related symptoms). The survey cannot be equated with a precise and selective measuring instrument. Hence the comment that these are not medical diagnoses within the scope of the study. We have now tried to emphasise this point even more in the limitations.
- The limitations mentioned are by itself limited, it is a cross-sectional study with subjective assessment of exposure, other risk factors, work-relatedness and symptoms. A lot of limitations can be mentioned, with the most important one that statements about cause and effect cannot be made.
Answer: Correct! Added in the limitations.
- In general, I find the discussion of moderate quality. I suggest to strive for some larger paragraphs that discusses some important issues more in-depth. In the present version too many small paragraph are included that only point to some issues.
Answer: It remains somewhat open which aspects exactly are meant. If there is a clear need here, it would be very helpful if a concretisation of such points could be proposed.
Otherwise, the aim was to correspond to the explorative character of the study and, above all, to point out starting points for future work. The authors wanted to avoid an overinterpretation of their own results. Because, as you thankfully correctly pointed out, statements on cause-effect relationships are not possible (added in the discussion as limitation) and there is a great danger of lapsing into speculation in the context of deeper discussion. We hope this is comprehensible?!
