Identifying Behaviour Change Techniques in Cancer Nutrition Interventions and Their Implementation Contexts: A Systematic Review

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

• Title: What does the “in context” refer to?
• Line 50: Why “estimated”?
• Lines 86-91: These questions sound much more like a scoping review instead of a systematic review.
• Line 100: Provide the search strategy here.
• Line 111: How can you include a comparator group? The focus is on implementation.
• Line 129: Please elaborate on the interrater-reliability.
• Line 166: This needs to be corrected. You did not check 4,055 full texts.
• There are too many tables within the manuscript. Think about synthesizing the content much more and provide further information in a supplementary appendix.

Author Response

Comment 1: Title: What does the “in context” refer to?

Response 1: Thank you for raising this point. To improve clarity, we have amended the title to “Identifying Behaviour Change Techniques in Cancer Nutrition Interventions and their Implementation Contexts: A Systematic Review”.

 

Comment 2: Line 50: Why “estimated”?

Response 2: Thank you for highlighting this. We have revised the wording to ‘close to 20 million cases’ to improve accuracy.

 

Comment 3: Lines 86-91: These questions sound much more like a scoping review instead of a systematic review.

Response 3: We agree that the objectives include elements commonly associated with scoping reviews, particularly the broad identification and mapping of barriers and BCTs. Nevertheless, this review was conducted and reported as a systematic review in accordance with PRISMA guidelines, including a comprehensive search strategy, predefined eligibility criteria, thorough screening and structured data extraction, and systematic coding using established theoretical frameworks. Importantly, the review sought not only to map existing evidence but also to synthesise findings across studies, including outcomes at patient, service, and implementation levels, and to examine alignment with established mechanisms of behaviour change. To strengthen clarity, we have revised as follows:

“This systematic review aimed to systematically identify, code, and synthesise evidence on the use of BCTs in nutrition interventions and their implementation as delivered in adult oncology care, using established theoretical frameworks to support analytical synthesis. The specific objectives were to:”

 

Comment 4: Line 100: Provide the search strategy here.

Response 4: Thanks for noting this. We have now provided the search strategy via an appropriate citation.

 

Comment 5: Line 111: How can you include a comparator group? The focus is on implementation.

Response 5: Thank you for raising this point. The predefined selection criteria allowed for comparator groups in instances where implementation controls were included, such as in Type 3 hybrid effectiveness-implementation trials.

 

Comment 6: Line 129: Please elaborate on the interrater-reliability.

Response 6: Thank you for this comment. We have clarified the approach as follows:

“Inter-rater reliability was monitored throughout the screening process by comparing reviewer decisions and resolving discrepancies to ensure consistent application of eligibility criteria. For quality control, a random sample of 15 full text articles was independently reviewed by implementation science experts (AM and SL), with 100% agreement observed.”

 

Comment 7: Line 166: This needs to be corrected. You did not check 4,055 full texts.

Response 7: Thank you for flagging this error. We have corrected the text to specify the number of full texts screened:

“Of the 5,168 records identified through the systematic search, 4,055 abstracts met the inclusion criteria and 165 records were screened at the full text level.”

 

Comment 8: There are too many tables within the manuscript. Think about synthesizing the content much more and provide further information in a supplementary appendix.

Response 8: Thank you for this suggestion. In response, we have moved Tables 5&6 to the supplementary materials.

Reviewer 2 Report

Comments and Suggestions for Authors

Review for Nutrients-4015735 “Identifying Behaviour Change Techniques in Cancer Nutrition Interventions and their Implementation Strategies in Context: 3 A Systematic Review”

 

Overall, this is a relevant and potentially useful systematic review of behaviour change techniques (BCTs) in oncology nutrition. However, several substantial issues in framing, methods, interpretation, and reporting need to be addressed before the manuscript is suitable for publication.

1. Conceptual framing and oxidative stress biomarkers in cancer
   The Introduction and Discussion focus mainly on implementation frameworks (TDF, BCT taxonomy, Proctor outcomes, CFIR 2.0), but they do not clearly explain why improving nutrition behaviours is biologically critical in oncology. There is a substantial body of work on oxidative stress and related biomarkers in cancer that links nutritional status, dietary patterns, and malnutrition to redox imbalance, inflammation, and treatment outcomes. I strongly recommend adding 1–2 paragraphs that explicitly connect cancer nutrition and behaviour change to oxidative stress biology and cancer related oxidative stress biomarkers, and citing representative studies in this area. This will strengthen the justification for the review and make the clinical importance of the mapped BCTs clearer.  Please cite the following study

494. Lin, E. T.; Bae, Y.; Birkett, R.; Sharma, A. M.; Zhang, R.; Fisch, K. M.; Funk, W.; Mestan, K. K. Cord Blood Adductomics Reveals Oxidative Stress Exposure Pathways of Bronchopulmonary Dysplasia. Antioxidants 2024, 13(4), 494. DOI: doi:10.3390/antiox13040494.
495. Funk, W. E.; Montgomery, N.; Bae, Y.; Chen, J.; Chow, T.; Martinez, M. P.; Lurmann, F.; Eckel, S. P.; McConnell, R.; Xiang, A. H. Human Serum Albumin Cys34 Adducts in Newborn Dried Blood Spots: Associations With Air Pollution Exposure During Pregnancy. Frontiers in Public Health 2021, 9(2062), Original Research. DOI: 10.3389/fpubh.2021.730369.
496. Mestan, K.; Gotteiner, N.; Porta, N.; Grobman, W.; Su, E.J.; Ernst, L.M. Cord Blood Biomarkers of Placental Maternal Vascular Underperfusion Predict Bronchopulmonary Dysplasia-Associated Pulmonary Hypertension. J. Pediatr. 2017, 185, 33–41
497. Voller, B.; Chock, S.; Ernst, L.M.; Su, E.; Liu, X.; Farrow, K.N.; Mestan, K.K. Cord blood biomarkers of vascular endothelial growth (VEGF and sFlt-1) and postnatal growth: A preterm birth cohort study. Early Hum. Dev. 2014, 90, 195–200.

 

1. Methods, PRISMA flow, and outcome classification
   The description of the search and selection process is confusing and internally inconsistent. For example, the statement that “4,055 abstracts met the inclusion criteria and were screened at the full text level” does not align with usual PRISMA logic or with the numbers in Figure 1. Please carefully reconcile all numbers across the text and the flow diagram. In addition, for the key implementation science elements (TDF coding of barriers, BCT coding of interventions and strategies), no information is given on inter rater agreement or how disagreements were resolved. Given that your main claims rest on these codings, you should report at least basic agreement statistics or a clearer description of the consensus process. Finally, the criteria for classifying implementation and service outcomes as “positive”, “neutral”, or “mixed” are not defined. These criteria need to be clearly stated to support later statements that certain BCTs are “frequently observed alongside positive outcomes”.
2. Interpretation of BCT patterns and “mechanisms”
   The Results and Discussion repeatedly describe certain BCTs, especially “Instruction on how to perform the behaviour” and “Credible source”, as “central”, “critical”, or “particularly influential” because they often co occur with positive outcomes. However, the analyses are descriptive frequency counts, without formal comparison between successful and unsuccessful interventions or any effect size estimation. Co occurrence alone is not sufficient to infer that these BCTs are active ingredients or mechanisms of change. Similarly, the discussion of “emerging mechanisms” and the reference to Kazdin’s criteria goes beyond what the data can support. I recommend tempering the language, explicitly framing these findings as hypothesis generating, and being clear that mechanistic claims require prospective testing rather than retrospective mapping alone.
3. Reporting quality, tables, and editorial issues
   There are a number of presentation problems that should be systematically corrected. Table 5 still contains template text (“This is a table. Tables should be placed…”), which must be replaced by a proper caption. Several tables are very long and include multiple rows for publications from the same underlying study (for example, the EAT program), making the manuscript difficult to read; consider consolidating these into a single row per intervention and moving detailed BCT lists to supplementary files. There are also numerous typographical and terminology issues (for example, inconsistent spelling of author names, minor English errors) that require careful copy editing. Overall, tightening and partially shifting detail to the Supplementary Material would help the paper focus more clearly on the main findings and implications.

Comments for author File: Comments.pdf

Author Response

Comment 1: Overall, this is a relevant and potentially useful systematic review of behaviour change techniques (BCTs) in oncology nutrition. However, several substantial issues in framing, methods, interpretation, and reporting need to be addressed before the manuscript is suitable for publication.

Conceptual framing and oxidative stress biomarkers in cancer
The Introduction and Discussion focus mainly on implementation frameworks (TDF, BCT taxonomy, Proctor outcomes, CFIR 2.0), but they do not clearly explain why improving nutrition behaviours is biologically critical in oncology. There is a substantial body of work on oxidative stress and related biomarkers in cancer that links nutritional status, dietary patterns, and malnutrition to redox imbalance, inflammation, and treatment outcomes. I strongly recommend adding 1–2 paragraphs that explicitly connect cancer nutrition and behaviour change to oxidative stress biology and cancer related oxidative stress biomarkers, and citing representative studies in this area. This will strengthen the justification for the review and make the clinical importance of the mapped BCTs clearer.  Please cite the following study

1. Lin, E. T.; Bae, Y.; Birkett, R.; Sharma, A. M.; Zhang, R.; Fisch, K. M.; Funk, W.; Mestan, K. K. Cord Blood Adductomics Reveals Oxidative Stress Exposure Pathways of Bronchopulmonary Dysplasia. Antioxidants 2024, 13(4), 494. DOI: doi:10.3390/antiox13040494.
2. Funk, W. E.; Montgomery, N.; Bae, Y.; Chen, J.; Chow, T.; Martinez, M. P.; Lurmann, F.; Eckel, S. P.; McConnell, R.; Xiang, A. H. Human Serum Albumin Cys34 Adducts in Newborn Dried Blood Spots: Associations With Air Pollution Exposure During Pregnancy. Frontiers in Public Health 2021, 9(2062), Original Research. DOI: 10.3389/fpubh.2021.730369.
3. Mestan, K.; Gotteiner, N.; Porta, N.; Grobman, W.; Su, E.J.; Ernst, L.M. Cord Blood Biomarkers of Placental Maternal Vascular Underperfusion Predict Bronchopulmonary Dysplasia-Associated Pulmonary Hypertension. J. Pediatr. 2017, 185, 33–41
4. Voller, B.; Chock, S.; Ernst, L.M.; Su, E.; Liu, X.; Farrow, K.N.; Mestan, K.K. Cord blood biomarkers of vascular endothelial growth (VEGF and sFlt-1) and postnatal growth: A preterm birth cohort study. Early Hum. Dev. 2014, 90, 195–200.

Response 1: We thank the reviewer for this comment. We fully acknowledge the important body of literature linking nutrition, oxidative stress biology, and cancer outcomes. However, the primary aim of this review was to examine the implementation of nutrition interventions and behaviour change techniques in adult oncology care. For this reason, the

 

review was intentionally scoped around behavioural and implementation mechanisms rather than underlying biological pathways. To maintain conceptual coherence and focus, we did not amend the manuscript to include additional discussion of oxidative stress biomarkers.

 

Comment 2: Methods, PRISMA flow, and outcome classification
The description of the search and selection process is confusing and internally inconsistent. For example, the statement that “4,055 abstracts met the inclusion criteria and were screened at the full text level” does not align with usual PRISMA logic or with the numbers in Figure 1. Please carefully reconcile all numbers across the text and the flow diagram. In addition, for the key implementation science elements (TDF coding of barriers, BCT coding of interventions and strategies), no information is given on inter rater agreement or how disagreements were resolved. Given that your main claims rest on these codings, you should report at least basic agreement statistics or a clearer description of the consensus process. Finally, the criteria for classifying implementation and service outcomes as “positive”, “neutral”, or “mixed” are not defined. These criteria need to be clearly stated to support later statements that certain BCTs are “frequently observed alongside positive outcomes”.

Response 2: Thank you for these helpful comments. We have corrected the text to include the number of full text screened:

“4,055 abstracts met the inclusion criteria and 165 records were screened at the full text level”

We appreciate the request for greater clarity regarding the coding process. We have revised this section accordingly:

“Data extraction was conducted by 1 reviewer and independently verified by a second. Extraction related to implementation science were performed by researchers with expertise in the field (SL, AM), while clinician researchers were responsible for extracting data related to clinical outcomes (NCF, AL, FF, CR, HB). Coding to the TDF, BCT Taxonomy v1,[16] and Proctor’s outcomes framework was conducted independently by reviewers (SL, AM). Discrepancies were resolved through discussion and, where necessary, consultation with a third reviewer (NT). To further ensure coding reliability, NT independently reviewed a random 10% sample of the coded data. The data extraction and coding template was developed and pilot tested using REDCap (Research Electronic Data Capture)[17] and Microsoft Excel[18] prior to application (SL and NCF).”

Thank you for the comment regarding classification of outcomes. We have clarified that the classification was based on author interpretation. Given the breadth and heterogeneity of outcomes (including qualitative and quantitative measures across multiple domains), it was not feasible to apply a single uniform rule for categorisation. This classification has now been explicitly stated in the manuscript:

“Outcomes were categorised using Proctor’s framework into implementation, service, and patient-level outcomes,[21] and were further classified as positive, neutral or negative based on author interpretation.”

 

Comment 3: Interpretation of BCT patterns and “mechanisms”
The Results and Discussion repeatedly describe certain BCTs, especially “Instruction on how to perform the behaviour” and “Credible source”, as “central”, “critical”, or “particularly influential” because they often co occur with positive outcomes. However, the analyses are descriptive frequency counts, without formal comparison between successful and unsuccessful interventions or any effect size estimation. Co occurrence alone is not sufficient to infer that these BCTs are active ingredients or mechanisms of change. Similarly, the discussion of “emerging mechanisms” and the reference to Kazdin’s criteria goes beyond what the data can support. I recommend tempering the language, explicitly framing these findings as hypothesis generating, and being clear that mechanistic claims require prospective testing rather than retrospective mapping alone.

Response 3: We appreciate this comment. Agree and we have carefully reviewed and tempered the language throughout to explicitly frame these findings as descriptive and hypothesis-generating (Table 4 and Results).

 

Comment 4: Reporting quality, tables, and editorial issues
There are a number of presentation problems that should be systematically corrected. Table 5 still contains template text (“This is a table. Tables should be placed…”), which must be replaced by a proper caption. Several tables are very long and include multiple rows for publications from the same underlying study (for example, the EAT program), making the manuscript difficult to read; consider consolidating these into a single row per intervention and moving detailed BCT lists to supplementary files. There are also numerous typographical and terminology issues (for example, inconsistent spelling of author names, minor English errors) that require careful copy editing. Overall, tightening and partially shifting detail to the Supplementary Material would help the paper focus more clearly on the main findings and implications.

Response 4: We appreciate your careful attention to presentation details. We have undertaken a comprehensive review of the manuscript to address these issues. Specifically:

The placeholder text in Table 5 has been corrected, and the title revised to: “Behaviour Change Techniques identified in implementation strategies that aligned with positive patient outcomes.”

Tables 5&6 have been moved to the supplementary materials, with Tables 3&4 remain in the main manuscript as they represent the core focus of this review.

We have conducted a thorough review to fix typographical errors, and author name spelling.

 

We also wish to clarify that intervention components are presented at the study level, whereas implementation strategies are reported at the article level. This distinction reflects that nutrition interventions were often identical across publications, while implementation contexts and strategies varied between articles, within a given study. We have retained this structure to preserve conceptual accuracy while reducing redundancy.

Reviewer 3 Report

Comments and Suggestions for Authors

First, I would like to thank you for the opportunity to review the article “Identifying Behavior Change Techniques in Cancer Nutrition Interventions and their Implementation Strategies in Context: A Systematic Review.” I believe this article provides a comprehensive overview of how behavior change techniques have been used in oncology nutritional interventions. Its main contribution is highlighting the lack of clarity and consistency in the application of behavior change techniques (BCTs), as well as opportunities to improve their reporting and design. However, I would like to suggest several improvements that I believe could enhance the quality of the document.

Regarding the abstract, I believe the authors should summarize methodological details and include a final summary with recommendations for clinical practice and research, as well as contextualize the importance and relevance of BCT analysis in oncology.
The introduction justifies the need for interventions and clarifies the lack of a systematic review of BCTs applied in oncology nutrition. In this section, I recommend summarizing or integrating the initial paragraphs to avoid redundancy and including a summary paragraph regarding the nutritional interventions and their behavioral design, as well as the rationale for why understanding the mechanisms of change is relevant for improving outcomes.
The methodology is clear, follows PRISMA, and cites the checklist and prior registration in PROSPERO. The authors provide a clear description of the databases and inclusion criteria, as well as the double-checking and third-party review process. It is not specified how the team was trained to code BCTs (essential when coding complex components). I would recommend adding the calculation of reliability measures (Kappa, discrepancies, consensus method) and clarifying whether official versions of the taxonomies were used (e.g., BCTTv1).
The results section is very detailed, with a large volume of data structured in an understandable way, identifying barriers by TDF domains, and clearly categorizing BCTs within interventions. I believe the manuscript could be improved by including matrices or diagrams to allow for a more interpretive visualization, and by incorporating an analysis of comparisons and differential patterns between types of interventions, countries, or methodologies. At the same time, although it describes the frequency of BCTs and barriers, the manuscript does not assess the quality or intensity of each BCT within each intervention and does not distinguish whether the BCTs were active, nominal, or simply mentioned.

Furthermore, the assessment of bias and quality should be evaluated. Although QUADS is used, the analysis does not discuss the implications of the risk of bias in the interpretation, nor does it analyze differences in results according to methodological quality. Finally, when referring to "full sequence of change," it only identifies two studies, but does not delve into why the others do not show a relationship.
Regarding the discussion, I believe a critical discussion of the quality of the included studies is lacking, and unexpected findings are not sufficiently explained (for example, the poor alignment between TDF and BCT in some cases). At the same time, I would recommend relating the findings to broader implementation frameworks (CFIR, ERIC) and including an explanation of why the most frequently used BCTs might be the most effective. I believe you should discuss which barriers tend to be resolved most often and which BCT works best for each type of barrier. You should also indicate clear implications for clinical practice and the design of future interventions. Additionally, I believe a section should be added discussing how QUADS scores affect the confidence in the evidence and indicating whether higher-quality studies show different patterns in BCT use.

Regarding limitations, the authors acknowledge the heterogeneity among studies and mention limitations not only of the study itself but also of the methodologies in the field. I believe they should include the limitations inherent in the coding process, which is a critical part of the study.
Regarding the conclusions, operational recommendations should be included, such as “explicitly reporting interventions with TIDieR and BCTTv1” and integrating concepts from the analysis itself (e.g., emergent mechanisms).

Author Response

Comment 1: First, I would like to thank you for the opportunity to review the article “Identifying Behavior Change Techniques in Cancer Nutrition Interventions and their Implementation Strategies in Context: A Systematic Review.” I believe this article provides a comprehensive overview of how behavior change techniques have been used in oncology nutritional interventions. Its main contribution is highlighting the lack of clarity and consistency in the application of behavior change techniques (BCTs), as well as opportunities to improve their reporting and design. However, I would like to suggest several improvements that I believe could enhance the quality of the document.

 

Regarding the abstract, I believe the authors should summarize methodological details and include a final summary with recommendations for clinical practice and research, as well as contextualize the importance and relevance of BCT analysis in oncology.

Response 1: Thank you for your thoughtful review and constructive comments. In response, we have revised the abstract to include clearer methodological detail and strengthened the concluding statements accordingly.

 

Comment 2: The introduction justifies the need for interventions and clarifies the lack of a systematic review of BCTs applied in oncology nutrition. In this section, I recommend summarizing or integrating the initial paragraphs to avoid redundancy and including a summary paragraph regarding the nutritional interventions and their behavioral design, as well as the rationale for why understanding the mechanisms of change is relevant for improving outcomes.

Response 2: Thank you for this comment. We carefully reviewed the introduction in light of this suggestion. The section was intentionally structured to progress from the burden of cancer, to the role of nutrition in oncology care, and then to the implementation challenges associated with integrating nutrition care into routine practice. We did not identify substantive redundancy within this flow and believe it remains fit for purpose. We also note that, as stated in Objective 4, the exploration of mechanisms of behaviour change is explicitly exploratory rather than a central focus of the review. We appreciate the suggestion and would welcome further clarification if we have misunderstood the intent of this comment.

 

Comment 3: The methodology is clear, follows PRISMA, and cites the checklist and prior registration in PROSPERO. The authors provide a clear description of the databases and inclusion criteria, as well as the double-checking and third-party review process. It is not specified how the team was trained to code BCTs (essential when coding complex components). I would recommend adding the calculation of reliability measures (Kappa, discrepancies, consensus method) and clarifying whether official versions of the taxonomies were used (e.g., BCTTv1).

Response 3: Thank you for this helpful suggestion. We have revised this section to provide greater clarity regarding the coding component, the use of the official BCTTv1, with a reference added.

“Extraction related to implementation science were performed by researchers with expertise in the field (SL, AM), while clinician researchers were responsible for extracting data related to clinical outcomes (NCF, AL, FF, CR, HB). Coding to the TDF, BCT Taxonomy v1,[16] and Proctor’s outcomes framework was conducted independently by reviewers (SL, AM). Discrepancies were resolved through discussion and, where necessary, consultation with a third reviewer (NT). To further ensure coding reliability, NT independently reviewed a random 10% sample of the coded data. The data extraction and coding template was developed and pilot tested using REDCap (Research Electronic Data Capture)[17] and Microsoft Excel[18] prior to application (SL and NCF).”

 

Comment 4: The results section is very detailed, with a large volume of data structured in an understandable way, identifying barriers by TDF domains, and clearly categorizing BCTs within interventions. I believe the manuscript could be improved by including matrices or diagrams to allow for a more interpretive visualization, and by incorporating an analysis of

 

comparisons and differential patterns between types of interventions, countries, or methodologies. At the same time, although it describes the frequency of BCTs and barriers, the manuscript does not assess the quality or intensity of each BCT within each intervention and does not distinguish whether the BCTs were active, nominal, or simply mentioned.

Response 4: Thank you for your positive assessment of the Results section. Given that the primary aim of this review was comprehensive evidence mapping, the substantial heterogeneity across interventions, contexts, and outcome measures limited the feasibility of formal comparisons. We agree that the quality, intensity, and functional role of BCTs could not be assessed based on available reporting and have now explicitly expanded the Limitations section to acknowledge this constraint and its implications for interpretation:

“Despite its strengths, this review has some limitations that should be acknowledged. The patterns identified between BCTs and outcomes were observational in nature, limiting the ability to draw causal inferences. The effectiveness of individual BCTs or specific combinations was not experimentally evaluated, and as such, conclusions about their relative impact remain uncertain. A further limitation is the reliance on retrospective coding of published descriptions to theoretical frameworks, which is constrained by the complete-ness and clarity of reporting. As a result, while the review describes the frequency of re-ported BCTs, it was not possible to assess the quality, intensity, or functional role of each BCT within interventions, nor to distinguish whether BCTs were actively delivered, nominally included, or simply mentioned. In addition, implementation and service outcomes were often reported qualitatively, with limited use of standardised measures. This variability in reporting may have influenced the comparability of findings across studies.”

 

Comment 5: Furthermore, the assessment of bias and quality should be evaluated. Although QUADS is used, the analysis does not discuss the implications of the risk of bias in the interpretation, nor does it analyze differences in results according to methodological quality. Finally, when referring to "full sequence of change," it only identifies two studies, but does not delve into why the others do not show a relationship.

Response 5: We appreciate you raising these important points. We have expanded the Discussion to provide a clearer interpretation of the QuADS scores:

“QuADS assessments indicated moderate to high methodological quality across all included studies. While the total scores suggest most studies met a substantial proportion of QuADS criteria, cut-offs were based on those applied in previously published systematic reviews.[62] Due to the lack of official guidance, these user-defined thresholds are for convenience and are not empirically validated. They should therefore be considered pragmatic descriptors rather than formal indicators of study quality, consistent with the tool’s design for item-level appraisal rather than categorical classification.[15] Nonetheless, the predominance of studies meeting high-quality criteria support the credibility of the findings of this review.”

 

Regarding ‘full sequence of change’, we have clarified in the Methods and Discussion that these were identified only in studies reporting BCTs across both intervention and implementation strategies and outcomes at both implementation/service and patient levels. The limited number of such studies reflects reporting constrains/evidence availability rather than an absence of relationships, and this has now been explicitly stated:

“To explore mechanistic pathways, full sequences of observed change were analysed in studies where BCTs were present in both intervention and implementation strategies and outcomes were reported at both implementation/service and patient levels.”

 

Comment 6: Regarding the discussion, I believe a critical discussion of the quality of the included studies is lacking, and unexpected findings are not sufficiently explained (for example, the poor alignment between TDF and BCT in some cases). At the same time, I would recommend relating the findings to broader implementation frameworks (CFIR, ERIC) and including an explanation of why the most frequently used BCTs might be the most effective. I believe you should discuss which barriers tend to be resolved most often and which BCT works best for each type of barrier. You should also indicate clear implications for clinical practice and the design of future interventions. Additionally, I believe a section should be added discussing how QUADS scores affect the confidence in the evidence and indicating whether higher-quality studies show different patterns in BCT use.

Response 6:

Thank you for these detailed comments.

First, we have added a clearer discussion of study quality based on QuADS assessments and its implications for confidence in the findings.

“QuADS assessments indicated moderate to high methodological quality across all included studies. While the total scores suggest most studies met a substantial proportion of QuADS criteria, cut-offs were based on those applied in previously published systematic reviews.[62] Due to the lack of official guidance, these user-defined thresholds are for convenience and are not empirically validated. They should therefore be considered pragmatic descriptors rather than formal indicators of study quality, consistent with the tool’s design for item-level appraisal rather than categorical classification.[15] Nonetheless, the predominance of studies meeting high-quality criteria support the credibility of the findings of this review.”

 

Second, we have also strengthened links to broader implementation frameworks by referencing recent related work applying CFIR and ERIC:

“These structural challenges may be more comprehensively captured through the application of multi-level frameworks such as the updated Consolidated Framework for Implementation Research (CFIR 2.0)[42], which offers a detailed lens for understanding constraints experienced by individuals,[14] particularly within the Environmental Context and Resources domain.”

 

Third, while analysis of differential effectiveness of BCTs or barrier-BCT matching is beyond the scope of this descriptive review and not supported by the available evidence, we highlight this as a priority for future research.

 

Finally, recommendations for more explicit specification and reporting of BCTs in oncology nutrition have been added:

“For clinical practice, more explicit specifications of BCTs may facilitate improved integration of nutrition interventions in oncology care. Future research may…”

 

Comment 7: Regarding limitations, the authors acknowledge the heterogeneity among studies and mention limitations not only of the study itself but also of the methodologies in the field. I believe they should include the limitations inherent in the coding process, which is a critical part of the study.

Response 7:

Thank you for highlighting this important point. We have added an explicit limitation addressing the reliance on retrospective coding:

 

“A further limitation is the reliance on retrospective coding of published descriptions to theoretical frameworks, which is constrained by the completeness and clarity of reporting.”

 

Comment 8: Regarding the conclusions, operational recommendations should be included, such as “explicitly reporting interventions with TIDieR and BCTTv1” and integrating concepts from the analysis itself (e.g., emergent mechanisms).

Response 8:

Thank you for the suggestion. We have revised the Conclusions accordingly:

“…For clinical practice, more explicit specifications of BCTs may facilitate improved integration of nutrition interventions in oncology care. Future research may help evaluate the effectiveness of specific individual BCTs and their combinations across both interventions and approaches to their implementation. Additionally, greater attention to contextual factors that influence implementation, inclusion of diverse interest holder perspectives, particularly those of patients and caregivers, and explicit adherence to reporting guidelines and implementation and behavioural frameworks would support more systematic accumulation of implementation evidence...”

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

All previous comments have adequately been addressed.