Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This cross-sectional study examines the discriminative power of total and segmental body composition indicators, such as lean mass, fat mass, and bone mineral content, for muscle weakness and low physical performance in community-dwelling older adults, using sex-stratified receiver operating characteristic analyses, and the topic is undoubtedly relevant, given the growing public health burden of sarcopenia and functional decline in ageing populations. The use of dual-energy X-ray absorptiometry as the reference method for body composition and sex-stratified analysis are clear strengths, however, several methodological limitations constrain the extent to which conclusions can be drawn, including the use of non-probabilistic convenience sampling, the marked sex imbalance, and the relatively small number of men with muscle weakness or low physical performance, which raises concerns about the reliability and generalisability of the ROC-derived cut-offs, especially for men. The paper would benefit from a more explicit hypothesis, a clearer discussion of the clinical meaning of the modest AUCs for physical performance, and more transparent statistical decision-making.

The Introduction is clear and well organized, and it clearly sets out a logical conceptual pathway from biological ageing and body composition changes to sarcopenia and functional decline, and the argument for focusing specifically on bone mineral content is well articulated and clearly represents a gap in existing literature. However, there is no explicit hypothesis, and the Introduction finishes with the study aim, but does not clearly articulate a directional hypothesis, and the novelty of the study is not highlighted, as the authors indicate that bone mineral content has been less studied than bone mineral density, but they do not provide a clear demarcation of their work relative to prior studies that utilized segmental dual-energy X-ray absorptiometry-based approaches. The literature review is a bit narrow, and the Introduction focuses a lot on lean mass and bone mineral content, and less so on previous studies that have utilized receiver operating characteristic analyses to derive diagnostic cut-offs for body composition variables in older adults, and it would be helpful to expand this section to contextualize the methodology chosen.

The Methods are described clearly, and the dual-energy X-ray absorptiometry protocol, the handgrip strength assessment, and the Short Physical Performance Battery procedures are described in enough detail to enable replication, and are anchored to established guidelines, and the use of the Youden Index to derive optimal cut-off points is methodologically sound and well explained. However, there are several concerns, including the use of non-probabilistic sampling, which the authors admit to, but do not address what this means for representativeness, and the strong sex imbalance, which is not discussed in the Methods, and the authors should clarify whether this reflects the true composition of the senior centres, or whether there were issues in recruitment or selection, and explicitly acknowledge the consequences for the stability and precision of sex-stratified receiver operating characteristic estimates, in particular in men. The normality assessment is also a concern, as the authors mention that they visually inspected the histograms and examined skewness and kurtosis, and then used the Central Limit Theorem as an argument for using parametric tests, but it would be interesting to know whether any of the variables were clearly non-normally distributed, and whether they considered using non-parametric tests or performed sensitivity analyses.

The Discussion is logical and well engaged with prior literature, and the rationale for interpreting handgrip strength as a global indicator of musculoskeletal health is well developed and supported by appropriate references, and the limited discriminative value of adiposity measures is a clinically important finding that is handled sensitively. However, there are several concerns, including the interpretation of area under the curve values in men, which the Discussion states suggest robust predictive performance, but these high area under the curve values are based on very small numbers of male cases, and may therefore be unstable, and the total bone mineral content cut-off for low physical performance in men, which has an area under the curve of 0.66, and at a sensitivity of 100%, the specificity is only 33%, which is clearly suboptimal in the clinic. The sex differences discussion is also somewhat speculative, and the language should be more circumspect and framed as a hypothesis or interpretation, rather than a conclusion that is supported by the present data.

The reference list is broad and up-to-date, with most references dating from the last 10-15 years, and key consensus and guideline documents are appropriately cited, but there are some concerns, including a formatting issue with references 22 and 23, which seem to have been merged into a single continuous line, and the use of self-citations, which should be mostly used to support methods and concepts rather than to overemphasise the authors' previous work. The writing and clarity of the manuscript are generally good, but there are some concerns, including the abstract completeness, which appears to be truncated, and the figure legend, which contains an error, and the table formatting, which could be improved.

Author Response

Comment 1: This cross-sectional study examines the discriminative power of total and segmental body composition indicators, such as lean mass, fat mass, and bone mineral content, for muscle weakness and low physical performance in community-dwelling older adults, using sex-stratified receiver operating characteristic analyses. The topic is undoubtedly relevant given the growing public health burden of sarcopenia and functional decline in ageing populations.

Response 1: We thank the reviewer for this positive evaluation of our work and for recognizing the relevance of the research topic. We agree that understanding the role of body composition in the identification of functional decline in older adults is an important public health issue, particularly in the context of population ageing. We appreciate the reviewer’s acknowledgement of the methodological strengths of the study, including the use of DXA and sex-stratified analyses.

 

Comment 2: Several methodological limitations constrain the extent to which conclusions can be drawn, including the use of non-probabilistic convenience sampling, the marked sex imbalance, and the relatively small number of men with muscle weakness or low physical performance, which raises concerns about the reliability and generalisability of the ROC-derived cut-offs.

Response 2: We appreciate this important methodological observation. We agree that the use of non-probabilistic convenience sampling and the sex imbalance represent limitations that may affect the representativeness of the sample and the stability of sex-stratified ROC estimates. To address this point, we have expanded the Limitations section to explicitly acknowledge these issues and clarify that the ROC-derived cut-off values, particularly in men, should be interpreted cautiously and considered preliminary. These modifications have been incorporated in the revised manuscript (Discussion, Limitations section).

 

Comment 3: The paper would benefit from a more explicit hypothesis.

Response 3: We thank the reviewer for this suggestion. In response, we have added an explicit hypothesis at the end of the Introduction. Specifically, we now state that we hypothesized that lean mass and bone mineral content, particularly when assessed at the segmental level using DXA, would demonstrate greater discriminative capacity for identifying muscle weakness than adiposity-related variables, while their ability to identify low physical performance would be more limited. This addition strengthens the conceptual framing of the study and clarifies the expected direction of the associations.

 

Comment 4: The novelty of the study is not clearly highlighted in relation to previous studies using segmental DXA.

Response 4: We appreciate this comment and agree that the novelty of the study required clearer articulation. Therefore, we revised the Introduction to more explicitly highlight the contribution of our work relative to previous studies. In particular, we clarified that while prior studies have examined associations between body composition and functional outcomes, few have simultaneously evaluated total and segmental DXA-derived variables—including bone mineral content—using sex-stratified ROC analyses to determine their discriminative capacity for both muscle weakness and low physical performance.

 

Comment 5: The literature review is somewhat narrow and could include more studies that have used ROC analyses to derive diagnostic cut-offs for body composition variables.

Response 5: We sincerely thank the reviewer for this insightful comment. We agree that contextualizing our findings within the broader literature of ROC-derived diagnostic cut-offs significantly strengthens the manuscript and better justifies our methodological approach. In response to this suggestion, we have expanded our literature review and discussion by incorporating recent relevant studies.

Specifically, we have made the following additions to the manuscript:

• In the Introduction: We added a paragraph highlighting the scarcity of studies utilizing ROC curve analyses to derive specific diagnostic cut-offs for regional body composition variables, particularly bone mineral content (BMC). This addition clarifies the gap in the literature that our study aims to address.

• In the Discussion: We enriched this section by comparing our diagnostic accuracy results directly with other studies employing ROC analyses. We discussed how our high AUC values for lean mass and BMC (>0.80 in men) align with current evidence, and how the limited predictive capacity of adiposity found in our models is corroborated by similar recent studies that emphasize structural tissue quantity (muscle and bone) as a more reliable clinical marker.

 

Comment 6:  The use of non-probabilistic sampling and the strong sex imbalance should be clarified in the Methods.

Response 6: We appreciate this observation and agree that greater clarity was needed. The Methods section has been revised to explicitly acknowledge the non-probabilistic nature of the sampling strategy and to clarify that the observed sex imbalance reflects the typical demographic composition of community senior centers in the region, where female participation is substantially higher than male participation. This explanation has been incorporated in the Study Design and Participants subsection.

 

Comment 7: The normality assessment is a concern and the authors should clarify whether any variables were clearly non-normally distributed.

Response 7: We thank the reviewer for this observation. To provide a more objective assessment of the distributional assumptions, we repeated the normality analysis using the Kolmogorov–Smirnov test. The results confirmed that none of the variables showed significant deviations from normality. This information has been incorporated into the Statistical Analysis subsection of the Methods.

 

Comment 8: The interpretation of AUC values in men may be unstable due to the small number of cases.

Response 8: We appreciate this important observation. In response, we revised the Discussion to emphasize that the relatively small number of male cases may lead to instability in AUC estimates and the derived cut-off points. Therefore, these values should be interpreted as preliminary findings that require confirmation in larger samples.

 

Comment 9: The specificity of some cut-offs, such as total BMC for low physical performance in men, appears clinically suboptimal.

Response 9: We thank the reviewer for highlighting this issue. The Discussion has been revised to clarify the clinical implications of the sensitivity–specificity trade-off observed in some ROC-derived thresholds. We now explicitly acknowledge that some cut-offs, despite showing acceptable sensitivity, may have limited clinical applicability due to relatively low specificity.

 

Comment 10: The sex differences discussion is somewhat speculative.

Response 10: We appreciate this comment and agree that the interpretation of sex differences should be presented cautiously. The Discussion has been revised to adopt more circumspect language, framing these interpretations as potential explanations or hypotheses rather than definitive conclusions supported by the present data.

 

Comment 11: There are formatting issues in the reference list and some minor problems with the abstract, figures, and tables.

Response 11: We thank the reviewer for identifying these issues. The manuscript has been carefully revised to correct formatting problems in the reference list, including the separation of references 22 and 23. In addition, the abstract has been corrected, figure legends have been revised, and table formatting has been improved to enhance clarity and readability.

Reviewer 2 Report

Comments and Suggestions for Authors

The study is based on a cross-sectional design with non-probabilistic convenience sampling, recruiting participants from community senior centers in Talca, Chile. While the authors acknowledge this limitation, the implications for selection bias and generalizability should be discussed more thoroughly.

The total sample includes 268 participants, of whom 81% are women (218 women vs 50 men). 

This imbalance raises several concerns:

• ROC analyses stratified by sex may be unstable in the male subgroup.

• The number of outcome events in men is small (e.g., 10 cases of low physical performance).

The manuscript relies primarily on ROC curve analyses to evaluate discriminative capacity. While ROC analysis is useful for diagnostic studies, the current approach presents some limitations.

Missing multivariable adjustment

The analysis does not appear to account for potential confounders such as:

• age

• BMI

• comorbidities

• physical activity

• nutritional status

Given that body composition and functional performance are influenced by multiple factors, multivariable regression models (e.g., logistic regression) should be considered to estimate independent associations.

The authors should clarify whether:

• ROC analyses were conducted on unadjusted variables only, or

• adjusted models were explored.

Author Response

Comment 1: The study is based on a cross-sectional design with non-probabilistic convenience sampling, recruiting participants from community senior centers in Talca, Chile. While the authors acknowledge this limitation, the implications for selection bias and generalizability should be discussed more thoroughly.

Response 1: We thank the reviewer for this valuable observation. In response, we expanded the Discussion section to more explicitly address the potential implications of the non-probabilistic convenience sampling strategy. Specifically, we clarified that recruitment from community senior centers may represent a relatively active subgroup of the older population, which could introduce selection bias and limit the generalizability of the findings to the broader population of community-dwelling older adults.

 

Comment 2:  The total sample includes 268 participants, of whom 81% are women (218 women vs 50 men). This imbalance raises several concerns.

Response 2: We thank the reviewer for highlighting this important point. The sex distribution observed in our sample reflects the typical demographic composition of community senior centers, where female participation is substantially higher than male participation. We have clarified this aspect in the Methods section and acknowledged its implications for interpretation in the Discussion.

 

Comment 3:  ROC analyses stratified by sex may be unstable in the male subgroup.

Response 3: We thank the reviewer for this important observation. We agree that the relatively small number of male participants may affect the stability of ROC-derived estimates. In response, we revised the Discussion section to emphasize that the cut-off values obtained in men should be interpreted cautiously due to the limited number of outcome events, which may increase the variability of AUC estimates and affect the stability of the identified thresholds.

 

Comment 4: The number of outcome events in men is small (e.g., 10 cases of low physical performance).

Response 4: We thank the reviewer for this important observation. We have clarified in the Discussion that the number of outcome events in men was relatively small, particularly for low physical performance, which may affect the stability and precision of ROC-derived estimates. Accordingly, these findings should be interpreted with caution.

 

Comment 5: The manuscript relies primarily on ROC curve analyses to evaluate discriminative capacity. While ROC analysis is useful for diagnostic studies, the current approach presents some limitations.

Response 5: We thank the reviewer for this insightful comment. The primary aim of this study was to evaluate the discriminative capacity of body composition variables and to identify potential diagnostic cut-off values for muscle weakness and low physical performance. Accordingly, ROC curve analysis was selected as the most appropriate method to assess diagnostic performance and determine optimal thresholds using the Youden Index. We have clarified this rationale in the revised manuscript.

 

Comment 6: Missing multivariable adjustment. The analysis does not appear to account for potential confounders such as age, BMI, comorbidities, physical activity, and nutritional status.

Response 6: 

We thank the reviewer for this important and thoughtful observation. We agree that body composition and functional performance are influenced by multiple factors, including age, comorbidities, and lifestyle-related variables. However, the primary objective of the present study was to evaluate the discriminative capacity of body composition indicators and to identify potential diagnostic cut-off values for muscle weakness and low physical performance, rather than to estimate independent or causal associations.

In this context, ROC curve analysis was considered the most appropriate methodological approach, as it allows direct assessment of the ability of each variable to discriminate between conditions and to derive clinically interpretable thresholds. The inclusion of multivariable adjustment would shift the analytical focus toward modeling independent associations rather than evaluating diagnostic performance.

Nevertheless, we acknowledge the relevance of potential confounding factors, and this has now been explicitly addressed in the limitations section of the manuscript. Future studies incorporating multivariable approaches may provide additional insights into the independent contributions of these variables.

 

Comment 7: The authors should clarify whether ROC analyses were conducted on unadjusted variables only, or whether adjusted models were explored.

Response 7: We thank the reviewer for this helpful suggestion. We have clarified in the Statistical Analysis subsection that ROC curve analyses were conducted using unadjusted body composition variables, as the aim of the study was to evaluate their direct discriminative capacity for identifying muscle weakness and low physical performance.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript addresses a clinically relevant topic, namely the relationship between body composition compartments and functional outcomes in older adults. The use of DXA-derived segmental measures and sex-stratified analyses is valuable. However, in its current form, the manuscript requires substantial revision before it can be considered for publication. My main concerns relate to the analytical strategy, the interpretation of ROC-based findings, the strength of the conclusions, and several issues in reporting clarity.

First, the study is cross-sectional and based on a convenience sample with a marked sex imbalance (50 men vs. 218 women). This limitation is especially important because the most emphasized findings are the ROC-derived cut-off points in men, yet the number of male outcome events is very small. The manuscript itself states that only 16 men had muscle weakness and 10 had low physical performance. Under these conditions, AUC estimates, confidence intervals, and optimal thresholds derived from the Youden index are likely to be unstable. The cut-off values proposed for men should therefore be presented much more cautiously and explicitly framed as exploratory rather than clinically applicable thresholds.

Second, the manuscript relies almost exclusively on univariable ROC analyses. This is a major methodological limitation. Body composition variables are strongly correlated with sex, body size, height, weight, and age, and likely with each other. Without multivariable modeling or at least partial adjustment, it is difficult to determine whether the reported discrimination reflects true functional relevance or simply body size-related separation. The authors should consider adding regression models adjusted for key covariates, or at minimum provide a more explicit justification for why unadjusted ROC analysis is sufficient for the study objective.

Third, the framing of “diagnostic accuracy” may be overstated. The manuscript uses terms such as “predictive performance,” “diagnostic performance,” and “identify” in a way that may suggest clinical screening utility. However, this is not a diagnostic study against an external gold standard; it is a cross-sectional discrimination analysis using concurrent functional classifications. The language throughout the manuscript should be toned down. Terms such as “discriminative capacity” are more appropriate than “diagnostic accuracy” or “predictive ability,” especially given the absence of validation.

Fourth, the conclusions are broader than the data justify. The statement that adiposity is not a relevant determinant of functional decline is too strong. What the data support is that the adiposity measures used here had limited discriminatory performance for the chosen binary outcomes in this sample. That is not equivalent to demonstrating a lack of pathophysiological relevance of adiposity. The discussion should distinguish more carefully between mechanistic relevance and ROC-based discrimination.

Fifth, more detail is needed regarding the statistical workflow. The authors state that normality was assessed visually and that parametric tests were assumed to be robust because of sample size. However, the male subgroup is relatively small, and the event counts are smaller still. It would strengthen the manuscript to report whether any sensitivity analyses were performed, whether AUCs were compared statistically, and whether internal validation was attempted. At minimum, the limitations of cut-off derivation without validation should be discussed more explicitly.

Sixth, the presentation of the results can be improved. Table 2 is central to the paper, but it selectively presents only the best-performing variables rather than the full panel of tested predictors. This creates the impression of partial reporting. The authors should clarify whether all body composition variables were screened and only the top performers are shown, or provide the complete results in a supplementary table. This is important for transparency and to reduce the risk of overinterpretation.

Seventh, the figures should be improved. The narrative refers to separation of distributions and ROC performance, but the visual contribution of the figures appears limited relative to the central analytical message. The authors should ensure that all figures are sufficiently legible, include complete axis labeling and units, and add sample sizes per subgroup where relevant. If space is limited, some figure content may be better moved to supplementary material.

Eighth, the introduction and discussion would benefit from sharper positioning of the manuscript’s novelty. The general association between lean mass, bone-related measures, and physical function in older adults is already well established. The genuinely useful angle here seems to be the segmental DXA approach and the sex-stratified discrimination analysis. That specific contribution should be articulated more clearly, while avoiding claims that exceed the incremental novelty of the work.

Finally, the English is generally understandable but should be edited for precision and concision. There are several passages where the prose is repetitive, where causal wording is stronger than warranted by the design, or where the manuscript alternates between descriptive and clinically interpretive language too abruptly.

In summary, the study has merit, but the manuscript requires major revision. The authors should substantially strengthen the analytical justification, temper the interpretation of ROC-derived cut-offs, improve transparency of reporting, and align the conclusions more closely with what the data can actually support.

Author Response

Comment 1: The study is cross-sectional with convenience sampling and a strong sex imbalance. ROC cut-offs in men are based on very few events and should be interpreted as exploratory.

Response 1: We thank the reviewer for this important observation. We agree that the relatively small number of male participants and outcome events may affect the stability of ROC-derived estimates. In response, we have revised the Discussion to explicitly emphasize that the cut-off values identified in men should be interpreted cautiously and considered exploratory rather than clinically definitive. We have also reinforced the limitations related to sampling strategy and generalizability.

 

Comment 2: The manuscript relies on univariable ROC analyses without adjustment for confounders.

Response 2: We thank the reviewer for this thoughtful comment. The primary objective of the present study was to evaluate the discriminative capacity of body composition variables and to identify potential cut-off values for muscle weakness and low physical performance, rather than to estimate independent or causal associations. Accordingly, ROC curve analysis using unadjusted variables was considered the most appropriate approach to directly assess discrimination. We acknowledge that body composition is influenced by multiple factors, including age, body size, and clinical variables, and this has now been explicitly addressed in the limitations section. Future studies incorporating multivariable models may provide additional insights into independent associations.

 

Comment 3: The framing of “diagnostic accuracy” is overstated.

Response 3: We thank the reviewer for this valuable observation. We agree that the terminology used in the original version may have overstated the clinical implications of the findings. In response, we have revised the manuscript to adopt more precise and appropriate terminology throughout. Specifically, terms such as “diagnostic performance,” “predictive performance,” and “identify” have been replaced with “discriminative capacity,” “discriminative ability,” and “discriminate,” respectively. These changes better reflect the cross-sectional design of the study and avoid implying clinical diagnostic or screening applicability.

 

Comment 4

The conclusions are too strong, particularly regarding adiposity.

Response 4

We appreciate this important comment. We have revised the Discussion and Conclusions to adopt a more cautious interpretation. Specifically, we clarified that adiposity measures showed limited discriminative capacity for the selected outcomes in this sample, rather than suggesting a lack of pathophysiological relevance. This distinction has been explicitly incorporated into the revised manuscript.

 

Comment 5: More detail is needed regarding statistical workflow and validation.

Response 5: We thank the reviewer for this suggestion. We have expanded the Statistical Analysis and Discussion sections to improve transparency regarding the analytical approach. Specifically, we clarified the normality assessment procedures and the use of ROC analysis. In addition, we have explicitly acknowledged in the limitations that the derived cut-off values were not externally or internally validated and should therefore be interpreted as preliminary. Future studies are warranted to validate these findings in independent samples.

 

Comment 6: Figures should be improved and better aligned with the analytical message.

Response 6: We thank the reviewer for this helpful suggestion. We carefully reviewed all figures to ensure clarity, appropriate labeling, and consistency with the analytical message of the manuscript. Axis labels, units, and overall readability were verified and adjusted where necessary. We consider that the figures adequately represent the results and are consistent with the objectives of the study.

 

Comment 7: The novelty of the manuscript should be better articulated.

Response 7: We thank the reviewer for this observation. In response, we have revised the Introduction to more clearly highlight the specific contribution of the study, namely the use of segmental DXA-derived measures and sex-stratified discrimination analysis to evaluate body composition in relation to functional outcomes. We have also refined the discussion to better position the findings within the existing literature.

 

Comment 9: The English should be improved for precision and concision.

Response 9: We thank the reviewer for this suggestion. The manuscript has been carefully revised to improve clarity, precision, and conciseness of the language, and to ensure consistency in terminology throughout the text.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Despite the authors' stated commitment in their response letter to correct a formatting error, references 22 and 23 remain merged into a single line in the revised manuscript: " 22. Tavares NHC, Rodrigues BC, Arruda SP, Szlejf C, Suemoto CK, Griep RH, et al. Untangle the relationship of 534 muscle mass and bone mineral content on handgrip strength: Results of ELSA-Brasil. Ciência & Saúde Coletiva. 535 2023;28:3191-204. https://doi.org/10.1590/1413-812320232811.19372022 23. Shen S, Li J, Guo Q, Zhang W, Wang 536 X, Fu L, et al. Body mass index is associated with physical performance in suburb-dwelling older Chinese: A cross- 537 sectional study. PLoS One. 2015;10(3):e0119914. https://doi.org/10.1371/journal.pone.0119914" The authors are requested to separate these two references in the reference list.

Moreover, reference 21 has also erros: " 21. Shin H, Panton LB, Dutton GR, Ilich JZ. Relationship of physical performance with body composition and bone 531 mineral density in individuals over 60 years of age: a systematic review. Journal of aging research. 532 2011;2011(1):191896. https://doi.org/10.4061/2011/191896.  It should be: Shin, H., Panton, L. B., Dutton, G. R., & Ilich, J. Z. (2011). Relationship of Physical Performance with Body Composition and Bone Mineral Density in Individuals over 60 Years of Age: A Systematic Review. Journal of aging research, 2011, 191896. https://doi.org/10.4061/2011/191896
"

Author Response

Comments 1: Despite the authors' stated commitment in their response letter to correct a formatting error, references 22 and 23 remain merged into a single line in the revised manuscript. The authors are requested to separate these two references in the reference list. Moreover, reference 21 has also errors and should be corrected according to the journal format.

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, we have carefully revised the reference list to correct these formatting issues.

Specifically, references 22 and 23, which were previously merged into a single line due to a formatting error, have now been properly separated into two independent references. 

In addition, reference 21 has been corrected to ensure consistency with the journal’s formatting style, including proper title capitalization, removal of redundant issue numbering, and correct presentation of bibliographic details and DOI. 

Finally, we conducted a thorough review of the entire reference list to ensure consistency and accuracy throughout the manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

Satisfied with changes

Author Response

Thank you!!

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the careful revision. The manuscript has improved substantially and addresses the main concerns raised in the previous review. In particular, the terminology has been appropriately moderated from diagnostic language to discriminative language, the exploratory nature of the ROC-derived cut-offs in men is now clearly acknowledged, and the limitations regarding convenience sampling, sex imbalance, lack of validation, and absence of multivariable adjustment are more transparently discussed.

The revised discussion and conclusions are more balanced, especially regarding adiposity. I also appreciate that the manuscript now avoids overstating clinical applicability and frames the proposed thresholds as preliminary rather than definitive.

At this stage, I believe the manuscript is close to being acceptable, but a few minor issues should still be addressed before final acceptance:

1. Please ensure full consistency in terminology throughout the manuscript. Although “discriminative capacity” has replaced stronger diagnostic wording in most sections, some remaining expressions such as “predictors” or similar clinically loaded terms should be checked and harmonized.

2. The clinical applicability of the proposed cut-offs should remain explicitly limited, especially in men, given the small number of events. This point is already acknowledged in the limitations, but it would be useful to maintain the same caution wherever thresholds are mentioned in the Discussion.

3. The figures are generally acceptable, but they could still be improved in terms of visual clarity and alignment with the main analytical message. In particular, the ROC figures would benefit from maximal readability in the final formatted version.

4. The reference list should undergo one final technical revision for formatting consistency and accuracy. I noted at least some irregularities in punctuation and DOI formatting.

Overall, this is now a more rigorous and appropriately framed exploratory study. The manuscript makes a potentially useful contribution by examining sex-stratified DXA-derived body composition variables in relation to muscle weakness and low physical performance in older adults.

Author Response

Comments 1: Please ensure full consistency in terminology throughout the manuscript. Although “discriminative capacity” has replaced stronger diagnostic wording in most sections, some remaining expressions such as “predictors” or similar clinically loaded terms should be checked and harmonized.

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, we have carefully revised the manuscript to ensure full consistency in terminology.

Specifically, terms with stronger clinical or predictive implications (e.g., “predictors”) have been replaced or harmonized with more appropriate expressions such as “discriminative capacity” or “ability to discriminate,” in accordance with the analytical scope of ROC-based analyses. 

 

Comments 2: The clinical applicability of the proposed cut-offs should remain explicitly limited, especially in men, given the small number of events. This point is already acknowledged in the limitations, but it would be useful to maintain the same caution wherever thresholds are mentioned in the Discussion

Response 2: Thank you for this important observation. We agree with this comment. Therefore, we have reinforced the cautious interpretation of the proposed cut-off points throughout the manuscript.

In addition to the limitations section, we have now explicitly emphasized the exploratory nature and limited clinical applicability of these thresholds wherever they are mentioned in the Discussion, particularly in relation to men, given the relatively small number of events. These clarifications have been incorporated in the Discussion section.

 

Comments 3: The figures are generally acceptable, but they could still be improved in terms of visual clarity and alignment with the main analytical message. In particular, the ROC figures would benefit from maximal readability in the final formatted version.

Response 3: Thank you for this valuable suggestion. We appreciate the reviewer’s interest in improving the visual clarity of the figures.

However, we would like to note that the current figures were positively evaluated by the other reviewers and were designed to prioritize clarity, simplicity, and consistency with the analytical message of the study. In particular, the ROC curves were intentionally presented in a clean and straightforward format to facilitate interpretation without overloading visual elements.

For these reasons, and considering the balance between clarity and simplicity, we have decided to maintain the current figure design in this version of the manuscript.

 

Comments 5: The reference list should undergo one final technical revision for formatting consistency and accuracy. I noted at least some irregularities in punctuation and DOI formatting.

Response 5: Thank you for this observation. We agree with this comment. Therefore, we have conducted a thorough technical revision of the reference list.

All references have been carefully reviewed to ensure consistency in formatting, including punctuation, journal style, and DOI presentation, in accordance with the journal guidelines.