Characterizing the Exercise Behaviour, Preferences, Barriers, and Facilitators of Cancer Survivors in a Rural Canadian Community: A Cross-Sectional Survey

Round 1

Reviewer 1 Report

The authors adequately addressed previous concerns and no further revisions are requested at this time.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

General comments

This cross-sectional needs assessment survey study attempted to assess the factors that impact on the exercise behaviour of select cancer survivors. This reviewer is questioning why this rural community and hospital were selected.  It would appear from the number of questionnaires distributed to the respondents (n=449), that this catchment area has a relatively small base of patients.  Knowing that the response rate would likely fall between 10-15% (in this case 16%), the number of respondents expected would be relatively low.  As predicted, only 72 of the community members responded making this cross-sectional survey sample size very small. Small samples sizes irrevocably lead to major problems and issues with interpretation of the statistical outcomes.  Most notably, the authors cannot have much confidence in their statistical findings, irrespective of the outcome of each of the tests used, including the Fischer’s Exact test that is typically used when sample sizes are small. Thus, in planning this study, this reviewer is somewhat puzzled as to why the authors decided to choose to evaluate such a small rural community.  This is a major concern. A second concern relates to the respondents to the survey.  A large majority of respondents (49%; n=35) were breast cancer survivors, leaving 51% (n=37) of the responses from at least eight other cancer survivor groups.  As the authors have pointed out, previous studies of this nature have already addressed a breast cancer population. Unfortunately, this study cannot glean any major significant outcomes as they relate to other cancer survivors because as previously stated, the sample is too small. In summary, as the major purpose of this study was to contribute to the development and implementation of behaviour strategies in a rural population, the authors cannot state definitively or make assertions about other rural communities because of the small sample size.

Author Response

Reviewer 1

 

“This cross-sectional needs assessment survey study attempted to assess the factors that impact on the exercise behaviour of select cancer survivors. This reviewer is questioning why this rural community and hospital were selected.  It would appear from the number of questionnaires distributed to the respondents (n=449), that this catchment area has a relatively small base of patients.  Knowing that the response rate would likely fall between 10-15% (in this case 16%), the number of respondents expected would be relatively low.  As predicted, only 72 of the community members responded making this cross-sectional survey sample size very small. Small samples sizes irrevocably lead to major problems and issues with interpretation of the statistical outcomes.  Most notably, the authors cannot have much confidence in their statistical findings, irrespective of the outcome of each of the tests used, including the Fischer’s Exact test that is typically used when sample sizes are small. Thus, in planning this study, this reviewer is somewhat puzzled as to why the authors decided to choose to evaluate such a small rural community.  This is a major concern.”

 

• Response: Huntsville is the largest town in the Muskoka region of Ontario Canada. It is approximately 215km north of Toronto. The catchment area for Huntsville District Memorial Hospital includes a geographical area west towards Parry Sound (not including Parry Sound), east to Algonquin Park, south to Baysville, and north to South River, Ontario. It is representative of other Canadian rural communities in density and has a large geographical area. Currently the area has no exercise programs for cancer survivors. Clinicians and physicians within the area were interested in implementing an exercise program for cancer survivors but acknowledged that background information was first required to determine what survivors in the area needed. Needs assessments are used to determine the services required prior to implementation in order to make them more usable / effective for the population. We are not suggesting that the results of this Needs Assessment should be applied to all rural settings and have acknowledged this point in our limitations section: “While some rural communities may have similar characteristics, there may be important differences in the resources (natural, physical, professional) and support services across regions. Collectively, these factors may limit the generalizability of these results”. We have made sure this information is clear in the paper and that the need to do additional needs assessment based on region is clear throughout the manuscript.

 

“A second concern relates to the respondents to the survey.  A large majority of respondents (49%; n=35) were breast cancer survivors, leaving 51% (n=37) of the responses from at least eight other cancer survivor groups.  As the authors have pointed out, previous studies of this nature have already addressed a breast cancer population. Unfortunately, this study cannot glean any major significant outcomes as they relate to other cancer survivors because as previously stated, the sample is too small.”

 

• Response: Thank you for this comment. We agree the large majority of respondents being breast cancer survivors is a limitation (and have listed this within our Limitations section; please see manuscript page 2 [line 336] to page 3 [line 346]). However, this is the first needs assessment study to also include those living in a rural community with other forms of cancer. Knowing the sample was highly skewed to breast cancer survivors we did perform a detailed sub-analysis of the data by cancer type (BC versus other cancers) and of the 27 variables assessed looking at barriers, motivators, and facilitators, only 1 (‘cost’ as a barrier to exercise during treatment) was found to have a significant difference between groups. We appreciate that this does not give us specific information about the other types of cancer individually and have added this in as an area for future research in the conclusion (“uture research is required to expand upon our work in non-breast cancer rural survivors.”). Nonetheless, we feel this is an important first step in implementing exercise programming in this type of rural community. It is also important to share/publish this data so others can build off and compare findings of other rural communities.

 

“In summary, as the major purpose of this study was to contribute to the development and implementation of behaviour strategies in a rural population, the authors cannot state definitively or make assertions about other rural communities because of the small sample size.”

 

• We agree and have alluded to this repeatedly in our limitations and conclusion sections – we have reviewed and made sure this is clearly highlighted. See Limitations and Conclusion sections (example: “Other limiting factors include the majority of respondents being Caucasian, diagnosed with breast cancer, and living within a specific rural region of the province. While some rural communities may have similar characteristics, there may be important differences in the resources (natural, physical, professional) and support services across regions. Collectively, these factors may limit the generalizability of these results to other ethnicities, cancer types, and rural communities.”).

Reviewer 2 Report

Overall this is a timely, important review describing the exercise barriers and participation among rural cancer survivors.

 

Abstract:

Please describe how physical activity guidelines were captured within the survey.

 

Introduction:

Given the focus on rural cancer survivors, describing this population in detail to support the rationale for the purpose of the review, and earlier on in the introduction, is necessary.

 

2.3 Research instrument

Whether a validated survey was used is not clear.

Including the survey in an appendix or online supplement is critical in the rigor of this review. Questions asked to capture physical activity require additional detail.

 

Discussion:

Given the low response rate, are additional efforts available to enhance participation to better reflect the rural population for future studies on this topic.

Author Response

Reviewer 2

“Overall this is a timely, important review describing the exercise barriers and participation among rural cancer survivors.

 Abstract:

Please describe how physical activity guidelines were captured within the survey.”

• Questions were asked related to volume of aerobic and resistance exercise (minutes of moderate-to vigorous intensity aerobic and resistance exercise) and compared to current exercise guidelines for cancer survivors (based on American College of Sport Medicine guidelines). We have added in a few lines in abstract to try and clarify this. See tracked changes, Page 1 – “Abstract”.

“Introduction:

Given the focus on rural cancer survivors, describing this population in detail to support the rationale for the purpose of the review, and earlier on in the introduction, is necessary.”

• This is a great point. We have added in a few lines at the start of the manuscript and feel this helps contextualize the paper. See introduction, paragraph 1 (“Rural populations are defined as areas that have low population density and are dispersed over a large area. Individuals living in rural and remote communities are less active than their urban counterparts. Survivors of cancer living in rural and remote communities are considered ‘hard-to-reach’ due to their geographical location, engage less in cancer survivorship programming, and have low overall health status, well-being and quality of life.”).

 “2.3 Research instrument

Whether a validated survey was used is not clear.”

• This specific Needs Assessment survey had not been used previously, however; parts of the survey included a validated outcome measure (Godin Leisure Time Exercise Questionnaire) and used previously administered questions from other survey-based studies. While there is no validated survey to determine need, use of questions similar to those previously administered allows for comparison between urban based needs assessments. See Section 2.3 and associated references).
• References:
  • Fernandez S, Franklin J, Amlani N, et al. Physical activity and cancer: A cross-sectional study on the barriers and facilitators to exercise during cancer treatment. Can Oncol Nurs J. 42(1):37-72,
  • Smith-Turchyn J, Vani MF, Sabiston C. Understanding how to reach the hard to reach in cancer rehabilitation. Global J Nursing. 2020; 3:18.
  • Adams SC, Petrella A, Sabiston CM, et al. Preferences for exercise and physical activity support in adolescent and young adult cancer survivors: a cross-sectional survey. Support Care Cancer. 2021
  • Amireault, S., Godin, G., Lacombe, J., & Sabiston, C. M. (2015a). The use of the Godin-Shephard leisure-time physical activity questionnaire in oncology research: A systematic review. Bmc Medical Research Methodology, 15, 60. doi:10.1186/s12874-015-0045-7.

“Including the survey in an appendix or online supplement is critical in the rigor of this review.”

• We completely agree; this has been added as an appendix. (See Appendix 1)

“Questions asked to capture physical activity require additional detail.”

• Added this in briefly in section 2.3 (“Part 3: Exercise Needs Assessment (including exercise volume (how often and for how long do you take part in (1) strenuous, (2) moderate, and (3) mild exercise?), barriers to exercise during and after treatment, exercise motivators, and needed facilitators”) and referred to Appendix (specifically page 5-6). We used the Godin Leisure Time Exercise Questionnaire to assess exercise behaviour / volume within the Needs Assessment Survey.

“Discussion:

Given the low response rate, are additional efforts available to enhance participation to better reflect the rural population for future studies on this topic.”

• Yes, definitely. This is an important point, thank you for bringing it forward. We have added in some suggestions re: recruitment methods to enhance participation. See discussion section, 4.2 “Limitations” – where we discuss the low response rate (“Future research should expand recruitment duration and methods to enhance participation. Recruitment methods could include using web-based applications and social media platforms, as well as recruitment at physician offices and other community health and cancer support service locations.”).

Round 2

Reviewer 1 Report

Since this reviewer has an issue about the number of responses from the cross-sectional survey, can the author's provide their sample size calculation?  Evidently the number of respondents was below the sample size calculation or it would not have been mentioned as a limitation. 

Are the researchers willing to use tools like social media etc to increase their sample size in this study?

How are you going to convince hospital and health administrators that developing community-based exercise support services for all cancer survivors when the survey is underpowered?      

Author Response

Current Oncology

Manuscript ID#: 1200746

Response to Reviewers

Thank you to the reviewers for their continued support with our manuscript. We are pleased the majority of our past revisions met reviewers’ expectations.

We have made further edits, based on the most recent reviewer comments regarding sample size, the community surveyed, and the survey used, as well as email correspondence with the editor. Please let us know if you have any further comments or concerns.

 

Reviewers comments:

“Since this reviewer has an issue about the number of responses from the cross-sectional survey, can the author's provide their sample size calculation? Evidently the number of respondents was below the sample size calculation or it would not have been mentioned as a limitation.

Are the researchers willing to use tools like social media etc to increase their sample size in this study? How are you going to convince hospital and health administrators that developing community-based exercise support services for all cancer survivors when the survey is underpowered?”

 

Sample size calculation: Reviewer’s comments - “Can the authors provide a sample size calculation?” and “the study is underpowered – low response rate”.

 

Response: The stated purpose of the study was to characterize exercise behaviour, barriers, facilitators, and motivators within a sample of rural living cancer survivors. To our knowledge, with no ‘hypothesis to test’, this is no need for sample size or power calculations. This survey was sent to all eligible individuals in the community. We have acknowledged that our response rate of 16% is lower than desired; however, our response rate was comparable to or even greater than other mailbased surveys. For instance, Sinclair et al. (2012) reported a 10.5% response rate to personalized postal surveys and a 7.5% response rate to generic postal surveys. Notwithstanding the relatively better response rate for this study, we agree it is appropriate to list this as a limitation to facilitate the interpretation of our findings.

 

We have added additional contextualization around this point. See “Participants & Recruitment” (page 3, paragraph 2: highlighted “All individuals…were mailed a needs assessment survey package) and section ‘4.2: Limitations (here we contextualized findings related to other mail-based surveys and added in the Sinclair (2012) reference: “Firstly, only a small portion of individuals completed and returned the survey (16% response rate). While this response rate is consistent with similar mail-based surveys,49 the limited response rate may have led to a response bias as only individuals with an interest in this topic may have responded to this survey” and “Additionally, results of this study are specific to individuals living within a specific rural region of Ontario, Canada. While some rural communities may have similar characteristics, there may also be important differences in the resources (natural, physical, professional) and support services across regions”).

 

Sampling one rural community: Reviewer’s comments - “…your findings only reflect one rural

community”

 

Response: Needs assessments are used to determine the services required prior to implementation in order to make them more usable / effective for a specific population and are required to be completed for specific geographic communities (Community Needs Assessment. Atlanta, GA: Centers for Disease Control and Prevention (CDC), 2013.

 

Available at:

https://www.cdc.gov/globalhealth/healthprotection/fetp/training_modules/15/communityneeds_pw_final_9252013.pdf.

As we discussed, the catchment area for this rural community in Canada is quite large and we have not suggested that the results of this Needs Assessment be applied to all rural settings; we have acknowledged this point in our limitations section: “While some rural communities may have similar characteristics, there may be important differences in the resources (natural, physical, professional) and support services across regions. Collectively, these factors may limit the generalizability of these results” and conclusion section “This study provides important insights into exercise behaviour, barriers, facilitators and motivators of cancer survivors living in a rural Canadian region”. We have made sure this information is clear in the paper and that the need to do additional needs assessments based on region is highlighted. However, this needs assessment was quite important for this specific rural community prior to implementing exercise services for cancer survivors.

 

We have provided additional contextualization around this point. See update to title (“Characterizing the Exercise Behaviour, Preferences, Barriers, and Facilitators of Cancer Survivors in a Rural Canadian Community: A Cross-Sectional Survey”) and discussion (page 14, paragraph 1: The purpose of this study was to characterize exercise behaviour and determinants within a diverse sample of cancer survivors living in a single rural community in Canada to inform the development of local exercise support services). We have also made reference to a ‘single rural community’ or ‘this rural community’ or ‘a rural community’ (as oppose to ‘rural communities’) throughout so to not overstate our findings. We also ensured our conclusion had a statement related to this (“Future research is required to expand upon our work in other rural communities and in non-breast cancer rural survivors.”)

 

Survey Used: Reviewer’s comments - “a non-valid survey was used”

 

Response: We appreciate the need to use validated measurement tools in research whenever possible. However, as mentioned previously, currently no validated needs assessment survey exists which provides information on the scope of responses that was obtained in our survey. Our survey was based on previous research looking at the barriers / facilitators for individuals with cancer (ref: Adams SC, Petrella A, Sabiston CM, et al. Preferences for exercise and physical activity support in adolescent and young adult cancer survivors: a cross-sectional survey. Support Care Cancer. 2021; Smith-Turchyn J, Vani MF, Sabiston C. Understanding how to reach the hard to reach in cancer rehabilitation. Global J Nursing. 2020; 3:18) and included a validated measure (Godin Leisure Time Exercise Questionnaire; ref: Amireault S, Godin G, Lacombe J, Sabiston CM. The use of the Godin-Shephard leisure-time physical activity questionnaire in oncology research: A systematic review. Bmc Medical Research Methodology. 2015;15. doi:10.1186/s12874-015-0045-7.) to assess current physical activity level. If there was an existing validated needs assessment survey to assess the components that we wanted to assess we would have gladly used this. To this end, we feel we have been quite transparent within our submission and have provided the complete survey as an Appendix to facilitate study interpretation and replication.

 

However, we have provided additional context around this point. Please see section ‘2.3 Research Instrument’ (“As no validated needs assessment surveys currently exist on this topic, the survey used was created by three members of the study team (SA, JST, LA) informed by a previous needs assessment,36 a validated outcome measure assessing exercise volume37, and research on exercise-related barriers, facilitators, and needs for cancer survivors.25”) and Limitations section (“Finally, while we assessed physical activity behaviour using a validated questionnaire, the remaining components of the needs assessment have not previously been validated”).

Author Response File: Author Response.pdf