Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Reviewer Comments
Overall, the study is well-designed, uses an appropriate national dataset, and makes a significant contribution to the examination of metabolic syndrome determinants in rural elderly populations. However, the article has some methodological and reporting deficiencies. Firstly, statistical p-values ​​for the results are not given in the abstract section. In the introduction section, existing studies in the literature are not sufficiently discussed, the originality of the study is not clearly demonstrated, and some statements require references. Also, it is not stated that the STROBE checklist was used for the cross-sectional design, and compliance with the STROBE guideline should be clearly stated in the methods section. The incomplete data management should be passively expressed in the methods section, and the statistical analysis program used and the normality assessment of the data should be more clearly stated. The name of the ethics committee, approval date and number are missing. Abbreviations used in the tables should be explained in footnotes, and the table contents should be organized in a simpler and more understandable way. The discussion section, in its current state, remains superficial; the numerous findings of the study should be more comprehensively related to the literature, and the inferences should be structured more strongly. Specific suggestions are as follows.

Revisions
1. Page 1, line 18-27; Statistical p-values ​​should be added to the results section of the abstract.
2. Page 2, lines 46-48; “Yet determinants of MetS in rural older adults remain less well characterized; prior studies vary in measurement and adjustment, limiting comparability.” A source should be added for this information from the literature.
3. Page 2, lines 65-79; The introduction section does not clearly state the studies in the literature and how the current study differs from these studies. More mention should be made of the studies in the literature, and this paragraph should be linked to the purpose of the study to more clearly demonstrate the originality of the study.
4. Page 2, line 81; It is not stated that the STROBE reporting checklist was used for this cross-sectional analysis. The STROBE checklist should be used. In addition, it should be stated in the methods section that STROBE was used, and its source should be added.
5. Page 3, lines 93-96; “We included rural older adults with complete data on all variables required to define MetS and prespecified covariates, yielding a final analytic sample of 467 individuals (unweighted).” The sentence should be expressed using a passive verb.
6. Page 3, line 118; The program used in the statistical analysis should be given at the beginning of the text, and it should also be stated how the normality test of the data was performed and whether the data did not have a normal distribution.
7. Page 3, line 130; The name of the ethics committee, the approval date and number should be added.
8. All abbreviations in the tables should be explained in footnotes. Also, the variable descriptions in the table appear confusing; they should be expressed more clearly and simply.
9. Page 6, line 212; The discussion is somewhat superficial; the study has many results, and these results should be related to much more research in the literature, and the discussion should be revised by making inferences.

Comments on the Quality of English Language

While the article's language quality is generally good and fluent, some sections require minor linguistic adjustments to improve clarity of expression and enhance academic style.

Author Response

We sincerely thank the reviewer for the thorough and constructive evaluation of our manuscript and for recognizing the significance of our study using a nationally representative dataset to examine determinants of metabolic syndrome among rural older adults. We carefully addressed all methodological and reporting issues raised. Specifically, we revised the abstract to include statistical p-values, strengthened the Introduction to better situate our study within existing literature and clarify its originality, explicitly stated compliance with the STROBE guideline, improved clarity in the Methods section regarding statistical software, normality considerations, and data handling, and added complete ethics committee information. In addition, all tables were revised to improve readability and consistency, with abbreviations fully explained in footnotes, and the Discussion section was substantially expanded to more comprehensively interpret our findings in relation to prior research. All revisions have been highlighted in the revised manuscript with page and line numbers indicated below.

Comment 1: Statistical p-values should be added to the results section of the abstract.

Response: Thank you for this suggestion. We revised the Results section of the abstract to include statistical p-values for the key predictors identified in the fully adjusted complex-sample logistic regression model. Design-based p-values are now reported alongside adjusted odds ratios and 95% confidence intervals for body mass index ≥25 kg/m², HbA1c ≥7.0%, vitamin D deficiency (<20 ng/mL), and meeting the WHO physical activity guideline. Revised in the Abstract, Results section (Page 1, Lines 23–27).

Comment 2: A source should be added for the statement regarding limited characterization and comparability of MetS determinants in rural older adults.

Response: We agree and have added appropriate references to support this statement. The revised Introduction now cites prior studies documenting variability in measurement approaches and analytic strategies in studies of metabolic syndrome, as well as evidence highlighting gaps in research focusing specifically on rural older adult populations. These citations strengthen the rationale for the present study. Revised in the Abstract, Results section (Lines 46–50).

 

Comment 3: The Introduction does not clearly demonstrate how the current study differs from previous studies.

We revised the Introduction to more clearly synthesize prior literature and explicitly articulate the originality of our study. The revised text highlights gaps in previous research, including limited rural-specific analyses and inconsistent adjustment for relevant covariates, and clarifies how our study addresses these gaps by applying a complex-sample design to nationally representative KNHANES 2023 data restricted to rural adults aged ≥65 years. Revised in the Introduction (Lines 67–78).

Comment 4: The use of the STROBE checklist should be stated in the Methods section.

Response: Thank you for this important point. We have now explicitly stated in the Methods section that this cross-sectional study was reported in accordance with the STROBE guideline. Revised in the Methods section (Lines 88-96).

Comment 5: The sentence describing the analytic sample should be expressed in the passive voice.

Response: As suggested, the sentence has been revised to the passive voice. It now states that rural older adults with complete data on all required variables were included in the final analytic sample. Revised in the Methods section (Page X, Lines 100–103).

Comment 6: The statistical software and normality assessment should be stated more clearly.

Response: We revised the Statistical Analysis section to specify the statistical software (IBM SPSS Statistics, version 30.0) at the beginning of the section. We also clarified our approach to data distribution assessment and noted that design-based statistical procedures were used in accordance with the complex survey design. Revised in the Methods section (Lines 126–135).

Comment 7: The name of the ethics committee, approval date, and approval number are missing.

Response: We have added complete ethical approval information to the manuscript. The Ethical Considerations subsection now specifies the name of the reviewing board (Daejin University Institutional Review Board), the approval/exemption number (Approval No. 1040656-202512-HR-02-15), and the approval date. The initial submission blinded this information for peer review, which has now been corrected in the revised version. Revised in the Ethical Considerations subsection (Lines 137–144).

Comment 8: All abbreviations in the tables should be explained, and table contents simplified.

Response: All tables were revised to improve clarity and consistency. Abbreviations are now fully explained in the footnotes of each table, variable labels were simplified to enhance readability, and statistical tests are clearly indicated. Revised in Tables 1–4 and corresponding footnotes (Lines 158–216).

Comment 9: The Discussion section is superficial and should be strengthened.

Response: We substantially revised and expanded the Discussion section to provide a deeper interpretation of our findings. Each major result is now discussed in relation to relevant literature, with consideration of plausible mechanisms, public health implications for rural older adults, and study limitations. Revised in the Discussion section (Lines 236–306).

Reviewer 2 Report

Comments and Suggestions for Authors

Overall Assessment

The manuscript "Determinants of metabolic syndrome among rural older adults: A cross-sectional analysis of the 2023 Korea National Health and Nutrition Examination Survey" addresses a highly relevant public health problem, especially in the context of population aging and rural inequalities.

The use of recent data (KNHANES 2023), together with an analysis that respects the complex design of the survey, is an important strength.

The article shows internal consistency, I find it from the objectives, methods, results and discussion. It highlights clinical results, integrates recent evidence, addresses the rural context in a pertinent way, and presents practical and public policy implications.

 

Opportunities for improvement

Although the manuscript mentions the limitation of the cross-sectional design, I consider explicitly reinforcing the risk of reverse causality, especially in: physical activity and metabolic syndrome, as well as vitamin D and obesity/lifestyle. To improve, it could be said that:

"Reverse causality cannot be ruled out, as people with metabolic syndrome may have lower levels of physical activity or less sun exposure due to functional limitations."

In relation to the use of complete-case analysis, there may be selection bias, particularly in the elderly population. Therefore, it should be indicated what percentage of participants were excluded due to missing data. This bias can be recognized in the limitations section.

It is important to clarify that while an HbA1c ≥7.0% is clinically relevant, it is not a standard criterion for metabolic syndrome. Therefore, its use as an indicator of suboptimal glycemic control can be justified.

 

It should also be mentioned that other cut-off points (≥6.5%) could be explored in future or sensitivity analyses.

Although the manuscript reports FIV <2 (without statistical collinearity): I recommend clarifying in the discussion that BMI is not a direct part of the definition of metabolic syndrome and was maintained in the model due to its clinical and public health relevance.

Other improvements

Homogenize the use of “older adults” and “older adults aged ≥65 years”.

Reduce repetitions about physical activity and vitamin D in the discussion.

On the Boards

Add the unweighted n in the table headers.

It should be clarified that the percentages may not add up to 100% by rounding.

Regarding references

The reference to WHO guidelines for physical activity appears duplicated (refs. 15 and 22).

Author Response

Thank you for your constructive and insightful review. You highlighted important strengths of our manuscript, including the use of recent KNHANES 2023 data and the application of complex-sample methods, and you provided valuable suggestions to improve clarity and interpretability.

We sincerely appreciate the reviewer’s positive evaluation and thoughtful recommendations. In response, we revised the manuscript to (1) more explicitly acknowledge the possibility of reverse causality inherent to the cross-sectional design, particularly for physical activity and vitamin D status; (2) clarify the implications of complete-case analysis and address potential selection bias; (3) emphasize that HbA1c ≥7.0% is not a diagnostic criterion for metabolic syndrome and justify its use as a clinically meaningful indicator of suboptimal glycemic control, while noting alternative thresholds (e.g., ≥6.5%) for future sensitivity analyses; (4) clarify that BMI is not a MetS diagnostic component and explain its inclusion due to clinical and public health relevance; (5) standardize terminology related to older adults throughout the manuscript; and (6) improve table presentation (including adding unweighted n in headers, rounding notes) and correct reference duplication. These revisions strengthen methodological transparency and improve the readability and scientific rigor of the manuscript.

Comment 1: Reverse causality should be more explicitly addressed (physical activity/vitamin D and MetS).

Response 1: We agree and have strengthened the Limitations section to more explicitly acknowledge the risk of reverse causality inherent to the cross-sectional design. We now state that reverse causality cannot be ruled out because individuals with MetS may have lower physical activity levels or less sun exposure due to comorbidity burden or functional limitations, which may also contribute to lower 25(OH)D levels. Revised in the Discussion–Limitations (Lines 286–287).

Comment 2: Complete-case analysis may introduce selection bias; please report the proportion excluded due to missing data and acknowledge this limitation.

Response 2: We agree. In the Methods (Study population), we now report the extent of missing data handled via complete-case analysis. Among rural older adults aged ≥65 years (unweighted n=527), the final analytic sample included participants with complete data (unweighted n=467); therefore, 60 participants (11.4%) were excluded due to missing values. We also acknowledge the potential for selection bias in the Limitations, noting that excluded participants may have differed systematically from those included. Revised in the Methods (Section 2.2) (Lines 103–104) and Discussion–Limitations.

Comment 3: Please clarify that HbA1c ≥7.0% is clinically relevant but not a standard metabolic syndrome criterion, justify its use as an indicator of suboptimal glycemic control, and note that alternative cut-offs (e.g., ≥6.5%) could be explored in sensitivity analyses.

Response 3: We agree. We clarified in the Discussion that HbA1c ≥7.0% is not a diagnostic criterion for metabolic syndrome and should be interpreted as a clinically meaningful indicator of suboptimal glycemic control rather than part of the MetS case definition. We additionally noted in the Limitations/Future Directions that alternative HbA1c thresholds (e.g., ≥6.5%) could be evaluated in future sensitivity analyses to assess robustness. Revised in the Discussion (Sections 4.2 and 4.4) (Lines 299–301).

Comment 4: Please clarify that BMI is not part of the definition of metabolic syndrome and explain why BMI was retained in the model despite low VIF. Response 4: We agree. We clarified in the Discussion that BMI is not a diagnostic component of metabolic syndrome. We further explained that BMI was retained due to its clinical and public health relevance as a pragmatic indicator of overall adiposity that complements waist circumference–based central adiposity for risk stratification. (Our multicollinearity screening also indicated no problematic collinearity; all VIF values were <2.0.) Revised in the Discussion–Limitations (Section 4.4) (Lines 296–298).

Comment 5: Please homogenize terminology (“older adults” vs “older adults aged ≥65 years”) and reduce repetitions about physical activity and vitamin D in the Discussion.

Response 5: We agree. We standardized terminology throughout the manuscript by defining the study population as adults aged ≥65 years at first mention and using “older adults” consistently thereafter for readability. We also edited the Discussion to reduce repetition by streamlining overlapping statements regarding physical activity and vitamin D while retaining the core interpretation and implications. Revised throughout the manuscript (Lines 100–104, 268-275).

Comment 6: Tables: add unweighted n in headers; clarify that percentages may not sum to 100% due to rounding; references: WHO physical activity guideline appears duplicated.

Response 6: We agree and revised the tables and references accordingly. We added the unweighted sample size (n) to the headers of Tables 1–4 and included a footnote stating that percentages may not sum to 100% due to rounding. We also corrected the reference list by removing duplicate entries for the WHO physical activity guideline and retaining a single primary reference. Revised in Tables 1–4 and the Reference list.

 

Reviewer 3 Report

Comments and Suggestions for Authors It is necessary to discriminate by sex in the comparison results because it demonstrates better research experience and also presents results that are much more consistent with investigation.

The interest in working with this type of population is highlighted; however, it is necessary to complement the information with gender discrimination and also analyze the results regarding this difference. It is recommended to complement the methodological process with comparative characteristics in the research.

 

Author Response

Comment 1: The reviewer recommends sex-disaggregated analyses and examination of gender differences in the comparison results. Response 1: Thank you for this valuable suggestion. We carefully examined potential sex differences in our study. As shown in Table 1, the distribution of sex was already very similar between participants with and without metabolic syndrome (male: 49.0% vs. 49.7%; p = 0.893), indicating no meaningful baseline difference by sex. In response to the reviewer’s comment, we further strengthened the analysis by conducting additional sex-stratified analyses and testing sex-by-exposure interaction terms within the complex-sample logistic regression framework. These analyses demonstrated similar patterns of association in men and women, and no statistically significant sex-by-exposure interactions were observed. Accordingly, the Statistical Analysis and Results sections were revised to explicitly describe these additional analyses and findings, which confirm that the main results are consistent across sexes in this rural older-adult population. Revised in the Statistical Analysis and Results sections (Lines 136-138, 217-218).

Reviewer 4 Report

Comments and Suggestions for Authors

This manuscript examines determinants of metabolic syndrome (MetS) among rural Korean adults aged ≥65 years using the 2023 KNHANES and appropriately applies complex-sample survey methods. The topic is timely and relevant because  population aging, rural health disparities, and the growing burden of cardiometabolic disease all are a growing concern all around the world. It is well written manuscript with harmonized definitions, transparent analytic procedures, and results that are largely consistent with existing literature. Overall, the study adds good insights and evidence specific to rural older adults in Korea.

It is good that they have chosen National level data with good survey methods and data collection. The methods section is well written and easy to understand for a reader.

Restricting the analysis to rural older adults addresses an important gap, as determinants of MetS may differ from urban or mixed populations. 

Tables are informative, clearly labeled, and logically sequenced. The Discussion appropriately contextualizes findings with biologic plausibility and public health relevance.

One concern is the possible conceptual overlap between predictors and outcome. For example BMI, HbA1c, triglycerides, and HDL-C are closely related to MetS diagnostic components. Although BMI is not itself a MetS criterion, its very strong association (aOR ≈ 9) may partially reflect collinearity with central adiposity and glycemic status.

A sensitivity analysis excluding say waist circumference or related proxies (or like  modeling BMI continuously) could increase the confidence in their findings.

It would be good to address any missing data among the datasets 

Although the authors emphasizes rural setting, it would be interesting to examine the link with some key rural determinants (like food habits, physical activity, occupation, access to medical care) 

Grammer and spell check required. Overall it is a well written and thoughtfully designed and executed study and I would recommend for publication 

 

 

Author Response

We sincerely thank the reviewer for the positive and constructive evaluation of our manuscript. We appreciate the recognition of the study’s relevance, methodological rigor, and contribution to understanding metabolic syndrome among rural older adults using nationally representative data. In response to the reviewer’s valuable suggestions, we have clarified issues related to conceptual overlap between predictors and outcome, addressed missing data handling, strengthened the discussion of rural-specific determinants within the limits of available variables, and carefully revised the manuscript for grammar and clarity. Detailed responses to each comment are provided below.

Comment 1. Conceptual overlap between predictors and MetS components (BMI, HbA1c, lipids)
Response: We appreciate this important methodological comment. We acknowledge the potential conceptual overlap between certain predictors and metabolic syndrome pathways. As clarified in the revised manuscript, BMI is not a diagnostic component of metabolic syndrome and was retained as a clinically meaningful indicator of overall adiposity with strong public health relevance. HbA1c was included as an indicator of suboptimal glycemic control rather than as a diagnostic criterion for MetS. Multicollinearity was assessed and variance inflation factors were below accepted thresholds. We have expanded the Discussion to explicitly address this issue and to justify the inclusion of BMI in the multivariable model. In addition, we now highlight the need for future sensitivity analyses (e.g., alternative HbA1c cut-offs, continuous BMI modeling) to further assess robustness. Revised in the Discussion sections (Lines 298–302).

Comment 2. Sensitivity analysis suggestion (BMI/WC, continuous modeling)

Response: We agree that sensitivity analyses could further strengthen causal interpretation. However, given the cross-sectional nature of the study and the limited sample size of rural older adults, we prioritized a parsimonious model aligned with clinical practice. We have therefore acknowledged this point in the Discussion and suggested alternative modeling strategies (e.g., continuous BMI, alternative glycemic thresholds) as important directions for future research. Revised in the Discussion sections (Lines 304–309).

Comment 3. Missing data

Response: We thank the reviewer for this important point. We have clarified the handling of missing data in the Methods section and explicitly reported the proportion of participants excluded due to missing values. We also acknowledge the possibility of selection bias related to complete-case analysis in the Limitations section. Revised in the Methods - Study population and Discussion sections (Lines 100–104, 295-298).

Comment 4. Rural determinants (diet, occupation, access to care)

Response: We agree that rural-specific determinants such as diet, occupation, and access to care are highly relevant. However, the available KNHANES variables limited detailed assessment of these domains within the rural subsample. To address this, we expanded the Discussion to contextualize our findings within known rural structural constraints, including transportation barriers, limited access to preventive services, and environmental influences on physical activity and diet. We also highlight these factors as priorities for future research. Revised in the Discussion sections (Lines 267–288).

Comment 5. Grammar and spelling check

Response: The manuscript was carefully reviewed and edited for grammar, spelling, and clarity throughout. Minor language revisions were made to improve readability and precision.