Eating Disorders and Disordered Eating in Competitive Cycling: A Scoping Review

Round 1

Reviewer 1 Report

I enjoyed reading this scoping review on disordered eating in competitive cycling. The manuscript is well written with clear aims, seems methodologically sound and presents findings in  a clear manner.

The introduction provides adequate information to the background and justification for this review.

The methods are clearly written and adhere to PRISMA guidelines for scoping reviews. There is a mismatch between the items listed for data extraction and Table 1 in results. The list is the better of the two and I suggest re-formatting the table to reflect everything listed in the methods section.

Table 1 in the  results section could be better formatted and should include the author details in column 1. There doesn’t seem to be logic to their order (by author name / year / discipline), and this could be presented in a more structured manner. The citation number might be better in the last column or superscripted with the author name, but not as is. Where the table runs over multiple pages, it would be helpful to the reader to have the headers across the top of each new page.

Results:

Is Ref 39, Netherlands 2018 a good inclusion – there is only 1 cyclist and very little other data available / presented in Table 1.

Ref 36 UK 2002 has a typo in participant details (&)

It seems worthy of comment that no study included used the LEAF or LEAM questionnaires. I think this is worthy of comment.

There are few studies of any cyclist, but fewer still focusing on females. Is it possible to do a sub analysis on the female only data to  compare? This would be a valuable addition.

Line 198 studies reported on below should read above / in Table 1

Line 246 Please superscript -1 here and throughout

Quite a few studies included are missing data. Would the authors consider a quality assessment or risk of bias assessment to critically appraise the studies included.

Conclusions

This section might be improved with the addition of comparisons of prevalence to other sports, as it is not clear in the current format if athletes generally are at increased risk for DE or ED, or specifically cyclists.

Formatting

The references change in format from Line 555, Number 40. Please ensure you are citing / referencing as per author guidelines consistently throughout.

Overall, a well written manuscript with some formatting / presentation issues. My main recommendation is around Table 1 presentation and greater criticality in interpreting the results, as well as the inclusion of assessments of bias and quality assessments.

 

Author Response

Dear reviewer,

 

Thank you for your thorough review of our manuscript. Please see below for our responses to your comments. We look forward to hearing back from you.

Reviewer 1

The methods are clearly written and adhere to PRISMA guidelines for scoping reviews. There is a mismatch between the items listed for data extraction and Table 1 in results. The list is the better of the two and I suggest re-formatting the table to reflect everything listed in the methods section.

 

Table 1 in the  results section could be better formatted and should include the author details in column 1. There doesn’t seem to be logic to their order (by author name / year / discipline), and this could be presented in a more structured manner. The citation number might be better in the last column or superscripted with the author name, but not as is. Where the table runs over multiple pages, it would be helpful to the reader to have the headers across the top of each new page.

Thank you for the comments regarding the Table. We have implemented the changes requested and agree that this additional information will strengthen the table and thus the manuscript. Due to the changes the Table has become too large to incorporate into the manuscript and we have attached the Table as a separate document, however we would still like it to be in the manuscript. We also agree with the comment regarding the headers across the top of each new page. We will request these be incorporated from the editorial team.

Results:

Is Ref 39, Netherlands 2018 a good inclusion – there is only 1 cyclist and very little other data available / presented in Table 1.

We believe this is a good inclusion as the single participant in the only qualitative study that was found during searching supports our recommendation that further qualitative work should be conducted in the area.

Ref 36 UK 2002 has a typo in participant details (&)

Addressed in manuscript.

It seems worthy of comment that no study included used the LEAF or LEAM questionnaires. I think this is worthy of comment.

Thank you for the suggestion. The authors feel the LEAF and LEAM sit outside the purpose of the scoping review, with their focus being on LEA as a whole, and none of the questions particularly pertaining to ED or DE specifically. It would be worthwhile for future research to utilise multiple tools like the LEAF/LEAM and EAT-26 to build a broader picture on the inter-relationship between ED and LEA.

There are few studies of any cyclist, but fewer still focusing on females. Is it possible to do a sub analysis on the female only data to compare? This would be a valuable addition.

We agree that a sub-analysis would be useful, however don’t feel it is appropriate for our manuscript.  With the purpose of a scoping review being to “explore and systematically map the literature on a topic by identifying key concepts, theories and sources of evidence” it would seem out of place (but would be incredibly beneficial for a future systematic review and meta-analysis should more evidence become available).

Line 198 studies reported on below should read above / in Table 1

Addressed in manuscript.

Line 246 Please superscript -1 here and throughout

Addressed in manuscript.

Quite a few studies included are missing data. Would the authors consider a quality assessment or risk of bias assessment to critically appraise the studies included.

Whilst we agree this would strengthen the manuscript, it seems inappropriate for the purpose of a scoping review for the reason highlighted in the comment “There are few studies of any cyclist, but fewer still focusing on females. Is it possible to do a sub analysis on the female only data to compare? This would be a valuable addition.”

Conclusions

This section might be improved with the addition of comparisons of prevalence to other sports, as it is not clear in the current format if athletes generally are at increased risk for DE or ED, or specifically cyclists.

Thank you for highlighting this, particularly as we have mentioned that cyclists are uniquely susceptible to developing DE/ED. We have addressed this in the manuscript.

Formatting

The references change in format from Line 555, Number 40. Please ensure you are citing / referencing as per author guidelines consistently throughout.

Addressed in manuscript. The authors experienced technical issues with Mendeley Cite during the manuscript preparation!

Overall, a well written manuscript with some formatting / presentation issues. My main recommendation is around Table 1 presentation and greater criticality in interpreting the results, as well as the inclusion of assessments of bias and quality assessments.

 

Reviewer 2 Report

behavsci-2050707

1. The review analyzed 14 publications on eating disorders in cyclists, which is clearly insufficient to obtain any reliable conclusions about the problem under consideration, and even more so to formulate recommendations for the prevention of eating disorders.

2. The methodology of analysis and classification of literature does not raise objections. This part of the study was done very carefully and painstakingly.

3. From my point of view, the authors did not approach the theoretical justification of the study carefully enough. This led to the fact that a whole layer of concepts closely related to the topic under consideration were not included in the search and therefore not reflected in the main results. I am referring to the long-described concept of "The female athlete triad", an important component of which is "anorexia athletica". By the way, in the review (Tenforde et al. Sports Medicine, 2016. 46(2), 171-182) it is postulated that similar symptoms occur in men. I am sure that if the authors had included these terms in the search, they would have found significantly more publications on eating disorders in cyclists, described a wider range of health problems for them, and also provided theoretically more sound recommendations for their prevention.

4. Another important disadvantage of the study is that the authors have very much limited the scope of the study, including only those dedicated to cyclists. I looked at the lists of the authors' previous publications and found that they are mainly devoted to the research of cyclists. Apparently, the narrow specialization of the authors did not allow them to look at the problem in question more broadly. If they had carried out a preliminary theoretical analysis of the problem, they could have included in the study works devoted to skiers who previously (Sundgot-Borgen, Torstveit, 2010) had exactly the same list of risk factors for the development of eating disorders as cyclists. Perhaps a theoretical analysis would allow the authors to pick up some other sports that are similar in their mechanism of action on human eating behavior.

5. In conclusion, the authors write: "The sport presents unique risk factors ...". This statement can be interpreted in two senses, but in both it does not correspond to the content of the review. 1. If the authors mean cycling, then the uniqueness of risk factors is questionable (see comment 4). 2. If the authors mean sports in general, then this statement is also incorrect, since it can only be about competitive sports of high intensity (Sundgot-Borgen, Torstveit, 2010).

6. Further, the authors write: "Greater detail and consistency is required in the literature when reporting the anthropometric and demographic information of participants, cycling discipline, levels and quantification of training and racing volume." In general, I agree with this statement, but if the authors considered the problem more broadly, they could have proposed specific anthropometric and physiological indicators for selecting athletes at risk of eating disorders (the athlete triad symptoms).

Author Response

Reviewer 2

Thank you for your time reviewing this manuscript and your feedback on the paper. Please find our responses to each point raised below.

The review analyzed 14 publications on eating disorders in cyclists, which is clearly insufficient to obtain any reliable conclusions about the problem under consideration, and even more so to formulate recommendations for the prevention of eating disorders.

This was a scoping review, with the purpose to scope what is currently available in the literature. The authors agree that 14 publications is insufficient to obtain reliable conclusions, but this is not the aim of the paper (or scoping reviews in general). We aimed to address this by suggesting that “more research on the area is required” in the abstract and conclusions, alongside highlighting that the reliance on cross-sectional self-reported surveys is limiting. Furthermore, the recommendations made by the authors are not intended to prevent eating disorders, but are instead highlighting the direction for future research to formulate a deeper understanding of the problem, which may in turn influence practices and policy in the sport.

2. The methodology of analysis and classification of literature does not raise objections. This part of the study was done very carefully and painstakingly.

Thank you for this positive feedback.

3. From my point of view, the authors did not approach the theoretical justification of the study carefully enough. This led to the fact that a whole layer of concepts closely related to the topic under consideration were not included in the search and therefore not reflected in the main results. I am referring to the long-described concept of "The female athlete triad", an important component of which is "anorexia athletica". By the way, in the review (Tenforde et al. Sports Medicine, 2016. 46(2), 171-182) it is postulated that similar symptoms occur in men. I am sure that if the authors had included these terms in the search, they would have found significantly more publications on eating disorders in cyclists, described a wider range of health problems for them, and also provided theoretically more sound recommendations for their prevention.

Thank you for the suggestion. A cursory search for “female athlete triad” AND cycling produced 2 results on PubMed, none of which fit the inclusion criteria. Additionally, the purpose of the review was not to provide recommendations for their prevention but to highlight areas for future research.

4. Another important disadvantage of the study is that the authors have very much limited the scope of the study, including only those dedicated to cyclists. I looked at the lists of the authors' previous publications and found that they are mainly devoted to the research of cyclists. Apparently, the narrow specialization of the authors did not allow them to look at the problem in question more broadly. If they had carried out a preliminary theoretical analysis of the problem, they could have included in the study works devoted to skiers who previously (Sundgot-Borgen, Torstveit, 2010) had exactly the same list of risk factors for the development of eating disorders as cyclists. Perhaps a theoretical analysis would allow the authors to pick up some other sports that are similar in their mechanism of action on human eating behavior.

Whilst we appreciate that eating disorders and disordered eating are not unique to cycling, the purpose of the manuscript is to discuss these specifically in cycling and we therefore felt it appropriate to keep the scope narrow. Furthermore, the manuscript was submitted to the special issue entitled “Assessing Injury Risk and Cycling Performance” and including other disciplines would not fit within this.

5. In conclusion, the authors write: "The sport presents unique risk factors ...". This statement can be interpreted in two senses, but in both it does not correspond to the content of the review. 1. If the authors mean cycling, then the uniqueness of risk factors is questionable (see comment 4). 2. If the authors mean sports in general, then this statement is also incorrect, since it can only be about competitive sports of high intensity (Sundgot-Borgen, Torstveit, 2010).

Thank you for the comment. The sport of cycling presents unique risk factors when one considers the interplay of physiological demands of the sport, the perception of leanness correlating with improvements in performance, and the social environment of cycling. The last point has not been addressed much in the literature, hence the call from the researchers for further qualitative research in particular. Whilst we acknowledge that the social environment of other sports may also present a risk factor, our interpretation of cycling possessing “unique risk factors” isn’t that no other sport also contains these risk factors, but that cycling practitioners should be cognizant of these.  

6. Further, the authors write: "Greater detail and consistency is required in the literature when reporting the anthropometric and demographic information of participants, cycling discipline, levels and quantification of training and racing volume." In general, I agree with this statement, but if the authors considered the problem more broadly, they could have proposed specific anthropometric and physiological indicators for selecting athletes at risk of eating disorders (the athlete triad symptoms).

As mentioned previously, the aim of our scoping review was not to provide direct practical recommendations for the identification of athletes at risk of eating disorders, but more to highlight the current state of literature in the area and address the themes arising in this.

Round 2

Reviewer 2 Report

No comments.