The Role of Physical Activity in Moderating Psychopathological Symptoms and Quality of Life Among Adult Cancer Survivors: A Cross-Sectional Study

1. Summary

Thank you very much for taking the time to review this manuscript. Please find detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: The manuscript and/or abstract should clearly present this work as a cross sectional study, one possible title would be "The Role of Physical Activity in Moderating Psychopathological Symptoms and Quality of Life Among Adult Cancer Survivors: A Cross-Sectional Study".

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, we have changed the manuscript title for "The Role of Physical Activity in Moderating Psychopathological Symptoms and Quality of Life Among Adult Cancer Survivors: A Cross-Sectional Study" [page number 1 and line number 2 to 4].

Comments 2: Following the same order of ideas, Introduction could be further elaborated to include a good explanation on what the link between the physical exercise and mental health, also please include why is it expected to have an impact on cancer survivors, and what is known on the differences between rural and urban survivors, all these aspects are shallowly mentioned in the introduction but can be presented in a clearer, evidence-based way.

Response 2: Agree. We have, accordingly, modified the introduction to emphasize the points mentioned more clearly. The relationship between physical activity and mental health in cancer survivors has been more clearly explained, including the physiological and psychosocial factors involved [page 2, lines 69 to 86]. We have also included a justification for the relevance of this impact on cancer survivors, highlighting the high prevalence of depressive anxiety, and fatigue symptoms, as well as evidence of how physical activity can reduce these effects and improve quality of life [page 2, lines 53 to 60]. In addition, the introduction was revised to include references that highlight barriers that survivors in rural settings face, particularly in accessing healthcare and opportunities to practice physical activity [page 2, lines 92 to 106]. Given this, we believe that the new version of the introduction is more clearly substantiated, and evidence based.

Comments 3: In methods, Outcomes and exposures are defined; but no clear statement of potential confounders or effect modifiers.

Response 3: We recognize the importance of considering potential confounders such as age, sex, socioeconomic status, and type of cancer, as well as possible effect modifiers in the relationship between physical activity and the variables of interest. However, these tests were performed as part of the validation study of the instrument used to assess physical activity (GLTEQ), which will be covered in another paper. In the present article, only the results related to the application of the instrument are presented, without including specific analyses of confounding and effect modification.

Comments 4: There is also a lack of information of how sampling was performed, how sample size was calculated and met. Limitations mention sampling bias, but no detailed strategy to mitigate it. Attrition flow should be presented.

Response 4: The sample was performed by convenience, including all eligible patients (referred by medical doctors) undergoing medical follow-up in the oncology consultation and participants in the Mama Move program, who agreed to participate in the study. Data collection took place among patients available during the waiting period prior to consultation and/or treatment. In order to minimize potential selection biases, previously defined inclusion and exclusion criteria were applied: individuals aged ≥18 years, living in the Cova da Beira region (interior of Portugal), with a cancer diagnosis (any anatomical location or tumor type), undergoing treatment or post-treatment, were included. Only patients in palliative care were excluded. No formal sample size calculation was performed, all eligible patients during the data collection period were included. Only one participant was excluded for not responding to most items in the protocol. Despite the convenience sample approach, efforts were made to ensure the greatest possible representativeness of the cancer population in the region [page 4, lines 162 to 184].

Comments 5: Regression and correlations described; no control for confounders or missing data handling are reported. Representativity and handling of potential confouders are essential to ensure or estimating external validity, therefore, it is important to clearly state.

Response 5: We appreciate the observation. We recognize that controlling for confounding variables is an important aspect to ensure more reliable conclusions. However, in the present study, we choose to focus on the variables under investigation (physical activity, depression, anxiety, and quality of life) in order to explore the relationship between them more directly. Therefore, other potential confounding variables were not included, which could be considered a limitation and explored in future research [page 9, lines 366 to 368 and lines 381 to 383]. In terms of missing handling data, only one participant was excluded for not responding to more than half of the protocol. For the rest of the sample, the proportion of missing data was minimal and didn´t compromise the analyses performed.

Our sample was not selected based on representativeness criteria for the general population, therefore we recognize that the generalization of results, as already mentioned in the limitations, should be done with caution. This point will also be emphasized in the manuscript in order to clarify external validity limitations [page 9, lines 371 to 376].

Comments 6: From an ethical point of view, what intervention/treatment was offered to patients in depression and/or anxiety?

Response 6: According to Appendix A3, participants were asked if they were being monitored by a psychology or psychiatry team, with 21.8% reporting “yes”. All participants were informed of the possibility of requesting psychological support at the hospital or at institutions that provide psychological support to cancer survivors. All procedures were previously approved by the Ethics Committee (protocol code 17/2024), guaranteeing that ethical principles, such as nonmaleficence and respect for autonomy, were respected.

Comments 7: How can you explain a better quality of life in those who exercise? how can you know if not the patients that are in better mental state are more prone to exercise? in a cross-sectional study variables are associated but it is difficult to define cause-effect.

Response 7: Indeed, the results of cross-sectional studies only allow us to identify associations between the variables under study, without establishing causal relationships. The hypothesis that cancer survivors with better psychological state and higher quality of life are more likely to engage in physical activity can’t be ruled out [page 9, lines 360 to 366]. This limitation is acknowledged, and it has been pointed out that longitudinal/intervention studies will be necessary to clarify this relationship and understand how exercise contributes to improvements in the mental health and quality of life of cancer survivors [page 9, lines 385 to 388].

Comments 8: Did you find differences between cancer type or stage in physical activity and or depression/anxiety?

Response 8: Good observation. As mentioned above, this study analyzed differences in physical activity scores (GLTEQ Total, without categories) based on cancer type, as well as the variables gender, age range, marital status, employment status, participation in physical exercise programs, and recommendations by health professionals. This analysis was part of the preliminary validation process of the instrument GLTEQ. Although no statistically significant differences were found (p = 0.063), it was observed that the mean for breast cancer (M = 26.62; SD = 15.38) was higher than the mean for the other types of cancer. No differences were calculated between cancer type and depression/anxiety variables.

1. Summary

Thank you very much for taking the time to review this manuscript. Please find detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: Abstract. The authors are encouraged to add brief description of study design, statistical tests used in the analysis and effect size of significant associations detected with 95% CI or p values.

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, in the revised version of the abstract, we added a concise description of the study design [abstract, page 1, line 19] and specified the statistical tests applied in the analysis [abstract, page 1, lines 25 to 27]. It was also reported the effect size of significant associations detected 95% confidence intervals and p-values, as recommended [abstract, page 1, lines 29 to 31].

Comments 2:

Line 40: Please add context to this statistics. Is this a global context?

Response 2: In the previous version of the manuscript, there is no statistics reported in the line referred to (line 40 – “cancer-related mortality is decreasing, and both the prognosis and course of the disease”). However, if the comment concerns the statistics presented at the end of paragraph 1 (currently line 45 and 46 in the revised version), we confirm that it refers to a global context, as reported by the International Agency for Research on Cancer (“Global cancer burden growing, amidst mounting need for services”). We have clarified and added context to avoid ambiguity [page 1, line 45 and 47].  

Comments 3:

Line 53: The authors are highly encouraged to add more effect size statistics about the effect of exercise on quality of life and mental health among cancer patients.

Response 3: We appreciate for pointing this out. The introduction (including line 53 in the previous version of the manuscript) has been revised in a clear and evidence-based way to elucidate the connection between physical activity and mental health among cancer survivors [page 2, lines 53 to 60]. In the revised version, we emphasize the reasons why physical is anticipated to significantly benefit cancer survivors [page number 2, lines 69 to 86]. Additionally, we included information on known variations between rural and urban survivors [page 3, lines 94 to 106]. To support the evidence of the influence of exercise on cancer survivors’ quality of life and mental health, additional information in evidence-based way has been included [page 2, lines 61 to 63, lines 71 to 74, lines 85 to 86].  

Comments 4: Materials and methods. Reporting demographic characteristics in the methodology is unusual. Please move all descriptive sample statistics to the results.

Response 4: Thank you for pointing this out. We agree with this comment. Therefore, we have moved all descriptive sample statistics to the results [page 5, lines 206 to 232].

Comments 5:

Line 108: the authors made the following statement

 ‘The sociodemographic and clinical questionnaire included several sections, namely information about the sample such as sociodemographic data, date of cancer diagnosis, type and stage of the disease, treatments conducted, to be carried out, and other relevant types of data.’

Please declare that are the other relevant types of data. The methodology should be clear enough to allow replication of the study.

Response 5: Thank you for pointing this out. We have revised the description of the sociodemographic and clinical questionnaire to include other relevant types of data in a clear way, allowing the replication of the study [page 3, line 116to 121].

Comments 6:

Line 111: Please add more details on how the physical activity questionnaire was developed and tested with regards to validity and reliability.

Response 6: The questionnaire designed to gather information on physical activity characteristics was created specifically for this research, aiming to obtain descriptive data about the sample. It comprised both open-ended and closed-ended questions and was not derived from an established validated tool. As a result, we didn’t conduct any validity and reliability tests for these questions. We would like to highlight that the only validated instrument used for evaluating the main study variables and associations was GLTEQ.

Comments 7: Please add more details on how the sample size estimation was performed and whether it was sufficient to detect the tested associations.

Response 7: The sample was conducted using a convenience method, including all eligible patients (referred by medical professionals) who consented to join the investigation. Data collection occurred among patients present during the waiting time before their consultation and/or treatment. To reduce possible selection biases, specific inclusion and exclusion criteria were established. No formal calculation and estimation for sample size was conducted [page 4, line 162 to 184]. Instead, all eligible patients during the data collection timeframe were included. One participant was excluded for failing to answer most questions in the protocol. Despite the relatively small sample size (n = 55), we were able to detect significant associations between the variables. This has been described in the limitation of the present study [page 9, lines 371 to 376].

Comments 8: Statistical analysis. Distribution assessment is needed to display whether the data are normally distributed or skewed. This is important to illustrate whether the used statistical tests are suitable or not.

Response 8: Thank you for pointing this out. We agree with this comment. Before performing any statistical tests, we assessed the data distribution using a normality test (Kolmogorov-Smirnov test). The results indicated that the data did not conform to a normal distribution. Therefore, nonparametric tests were applied to ensure the validity of the statistical analyses. In this revised manuscript, we have included this information [page 4, lines 186 to 189].  

Comments 9:

Line 176: It is advised to add more details about the dependent and independent variables within the regression analysis.

Response 9: Thank you for your observation. We added more details about the dependent and independent variables within the regression analysis [page 5, lines 196 and 202].

Comments 10: Results. The first section of the results should be descriptive statistics of the recruited sample, which is already referred to in a previous comment. It is also advised to add a new table within the main text to describe the demographic characteristics of the sample.

Response 10: In accordance with response 4, all sample descriptive statistics have been transferred to the Results section. The table containing the sociodemographic characteristics of the sample remains in the appendix due to its length, with citations in the main text (e.g., “Appendix A.1”) to promote a cleaner and clearer reading of the manuscript.

Comments 11:

Table one: as the data presented in the table are mostly frequencies and percentages, using min-max and SD is not informative. Please revise.

Response 11: To enhance the clarity and informativeness of Table 1 in the original manuscript, it was revised and separated into two different tables. Table 1 presents a descriptive statistical analysis of the weekly frequency of physical activity based on the GLTEQ [page 6, line 240], while Table 2 offers a descriptive statistical analysis of the GLTEQ total classifications [page 6, line 248]. In line with the modifications, the text content referring to the tables was revised [page 5, lines 234 to 239 and page 6, lines 243 to 247].

Comments 12:

Table two: all abbreviations should be spelled out beneath the table.

Response 12: Thank you for your observation. Due to the changes described in Response 11, Table 2 has been renumbered as Table 3. The table has been revised to include the GLTEQ total as well as the 95% confidence intervals [page 6, line 259]. All abbreviations are now full spelled out below the table, and its context has also been revised [page 7, lines 251 to 258 and lines 231 to 264].

Comments 13:

Table three: the correlation matrix is not clear. What is meant by the numbers in the first row (1, 2, 3)?

Response 13: I appreciate you highlighting this ambiguity. In the revised manuscript, we have modified Table 4 to clearly identify the variables (physical activity, quality of life, depression and anxiety) rather than using numerical labels [page 7, line 283].

Comments 14:

Line 221: The rationalization mentioned in this paragraph should be detailed in the methodology section and to explain why it was conducted.

Response 14: The rationalization mentioned has been detailed in the methodology section [page 5, lines 196 to 202].

Comments 15: Discussion. Interpretation of the interactions is not clear. Physical activity does not seem to influence quality of the life in the sample. However, interaction between depression and physical activity reduced quality of life, but interaction between anxiety and physical activity increases quality of life. Please explain why physical activity does not influence quality of life unless it is interacting with depression and anxiety and why it has beneficial effect with anxiety but not with depression.

Response 15: Thank you for your observation. There are several potential explanations for these patterns of interaction. Firstly, anxiety symptoms may serve as a catalyst/motivator for participating in physical activity, driven by worries about health decline or disease progression, which could clarify the association between anxiety and physical activity with enhanced quality of life. Secondly, the absence of a direct link between physical activity and quality of life might be related to methodological factors within our sample, such as the relatively limited range of physical activity levels potentially diminishing the statistical power to identify independent effects. All together, these factors imply that the impact of physical activity on quality of life is complex and may be more easily detected in interaction with particular psychological factors.

1. Summary

Thank you very much for taking the time to review this manuscript. Please find detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: Regarding the abstract, I believe it has a clear structure and well-defined objectives, focused on evaluating the relationship between physical activity, symptoms of depression and anxiety, and quality of life in cancer survivors. I see that the variables and measurement instruments are correctly identified, but the methodology lacks important details such as the type of study design, inclusion and exclusion criteria, and the sample selection procedure. Furthermore, it is not specified whether sociodemographic or clinical variables that could influence the results were controlled for, which limits the robustness of the conclusions. Regarding the results, I note that significant associations and relevant interactions are described, although no specific statistical data are presented to assess the magnitude of the findings. The recommendation to promote supervised physical activity in the comprehensive care of cancer survivors is relevant, but I believe it would be strengthened if it were supported by figures from the study and a discussion of its limitations. Overall, the abstract is clear and coherent, but I believe it could be strengthened with greater methodological precision and quantitative data to increase its scientific rigour.

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, we have clarified the study’s design and briefly outlined the inclusion and exclusion criteria [page 1, lines 19 and 20], along with the statistical tests employed [page 1, lines 25 to 27]. However, due to the abstract’s word limit, we didn’t specify the sociodemographic and clinical variables that were not accounted for, nor did we address other important limitations, such as the reduced generalizability of the findings. Additionally, we incorporated quantitative data that allow us to illustrate the strength of the associations identified with clarity [page 1, line 29 to 31].

Comment 2: However, I believe that the introduction could be strengthened with a more concise summary of the literature to avoid redundancy, as some benefits of physical activity are repeated in different paragraphs. It would also be valuable to clarify the research gap more clearly, detailing the limitations of previous studies in Portugal, especially regarding sample characteristics and methodological aspects. While the focus on rural and inland populations is relevant and justifies the study, it would be more convincing if supported by statistics or data illustrating inequality in access to exercise programmes.

Response 2: Thank you for your valuable feedback. We agree with your observations regarding the introduction and, therefore, revised to minimize redundancy and present a more concise overview of the literature on the benefits of physical activity. We clarified statistics concerning cancer incidence and prognosis to provide context and avoid any ambiguity [page 2, lines 45 to 46], as well as on the involvement in physical activity among adults’ cancer survivors [page 2, lines 61 to 63].

The link between physical activity and mental health among cancer survivors has been elucidated, incorporating both physiological and psychosocial aspects [page 2, lines 69 to 88], with an emphasis on the importance of the effects, particularly given the high occurrence of symptoms related to depression, anxiety and fatigue. Additionally, we have provided evidence indicating that physical activity can alleviate these symptoms and improve quality of life [page 2, lines 69 to 86].

We have more clearly addressed the research gap by outlining the limitations of prior studies conducted in Portugal, especially concerning sample characteristics and methodological issues [page 3, lines 98 to 103]. The emphasis on rural and inland populations has been reinforced by incorporating references that illustrate disparities in access to healthcare and physical exercise opportunities [page 2, lines 92 to 106].

Comment 3: With regard to methodology, I believe that the section clearly describes the quantitative approach adopted and the cross-sectional, descriptive and correlational design, which is consistent with the objectives set. The sample is characterised in detail in terms of sociodemographic, clinical and physical activity characteristics, which allows the results to be properly contextualised. Likewise, the inclusion and exclusion criteria are specified, and a comprehensive description of the instruments used is provided, including references to their validity and reliability, as well as the internal consistency coefficients obtained in this study. The data collection procedure is clearly outlined, with mention of ethical approval, informed consent and compliance with national and international regulations on research involving human subjects. However, I believe that there are elements that could be strengthened to increase methodological rigour. Firstly, the use of non-probability convenience sampling should be accompanied by a discussion of the limitations this implies for the generalisation of the results. It would also be advisable to specify more precisely the conditions under which the questionnaires were administered (e.g., whether they were self-administered or administered with researcher support) and to indicate whether adjustments were made for possible confounding variables in the statistical analyses. Although the use of descriptive statistics, non-parametric correlations and multiple linear regression is detailed, it would be useful to indicate whether the regression assumptions were verified and whether procedures were applied to control for error due to multiple comparisons

Responses 3: Thank you for your feedback. Regarding the applications for the questionnaires, we clarify that they were self-administered, with support from the researchers provided only when necessary [page 4, lines 172 to 175]. As for the statistical analysis, we did not account for additional confounders, this will be addressed as a limitation in the discussion section [page 10, lines 366 to 368 and 381 to 383]. However, some sociodemographic and clinical variables were included in the validation of the GLTEQ, without statistically significant differences being found. We outline the method used to assess the normality of the sample distribution, which is essential for selecting statistical tests [page 4, lines 186 to 189]. Nevertheless, we recognize that no specific procedures were applied to control the error associated with multiple comparisons.

Comment 4: With regard to the results, I believe that this section clearly presents both the descriptive analysis and the statistical associations between the variables, allowing the logic of the study to be followed. The initial characterisation of physical activity levels (GLTEQ) and mean scores for depression (PHQ-9), anxiety (GAD-7) and quality of life (EORTC QLQ-C30) is detailed, which facilitates interpretation of the sample profile. In addition, bivariate correlations are shown and r and p values are included, providing transparency on the strength and significance of the relationships. The use of multiple linear regression to assess the moderating role of physical activity on the relationship between psychopathological symptoms and quality of life is relevant and well justified, and standardised coefficients, standard errors and confidence intervals are presented. However, I believe that there are aspects that could be strengthened. First, the descriptive analysis is comprehensive but could be summarised to avoid information overload in tables and text. Furthermore, although non-significant associations between physical activity and psychological variables are indicated, possible causes for this lack of significance are not discussed, which could be relevant for interpretation. The regression explains 33.6% of the variance in quality of life, but I see that it is not specified whether the model assumptions were tested or whether possible collinearities were explored, especially given the high coefficient between depression and anxiety. Also, I believe it would be advisable to specify whether sociodemographic or clinical variables that could influence the results were controlled for.

Responses 4: We sincerely appreciate the reviewer’s valuable feedback. In response to the comments: (1) the tables have been simplified for better clarity and to prevent information overload, while still presenting the essential information; (2) the possible causes for the lack of significance between physical activity and psychological variables were explained in the discussion section [page 8, lines 323 to 325]; (3) the normality analysis of the sample was verified, however, we did not asses multicollinearity among the variables; (4) no sociodemographic or clinical control variables were incorporated into the regression model, which we recognize as a limitation [page 9, lines 366 to 368]. We believe these revisions improve the transparency and interpretation of the results.

Comment 5: Regarding the discussion, I believe that this section successfully connects the results with previous studies on physical activity, psychopathological symptoms, and quality of life in cancer survivors. I see that the authors clearly highlight the negative impact of anxiety on quality of life and how physical activity can moderate the effect of depression, which I find very relevant from a clinical point of view. I also think it is positive that they acknowledge the limitations of the study, such as the cross-sectional design, the small size and non-probability sample, and the use of self-reported measures for physical activity. However, I believe that there are aspects that could be reinforced to give greater depth to the discussion. For example, I would like to see further exploration of the possible mechanisms by which physical activity influences psychological symptoms and quality of life, as this would help to better understand the findings. I also note that you mention the lack of significant direct associations between physical activity and psychopathological symptoms, but you do not discuss possible reasons for this, such as the small sample size or measurement limitations, and I think it would be useful to consider this.

Response 5: We thank you for your thoughtful feedback. We have revised the discussion to further explore potential mechanisms by which physical activity might affect psychological symptoms and overall quality of life, which helps to better understand the findings [page 8, lines 330to 339]. Additionally, we included a discussion regarding the lack of significant direct associations between physical activity and psychopathological symptoms. We highlight possible reasons for this, such as the small sample size and the narrow range of physical activity levels observed [page 8, lines 323 to 325]. These revisions are intended to provide more depth and clarity in interpretation of the results of the manuscript.

Comment 6: Furthermore, I think it is important to mention that it is unclear whether sociodemographic or clinical variables that could affect the results were controlled for, which I believe is key to interpreting the data correctly. Finally, the strong correlation between depression and anxiety could indicate problems of multicollinearity, but this point is not addressed in the discussion, and I think it would be good to mention it for a more rigorous analysis.

Response 6: Thank you for pointing this out. We understand how crucial it is to control sociodemographic and clinical variables that may affect the findings. We included this limitation in the discussion section for better clarity [page 9, lines 366 to 368 and lines 381 to 383]. Regarding the strong connection between depression and anxiety, we recognize that multicollinearity may be a factor to consider. Although we did not perform this analysis, we acknowledge this as a limitation. Therefore, we addressed this issue in the discussion to emphasize its potential effects on the interpretation of the results [page 9, lines 383 to 385].

Comment 7: In my opinion that the conclusions of the study adequately summarise the main findings. However, I believe they could be improved with greater precision regarding the practical implications for clinical care. For example, it would be valuable to specify what type of recommendations or interventions healthcare professionals should prioritise in order to promote physical activity in an effective and personalised manner.

Response 7: Thank you for your valuable feedback. We agree with your observations regarding the Conclusions and, therefore, we revised to include specific recommendations that healthcare professionals should prioritize to promote physical activity in an effective and personalized way [page 10, lines 411to 417].

Comment 8: Overall, I value the effort put into this study and the relevance of the topic addressed. I believe that the results provide valuable information on the importance of physical activity in the quality of life of cancer survivors. However, I suggest that you review and clarify some methodological aspects, such as the control of sociodemographic and clinical variables and the verification of statistical assumptions, in order to strengthen the rigour of your analysis. In addition, it would be very enriching to deepen the discussion on the possible mechanisms linking physical activity to psychopathological symptoms and to address the limitations encountered in more detail. Finally, I encourage you to further specify the practical recommendations derived from your findings so that they can be of direct use in clinical care.

Response 8: Thank you for your valuable and constructive feedback. All recommended changes have been implemented: we clarified methodological aspects, statistical assumptions verified, the discussion was expanded to address potential mechanisms and limitations, and the practical recommendations were specified for clinical applicability.