Biopsychosocial Markers of Body Image Concerns in Patients with Head and Neck Cancer: A Prospective Longitudinal Study

Round 1

Reviewer 1 Report

We know that there are few studies with patients with head and neck cancer, given the vulnerability of this group of patients. We congratulate the authors for their initiative in studying this population, mainly through a longitudinal study, contributing to improving the scientific evidence.

Aware of the complexity of longitudinal studies, particularly concerning sample attrition, were there participants who dropped out for a different reason? What was the baseline sample of the study?

It would be interesting for the reader and future studies to better understand the characterization of the selection process: for example, identify which patients underwent surgery, radiotherapy, chemotherapy, or a combination of several treatments.

Also related to the sample selection, the authors said they use interviews (and questionnaires). Some patients with head and neck cancer may have speech changes (dysphonia, dysarthria, aphonia). How do you do with this kind of participant? They were excluded, or were there used some additional alternative augmentative communications strategies? Could the authors explain this process to the readers?

We suggested that updated some references for greater scientific soundness. We noticed that about half of the references present in the article are more than five years old.

The article presents relevant, adequate and interesting information for professionals who care for these patients.

Author Response

1. We know that there are few studies with patients with head and neck cancer, given the vulnerability of this group of patients. We congratulate the authors for their initiative in studying this population, mainly through a longitudinal study, contributing to improving the scientific evidence.
   1. Thank you for your appreciation of our article.

 

2. Aware of the complexity of longitudinal studies, particularly concerning sample attrition, were there participants who dropped out for a different reason? What was the baseline sample of the study?
   1. We have added a sentence about participant attrition stating that “Participant attrition at 3-months, 6-month, and 12-months was not found to be due to sociodemographic (sex, age) and medical (stage) differences at baseline; and attrition at 3 months was not due to psychological differences. Patients having dropped out at 6 and 12 months were found to have higher levels of psychological distress at baseline (p=0.09 and p=0.048, respectively; depression p=0.006-0.009; anxiety p=0.31-0.47). Those who did not complete follow-up at 3-months presented with significantly lower ECOG functioning at baseline (p<0.005; 6- and 12-months p=0.41-0.72). There was a trend for 3-month disfigurement rates to be higher in the drop-out group at 12 months (p=0.09) but not at 6 months (p=0.62-1.0). ” on lines 269-278.

 

3. It would be interesting for the reader and future studies to better understand the characterization of the selection process: for example, identify which patients underwent surgery, radiotherapy, chemotherapy, or a combination of several treatments.
   8. We have added this information in Table 1 as suggested on page 8.

 

4. Also related to the sample selection, the authors said they use interviews (and questionnaires). Some patients with head and neck cancer may have speech changes (dysphonia, dysarthria, aphonia). How do you do with this kind of participant? They were excluded, or were there used some additional alternative augmentative communications strategies? Could the authors explain this process to the readers?
   1. We added the following explanation about how difficulties in speech were dealt with during the study (see lines 152-154): “We used written means of communicating with patients when needed in consideration of their speech function/impairments”.

 

5. We suggested that updated some references for greater scientific soundness. We noticed that about half of the references present in the article are more than five years old.
   1. We have reviewed all of the references and have updated them when possible.

 

6. The article presents relevant, adequate and interesting information for professionals who care for these patients.
   1. Thank you for your comments and diligent review.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors Albert et al. present a longitudinal biopsychosocial study on body image in head and neck cancer (HNC) patients. The patients were asked to fill in several questionaires on quality of life, psychological disorders, and body image upon diagnosis, 3 months, 6 months and 1 year after diagnosis. They find that HNC patients have higher rates of body disfigurement and rates are highest 3 months after diagnosis.

Unfortunately, information about clinical characteristics of the patients is scarse and this might bias the results. In HNC collectives, both disease-related factors such as the localization (e.g. oropharynx vs. larynx), T and N stage as well as therapy-related morbidity (e.g. extensive surgery such as radial free flap surgery, laryngectomy, extensive radiotherapy of the neck in higher N stages). It is expectable that these changes are most prevalent shortly after treatment and, therefore at T1 of the study concept. But with excluding such important possible confounders, discussing and explaining the results found with underlying anxiety disorder/depression or even french language only seems devious.

Therefore, addition of this clinical information and including these time-invariant factors into the formal analysis of the results seems mandatory to strengthen the manuscript and make it suitable for publication.

Minor: Some style modifications are needed, e.g. the sub-headings in the methods section (page 5).

Author Response

The authors Albert et al. present a longitudinal biopsychosocial study on body image in head and neck cancer (HNC) patients. The patients were asked to fill in several questionnaires on quality of life, psychological disorders, and body image upon diagnosis, 3 months, 6 months and 1 year after diagnosis. They find that HNC patients have higher rates of body disfigurement and rates are highest 3 months after diagnosis.

1. Unfortunately, information about clinical characteristics of the patients is scarce and this might bias the results. In HNC collectives, both disease-related factors such as the localization (e.g. oropharynx vs. larynx), T and N stage as well as therapy-related morbidity (e.g. extensive surgery such as radial free flap surgery, laryngectomy, extensive radiotherapy of the neck in higher N stages). It is expectable that these changes are most prevalent shortly after treatment and, therefore at T1 of the study concept. But with excluding such important possible confounders, discussing and explaining the results found with underlying anxiety disorder/depression or even french language only seems devious. Therefore, addition of this clinical information and including these time-invariant factors into the formal analysis of the results seems mandatory to strengthen the manuscript and make it suitable for publication.
   1. We have added more clinical characteristics of the patients and general descriptions of our sample on lines 256-260 as well as in Table 1. We included many disease-related and psychosocial factors in our model, including tumor site, HPV status, work status, education, and living situation to see if they would remain in the model but the model seen in the manuscript was the best model found. Some demographic and medical variables remained in our model while others were non-significant. We have included an explanation in the limitation section on lines 477-483.

2. Minor: Some style modifications are needed, e.g. the sub-headings in the methods section (page 5).
   1. Thank you for bringing this to our attention. We have reviewed the sub-headings in the methods section.

Author Response File: Author Response.docx

Reviewer 3 Report

This study investigates body image concerns among head and neck cancer patients from start of treatment up to 1 year after treatment. It is an interesting study and important topic. I only  have a few remarks:

• In the methods section it is stated that medical variables such a tumor type, HPV status and treatments received were collected from medical files. This study might benefit from a description of the study population with regard to tumor type, HPV status, and treatment.
• Despite figure 1, it is not entirely clear which variables were investigated in relation to body image concerns. Is it correct that tumor site, HPV status, work status, education, and living situation were not investigated in relation to body image concerns (not part of Figure 1)? Why not?
• Drop-out over time is quite high (i.e. (218-117)/218=46%). Mixed model analyses were used, which perform quite well when there is missing data. Nevertheless, this study might benefit from more detailed information on reasons for drop-out. Based on the descriptives on disfigurement presented in Table 1 (disfigurement was, if I understand correctly, carried forward into 6 and 12 months) it seems that especially patients with disfigurement dropped out. As disfigurement is related to body image concerns, the prevalence of body image concerns at follow-up may be underestimated.
• It is stated that body image concerns were highest at 3 months after diagnosis. Is it correct that some patients were still undergoing treatment at this timepoint?
• Related to the question above, it is stated in the methods section that: “Observer-rated disfigurement was also measured at baseline (i.e., upon cancer diagnosis/ pre-treatment) and after HNC treatments.”. In addition it is stated that: “Disfigurement rating was only assessed at baseline and 3 months, with the 3-month timepoint carried forward into 6 and 12 months.” These two sentences are slightly different, which one is correct?
• Please take a look at the following sentence in the discussion section: “Furthermore, the findings regarding the individual trajectory of body image concerns can help inform the effective timing of delivery has based on our results, body image concerns are the highest at 3-months and, therefore between 3-months and 6-months post-treatment is when patients will need the most support and therapy.”

Author Response

This study investigates body image concerns among head and neck cancer patients from the start of treatment up to 1 year after treatment. It is an interesting study and an important topic. I only have a few remarks:

1. In the methods section, it is stated that medical variables such as tumour type, HPV status and treatments received were collected from medical files. This study might benefit from a description of the study population with regard to tumour type, HPV status, and treatment.

• • We have added a more detailed description of the study population with regard to tumour type, HPV status, and treatment along with other details in lines 255-264 and in Table 1 on page 8.

 

2.  Despite figure 1, it is not entirely clear which variables were investigated in relation to body image concerns. Is it correct that tumour site, HPV status, work status, education, and living situation were not investigated in relation to body image concerns (not part of Figure 1)? Why not?

• We did investigate tumour site, HPV status, work status, education, and living situation in relation to body image concerns to see if they would remain in the model but only a few variables did and the model seen in the manuscript was the best model found. We have detailed the variables entered in our model on page 4 and have added a section in the limitation in response to this comment on lines 477-483.

 

3. Drop-out over time is quite high (i.e. (218-117)/218=46%). Mixed model analyses were used, which perform quite well when there is missing data. Nevertheless, this study might benefit from more detailed information on the reasons for drop-out. Based on the descriptives on disfigurement presented in Table 1 (disfigurement was, if I understand correctly, carried forward into 6 and 12 months) it seems that especially patients with disfigurement dropped out. As disfigurement is related to body image concerns, the prevalence of body image concerns at follow-up may be underestimated.

• • We have added more detailed information regarding reasons for participant drop-out and acknowledged limitations to the generalizability of results in the limitation section of the paper on page 18.

 

4. It is stated that body image concerns were highest at 3 months after diagnosis. Is it correct that some patients were still undergoing treatment at this time point?

• • Almost all patients had completed treatments at 3-months post-diagnosis, which we now specify on page 7.

 

5. Related to the question above, it is stated in the methods section that: “Observer-rated disfigurement was also measured at baseline (i.e., upon cancer diagnosis/ pre-treatment) and after HNC treatments.”. In addition, it is stated that: “Disfigurement rating was only assessed at baseline and 3 months, with the 3-month timepoint carried forward into 6 and 12 months.” These two sentences are slightly different, which one is correct?

• • We have made sure to keep the description of the use of the disfigurement scale consistent throughout the article to reflect the timepoints of baseline and 3-month follow-up. Please see pages 5 and 6.

 

6. Please take a look at the following sentence in the discussion section: “Furthermore, the findings regarding the individual trajectory of body image concerns can help inform the effective timing of delivery has based on our results, body image concerns are the highest at 3-months and, therefore between 3-months and 6-months post-treatment is when patients will need the most support and therapy.”

• • We have revised the sentence to make it intelligible.

 

We wish to thank the reviewers for their thoughtful comments.

Author Response File: Author Response.docx