Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript is an umbrella review of 23 systematic reviews/meta-analyses covering 459 primary-study occurrences, 333 unique studies, and 44,560 participants across reviews reporting sample size. Overall, I believe this is a timely, clinically relevant, and methodologically strong review. However, I have some comments as follows:

1. I recommend moderating statements that digital interventions “consistently improved” outcomes. Since 83% of reviews included were rated low confidence by AMSTAR-2, the Abstract and Conclusion should better reflect this limitation. A more appropriate wording would be: “Digital interventions generally showed favorable effects on quality of life, physical activity, and psychosocial outcomes, although confidence in many reviews was low and implementation evidence remains limited.”

2. The eligibility criteria refer to survivors who completed primary treatment, yet several reviews included participants during treatment or across the care continuum. Please clarify how “survivor” was defined, how mixed treatment-phase populations were handled, and whether the conclusions apply mainly to post-treatment survivorship or more broadly to supportive cancer care.

3. The Discussion about AMSTAR-2 and GRADE should better explain how methodological limitations affect confidence in the findings. I suggest distinguishing between the generally favorable direction of effect, the often-limited certainty of evidence, and the still insufficient implementation readiness.

4. Table 1 is informative but overly dense. I recommend retaining only key items in the main manuscript, such as author/year, modality, number of studies, primary outcomes, direction of effect, and AMSTAR-2 confidence, while moving detailed information to the Supplementary Material.

5. The Conclusion should better distinguish between intervention types. mHealth and web-based interventions have the broadest evidence base, telephone interventions may support digital equity, wearables are most relevant to physical activity, VR remains promising but less certain, and evidence for AI chatbots is still limited.

6. Although overall overlap was slight, some mHealth reviews shared substantial primary evidence. I recommend briefly noting that this overlap may partly contribute to the apparent consistency of findings.

7. The four-tier integration ladder is a useful contribution. I suggest briefly incorporating practical considerations such as referral pathways, clinician responsibility, electronic medical record integration, privacy, workflow burden, reimbursement, digital literacy, language adaptation, and access for older, rural, and lower-income survivors.

8. Some supplementary tables, particularly Table S3, are difficult to navigate. I recommend landscape formatting for large matrices, clearer headers and symbols, improved GRADE table alignment, and considering separate Excel files for Tables S3 and S4.

Author Response

Comment 1: I recommend moderating statements that digital interventions “consistently improved” outcomes. Since 83% of reviews included were rated low confidence by AMSTAR-2, the Abstract and Conclusion should better reflect this limitation. A more appropriate wording would be: “Digital interventions generally showed favorable effects on quality of life, physical activity, and psychosocial outcomes, although confidence in many reviews was low and implementation evidence remains limited.”

Response 1: Thank you for the comment and suggestion. We agree that the original wording may have overstated the strength and consistency of the evidence, particularly given that 19 of the 23 included reviews (83%) were rated as having low methodological confidence using AMSTAR-2. As such, we have moderated the language in the Abstract and Conclusions to distinguish the generally favorable direction of reported effects from the limitations in methodological quality and certainty of evidence. We have also emphasized that favorable clinical outcomes do not necessarily indicate readiness for widespread implementation, particularly in the absence of robust long-term, economic, and health-system integration data.

 

Comment 2: The eligibility criteria refer to survivors who completed primary treatment, yet several reviews included participants during treatment or across the care continuum. Please clarify how “survivor” was defined, how mixed treatment-phase populations were handled, and whether the conclusions apply mainly to post-treatment survivorship or more broadly to supportive cancer care.

Response 2: We thank the reviewer for identifying this important inconsistency between our stated eligibility criteria and the populations represented in several included reviews. We acknowledge that our prospectively registered protocol originally specified breast cancer survivors who had completed primary treatment, whereas several included systematic reviews encompassed participants undergoing active treatment or populations spanning the cancer care continuum. During study selection, reviews involving mixed treatment-phase populations were retained to capture evidence relevant to breast cancer supportive care, particularly where findings for post-treatment survivors were not separately reported. We recognise that this constituted a broadening of the originally specified population eligibility criteria and should have been explicitly documented. We have therefore revised Section 2.2 of Methods to distinguish the original protocol-defined population from the broader population included in the final synthesis and have transparently reported this deviation from the registered protocol. We have also expanded Section 4.5 Limitations of Discussion to acknowledge the potential indirectness introduced by mixed treatment-phase populations and have moderated the text to clarify that the findings apply more broadly to supportive breast cancer care and cannot necessarily be generalized exclusively to post-treatment survivorship.

 

Comment 3: The Discussion about AMSTAR-2 and GRADE should better explain how methodological limitations affect confidence in the findings. I suggest distinguishing between the generally favorable direction of effect, the often-limited certainty of evidence, and the still insufficient implementation readiness.

Response 3: Thank you for the comment and suggestion. We agree that the discussion about methodological limitations could be further expounded on and the organisation improved for greater clarity. We have expanded Section 4.5 of Discussion to discuss the AMSTAR-2 findings to a greater extent and clarified that methodological limitations, including incomplete consideration of risk of bias, limited assessment of publication bias, and inadequate reporting of long-term outcomes, constrain the strength of clinical recommendations. We have also separated the Discussion into paragraphs to explore each of the pointers separately.

 

Comment 4: Table 1 is informative but overly dense. I recommend retaining only key items in the main manuscript, such as author/year, modality, number of studies, primary outcomes, direction of effect, and AMSTAR-2 confidence, while moving detailed information to the Supplementary Material.

Response 4: We thank the reviewer for this practical suggestion. We agree that the original Table 1 contained substantial detail that, although useful for transparency, reduced readability and made it difficult for readers to identify the main findings. We have now streamlined Table 1 to retain the most clinically relevant characteristics as per the reviewer’s suggestion. A more complete table that was the original Table 1 has been moved to the Supplementary Material as Table S1.

 

Comment 5: The Conclusion should better distinguish between intervention types. mHealth and web-based interventions have the broadest evidence base, telephone interventions may support digital equity, wearables are most relevant to physical activity, VR remains promising but less certain, and evidence for AI chatbots is still limited.

Response 5: Thank you for the comment. We agree that the previous Conclusion did not adequately reflect the heterogeneity of intervention modalities and the differences in the quantity and certainty of evidence supporting each. We have thus revised the Conclusion to distinguish the relatively extensive evidence for mHealth and web-based interventions from the more targeted applications of telephone-based interventions and wearable activity trackers. We have also emphasised the potential accessibility advantages of telephone-based care, while acknowledging that its equity benefits require further evaluation. Finally, we have highlighted the promising but less certain evidence for VR-based interventions and the preliminary nature of evidence for AI chatbots.

 

Comment 6: Although overall overlap was slight, some mHealth reviews shared substantial primary evidence. I recommend briefly noting that this overlap may partly contribute to the apparent consistency of findings.

Response 6: We thank the reviewer for highlighting this important consideration. Although the overall corrected covered area was low (1.72%), we agree that this aggregate measure may obscure substantial overlap between individual reviews addressing similar intervention modalities. We have therefore expanded Section 4.5 of Discussion to acknowledge that the highest pairwise overlap was 40.0%, observed between the mHealth-focused reviews by Saevarsdottir et al. and Wanchai et al.

 

Comment 7: The four-tier integration ladder is a useful contribution. I suggest briefly incorporating practical considerations such as referral pathways, clinician responsibility, electronic medical record integration, privacy, workflow burden, reimbursement, digital literacy, language adaptation, and access for older, rural, and lower-income survivors.

Response 7: We thank the reviewer for recognizing the potential utility of the proposed four-tier integration ladder. We agree that its clinical relevance can be strengthened by explicitly identifying the operational, organizational, and equity-related considerations that may influence implementation. We have expanded the accompanying Section 4.4 of Discussion to address the aforementioned pointers. We have also clarified that the integration ladder is a proposed conceptual framework rather than a validated implementation model, and that progression to higher levels of integration requires additional evaluation of feasibility, acceptability, safety, and resource implications.

 

Comment 8: Some supplementary tables, particularly Table S3, are difficult to navigate. I recommend landscape formatting for large matrices, clearer headers and symbols, improved GRADE table alignment, and considering separate Excel files for Tables S3 and S4.

Response 8: We thank the reviewer for this helpful recommendation. We agree that the large supplementary matrices would benefit from improved formatting and navigation. We have now reformatted the supplementary tables to improve readability, including landscape orientation where appropriate, clearer column headings, standardized abbreviations and symbols, and improved alignment of the GRADE evidence summaries. We hope this improves clarity.

Reviewer 2 Report

Comments and Suggestions for Authors

It is a very exhaustive revision on digital tools to optimize patient care after breast cancer care, and the conclusion is thart digital health interventions improve QoL, physical well-being and also the psycological aspects. The paper is well structured, the conclussions are adequate, the referencces are correct, the English is perfect. I have no comments further comments on it.

Author Response

Comment 1: It is a very exhaustive revision on digital tools to optimize patient care after breast cancer care, and the conclusion is thart digital health interventions improve QoL, physical well-being and also the psycological aspects. The paper is well structured, the conclussions are adequate, the referencces are correct, the English is perfect. I have no comments further comments on it.

Response 1: We thank the reviewer for the encouraging assessment of the manuscript and for considering the review relevant and well structured. We have retained its overall organisation while incorporating the other reviewers’ suggestions to improve clarity and appropriately qualify the conclusions.

Reviewer 3 Report

Comments and Suggestions for Authors

Review of the article "Digital Health Supportive Care Interventions for Breast Cancer Survivors: An Umbrella Review of Systematic Reviews and Meta-Analyses".

The article is not easy to understand. For this reason, the Authors should consider moving part of the information presented in the article that is not directly related to cancers to the Appendix.

Points to correct the article:
1. It is very difficult to understand from the article what "digital health interventions" is. Readers unfamiliar with the term will find it difficult to grasp, based on the article, what this technology is and how it works.

2. The impact of "digital health interventions" on people depends, among others, on the stage of the cured cancer, as well as the patients' nationality and age - which should be examined in the article.

3. In my opinion, it would be better to present the very huge amount of information presented in Table 1 as, for example, a bar chart, along with moving Table 1 to the Appendix.

4. In the Conclusions section, there is:
"Digital health interventions consistently improve QoL, physical activity and psychosocial outcomes among breast cancer survivors when delivered as standalone tools."

This conclusion is too general, i.e., the Authors should try to add interpretation of why this improvement occurs. The Conclusions section has been written in terms that are too general. It should specifically relate to the analysis conducted in the article.

5. The quality of Figure 3 should be improved.

6. The term "digital health interventions" encompasses a wide range of techniques; some of them may have a positive impact on humans, while others may have a negative impact, which should be carefully analyzed in this article.

7. "Digital health interventions" may also have numerous drawbacks, which the Authors should analyze in detail.

Author Response

Comment 1: The article is not easy to understand. For this reason, the Authors should consider moving part of the information presented in the article that is not directly related to cancers to the Appendix.

Response 1: We thank Reviewer 3 for their careful assessment of our manuscript and for highlighting opportunities to improve its accessibility and clarity. We recognize that the broad scope of digital health interventions and the substantial amount of evidence synthesized may make the manuscript challenging for readers unfamiliar with the field. To address this concern, we have simplified the presentation of review characteristics, clarified the definitions and practical applications of the intervention modalities, and revised the Discussion and Conclusions to focus more explicitly on breast cancer supportive care. We have also moved detailed methodological information to the Supplementary Materials where appropriate, while retaining information essential to understanding the review's objectives, methods, and findings in the main text.

 

Comment 2: It is very difficult to understand from the article what "digital health interventions" is. Readers unfamiliar with the term will find it difficult to grasp, based on the article, what this technology is and how it works.

Response 2: We thank the reviewer for identifying the need for a more accessible explanation of digital health interventions. Although a definition was provided in the original Introduction, we agree that additional practical examples would help readers unfamiliar with digital health technologies understand their application in breast cancer supportive care. We have thus expanded the Introduction to properly define digital health interventions and provide some illustration as to what some of the examples would entail in context. We hope this would suffice.

 

Comment 3: The impact of "digital health interventions" on people depends, among others, on the stage of the cured cancer, as well as the patients' nationality and age - which should be examined in the article.

Response 3: We agree that patient characteristics may influence the effectiveness, acceptability, and accessibility of digital health interventions. As this study is an umbrella review of published systematic reviews and meta-analyses, our ability to examine these factors depends on the extent to which they were reported and analysed in the included reviews. Unfortunately, due to reporting heterogeneity, the extent to which subgroup-specific conclusions can be drawn is limited. The available extraction does not provide comparable stratified effect estimates across reviews, so we cannot quantify age-, stage- or nationality-specific effects. We state this limitation and recommend reporting these characteristics and prespecifying adequately powered interaction analyses in future studies. We seek your kind understanding in this matter.

 

Comment 4: In my opinion, it would be better to present the very huge amount of information presented in Table 1 as, for example, a bar chart, along with moving Table 1 to the Appendix.

Response 4: We thank the reviewer for this suggestion, which echoes Comment 1.4 raised by Reviewer 1. We agree that the original Table 1 was excessively detailed. However, we do not agree that a visual representation like a bar chart would necessarily be superior. The bar chart enables visualisation of only one variable, while the table allows comparison across multiple columns and rows depending on the variable readers are looking into, and thus should remain as the core display of information in a review study such as ours. To address this concern, we have substantially condensed Table 1 to retain only essential review characteristics and findings, while transferring the detailed information to the Supplementary Materials as Table S1. We believe that retaining a concise main-text table provides an appropriate balance between readability and scientific transparency, as the table allows readers to identify the findings and methodological confidence of individual reviews which cannot be fully represented by visualisations such as a bar chart.

 

Comment 5: In the Conclusions section, there is: "Digital health interventions consistently improve QoL, physical activity and psychosocial outcomes among breast cancer survivors when delivered as standalone tools." This conclusion is too general, i.e., the Authors should try to add interpretation of why this improvement occurs. The Conclusions section has been written in terms that are too general. It should specifically relate to the analysis conducted in the article.

Response 5:  We thank the reviewer for this insightful comment. We agree that the original Conclusion was overly general and the manuscript did not adequately explain the potential mechanisms underlying the favorable effects observed across different intervention modalities. We have revised Section 4.2 of Discussion to relate the observed outcomes more explicitly to the supportive care functions of each intervention type. In particular, we discuss how structured education, symptom self-monitoring, personalized feedback, behavioural reinforcement, remote counselling, and rehabilitation support may contribute to improvements in quality of life, physical activity, and psychosocial well-being. We have also distinguished these plausible mechanisms from mechanisms directly established by the included evidence, as the umbrella review was not designed to establish causal mediation. Finally, we have moderated the Conclusion to reflect the methodological limitations and heterogeneity of the included reviews.

 

Comment 6: The quality of Figure 3 should be improved.

Response 6: We acknowledge this point that Figure 3 can be improved to facilitate interpretation. Figure 3 updated and re-uploaded accordingly.

 

Comment 7: The term "digital health interventions" encompasses a wide range of techniques; some of them may have a positive impact on humans, while others may have a negative impact, which should be carefully analyzed in this article.

Response 7: We agree. We have removed blanket statements of benefit and brought mixed and null findings into the Results and abbreviated Table 1. We concur that future studies should however systematically evaluate limitations of these interventions, such as safety and privacy, alongside clinical effectiveness. We also distinguish review-wide effects from effects that can be assigned to a particular technology, especially AI chatbots, and clarify that the available data cannot support a comparative harms meta-analysis.

 

Comment 8: "Digital health interventions" may also have numerous drawbacks, which the Authors should analyze in detail.

Response 8: We agree and have expanded the discussion of drawbacks, "Potential harms and practical drawbacks also differ by modality. Information-heavy apps may increase distress, and VR can cause physical discomfort such as nausea or dizziness [45,48]. Repeated prompts or symptom monitoring may be burdensome; wearables require charging and sustained use, while telephone support requires staff time and appointments. Privacy breaches, inaccurate automated advice and unclear responsibility for responding to alerts are potential implementation risks, not event rates established by this review. Programmes should therefore specify content review, data protection, clinical escalation and alternatives for patients who disengage. The available synthesis does not permit reliable comparisons of harm rates across modalities." We believe the current Discussion provides a balanced appraisal of the benefits and limitations of digital health interventions in breast cancer supportive care.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

All comments have been addressed.

Reviewer 3 Report

Comments and Suggestions for Authors

The Authors have appropriately addressed my comments. The article, after improving, is better and, in my opinion, is suitable for publication in Cancers.