Grip Strength, Fall Efficacy, and Balance Confidence as Associated Factors with Fall Risk in Middle-Aged and Older Adults Living in the Community

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review your manuscript. Your study has several strengths I would like to comment on - The introduction and background demonstrate significance. The literature reviewed is relevant and up to date. The research question is relevant to fall prevention in community dwelling active older adults. The methods
are described in sufficient detail to allow for replication and the sample size of 280 provides adequate data for analysis. The analysis is ppropriate given the data.The conclusions are well stated and linked to the original research questions. I do have recommendations to strengthen your manuscript. 

1. Your results rely on the FES-1 and the Activities Specific Balance Confidence Scale, both of which are frequently utilized as efficacy measures in fall prevention studies. Can you add the psychometric properties of the instruments to your description?

2. It is well established that falls efficacy measures should be used with physical measures to understand perceived and actual abilities. Can you explain why you chose grip strength instead other meaures of ability such as  Timed Up and Go or other balance and gait measures?

3. Psychological measures are well established predictors of falls in the literature. Please further explain how your study makes a contribution to the science of fall prevention.

Author Response

Review Report 1

1. Your results rely on the FES-1 and the Activities Specific Balance Confidence Scale, both of which are frequently utilized as efficacy measures in fall prevention studies. Can you add the psychometric properties of the instruments to your description?

Thank you for this helpful suggestion. We have now included a description of the psychometric properties of both the Falls Efficacy Scale-International (FES-I) and the Activities-specific Balance Confidence (ABC) Scale in the Methods section. Specifically, we have added information regarding their reliability, validity, and internal consistency as supported by previous studies.

2. It is well established that falls efficacy measures should be used with physical measures to understand perceived and actual abilities. Can you explain why you chose grip strength instead other meaures of ability such as Timed Up and Go or other balance and gait measures?

Thank you for this important observation. We agree that combining falls efficacy measures with objective physical assessments is critical to capturing both perceived and actual physical abilities.

We chose to include grip strength in our study for several reasons. First, grip strength is a well-established indicator of overall muscle strength and functional status in older adults and is strongly associated with adverse outcomes such as disability, frailty, and falls. Unlike performance-based tests such as the Timed Up and Go (TUG) or gait assessments, grip strength is quick, non-invasive, requires minimal space or equipment, and is feasible to administer in community settings and large-scale studies.

Additionally, research has shown that grip strength correlates with lower limb strength and postural control, making it a valuable proxy when more comprehensive mobility tests are not available. We acknowledge, however, that including additional functional measures such as TUG or balance tests could provide further insight, and we highlight this as a limitation in the revised Discussion section.

3. Psychological measures are well established predictors of falls in the literature. Please further explain how your study makes a contribution to the science of fall prevention.

First, while many studies focus exclusively on older adults (typically aged 65 and above), our sample includes both middle-aged and older adults, offering a broader perspective on fall risk across the aging continuum. This approach highlights the importance of early identification of psychological and physical fall risk factors before significant physical decline occurs.

Second, by examining the combined predictive value of grip strength, falls efficacy (FES-I), and balance confidence (ABC Scale), we provide insight into how these variables interact and contribute uniquely to fall risk. This multidimensional approach can inform more tailored and effective fall prevention strategies.

Finally, our findings support the integration of simple, scalable tools—such as grip strength testing and self-report psychological scales—in community and primary care settings. This has practical implications for early screening and intervention in populations not yet considered at high risk.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript investigates the relationship between handgrip strength (HGS), fall efficacy, and balance confidence in predicting fall risk among physically active older adults. The study is relevant and addresses an important issue in geriatric health. It is well-structured, and the writing is generally clear. The use of validated tools (FES-I, ABC Scale) and objective measurement of HGS strengthens the methodology. However, several areas require clarification, improvement, or expansion for the manuscript to be considered for publication.

The title accurately reflects the content. The abstract clearly defines the aim, methods, main findings, and conclusion.

Abstract line 36: Reword “Fall efficacy showed a positive correlation with fall risk (OR = 3.37…)” – technically, odds ratios are derived from regression, not correlation.

Emphasize in the abstract that the sample includes physically active individuals, which may limit generalizability.

The introduction is comprehensive, well-cited, and logically builds the rationale for the study. The integration of physical and psychological risk factors is appropriate and timely.

Lines 47–48: The phrase “The During the aging process…” contains a grammatical error.

Lines 75–77: The psychological section could benefit from clearer articulation about why self-efficacy is particularly relevant in active populations.

Add a hypothesis or set of expected outcomes at the end of the introduction to enhance clarity.

The use of validated scales in Portuguese, clear eligibility criteria, and ethical compliance is commendable. As a cross-sectional study, the use of terms like “predictors” and “prediction” throughout the paper should be re-evaluated, as no temporal causality can be inferred.

Additional descriptive statistics (e.g., distribution of exercise type or frequency) could provide important context. It is unclear whether multicollinearity was tested among predictors in the logistic regression. No justification is provided for dichotomizing the fall variable as “any fall vs. none” — a multinomial or ordinal regression might better reflect the range of fall experiences (e.g., single vs. multiple falls).

Confidence interval for balance confidence (Table 5) is suspiciously reported as “0.93–0.071”, which seems to be a typo.

Tables are well-organized making clear distinction between male and female participants. Use of Spearman correlation is suitable.

Table 3: Rename “Grip Strenght” to “Grip Strength”.

Table 5: Clarify that the logistic regressions were univariate and provide model fit statistics (e.g., Nagelkerke R², Hosmer-Lemeshow test).

The fall rate of 26.4% is substantial; explore in more depth how this compares to literature on physically active older adults.

The manuscript provides a thoughtful discussion of the psychological predictors and aligns with prior literature. Well-cited. The discussion repeatedly refers to "predictors" despite the cross-sectional nature of the study. Consideration of potential selection bias (motivated, healthier older adults) is not addressed. The discussion on “ceiling effects” and functional reserve could benefit from more data or stratified analysis (e.g., HGS tertiles).

The conclusion appropriately summarizes key findings and their implications. Reframe causal language to better reflect cross-sectional design (e.g., replace “predict” with “are associated with”).

Comments on the Quality of English Language

The language quality is generally good, but some grammatical and typographic errors are present (e.g., repeated use of comma instead of decimal point in statistical notation). Also, ensure consistent use of terminology for “fall efficacy” and “fear of falling” — clarify that “fall efficacy” refers to perceived capability despite the risk, and is conceptually the opposite of “fear of falling”. Figures/Tables: Tables are informative, but one graphical summary (e.g., a forest plot or path diagram) could aid interpretation.

Author Response

Review Report 2

This manuscript investigates the relationship between handgrip strength (HGS), fall efficacy, and balance confidence in predicting fall risk among physically active older adults. The study is relevant and addresses an important issue in geriatric health. It is well-structured, and the writing is generally clear. The use of validated tools (FES-I, ABC Scale) and objective measurement of HGS strengthens the methodology. However, several areas require clarification, improvement, or expansion for the manuscript to be considered for publication.

The title accurately reflects the content. The abstract clearly defines the aim, methods, main findings, and conclusion.

Abstract line 36: Reword “Fall efficacy showed a positive correlation with fall risk (OR = 3.37…)” – technically, odds ratios are derived from regression, not correlation.

Thank you for pointing this out. We agree that using the term “correlation” in reference to odds ratios is imprecise, as OR values are derived from logistic regression rather than correlation analysis. We have revised the sentence in the abstract accordingly to reflect the correct statistical interpretation.

Emphasize in the abstract that the sample includes physically active individuals, which may limit generalizability.

We have revised the abstract to clearly state that participants were physically active individuals, and we have acknowledged that this may limit the applicability of the results to more sedentary or frail populations.

The introduction is comprehensive, well-cited, and logically builds the rationale for the study. The integration of physical and psychological risk factors is appropriate and timely.

Lines 47–48: The phrase “The During the aging process…” contains a grammatical error.

We have corrected the phrase by removing the duplicated article to improve clarity and grammar.

Lines 75–77: The psychological section could benefit from clearer articulation about why self-efficacy is particularly relevant in active populations.

Add a hypothesis or set of expected outcomes at the end of the introduction to enhance clarity.

Thank you for this constructive suggestion. We have revised the psychological section of the introduction to more clearly explain why self-efficacy is particularly relevant in physically active individuals. Specifically, we highlight that even in active populations, reduced confidence or fear of falling can lead to self-imposed limitations in movement, avoidance behaviors, and a decline in overall function—thus representing an important psychological dimension of fall risk. Additionally, we have now added a clear hypothesis at the end of the introduction to guide the reader and clarify the expected direction of associations between variables.

The use of validated scales in Portuguese, clear eligibility criteria, and ethical compliance is commendable. As a cross-sectional study, the use of terms like “predictors” and “prediction” throughout the paper should be re-evaluated, as no temporal causality can be inferred.

We agree that the terms “predictors” and “prediction” imply temporal sequencing and potential causality, which are not appropriate in the context of a cross-sectional design. In response, we have reviewed the manuscript thoroughly and replaced such terminology with more appropriate alternatives, such as “associated factors,” “correlates,” or “variables associated with fall risk,” depending on the context.

Additional descriptive statistics (e.g., distribution of exercise type or frequency) could provide important context. It is unclear whether multicollinearity was tested among predictors in the logistic regression. No justification is provided for dichotomizing the fall variable as “any fall vs. none” — a multinomial or ordinal regression might better reflect the range of fall experiences (e.g., single vs. multiple falls).

We agree that information on the type and frequency of exercise would offer valuable context for interpreting our findings. However, we did not collect detailed data on exercise modalities or frequency in this study, as our inclusion criteria were based on regular participation in structured exercise programs verified through program enrollment records.

We have now included a description in the Methods section stating that multicollinearity was assessed using Variance Inflation Factor (VIF) values. All VIF values were below the commonly accepted threshold of 5, indicating that multicollinearity was not a concern among the independent variables.

The fall variable was dichotomized as “any fall” vs. “none” to align with clinical screening practices, where the presence of any fall is often sufficient to prompt further assessment and intervention. However, we recognize that collapsing the data in this way may have reduced the granularity of fall outcomes. We have now addressed this limitation explicitly in the Discussion section and have suggested that future research explore the use of multinomial or ordinal regression models to better capture differences between single and recurrent fallers.

Confidence interval for balance confidence (Table 5) is suspiciously reported as “0.93–0.071”, which seems to be a typo.

Thank you for noting this error. We acknowledge the mistake and have corrected the confidence interval for balance confidence in the revised version of Table 5.

Tables are well-organized making clear distinction between male and female participants. Use of Spearman correlation is suitable.

Table 3: Rename “Grip Strenght” to “Grip Strength”.

Thank you. We acknowledge the mistake.

Table 5: Clarify that the logistic regressions were univariate and provide model fit statistics (e.g., Nagelkerke R², Hosmer-Lemeshow test).

Thank you for this important observation. We clarify that the logistic regression presented in Table 5 was a multivariable model, including grip strength, balance confidence, and fall efficacy as simultaneous predictors of fall risk. We have updated the Methods section and the table caption to make this clear.

The fall rate of 26.4% is substantial; explore in more depth how this compares to literature on physically active older adults.

Thank you for highlighting this point. We agree that the observed fall rate of 26.4% is noteworthy, particularly considering that our sample consists of physically active older adults. In response, we have expanded the Discussion section to compare our findings with existing literature. While physically active older adults tend to show lower fall rates than sedentary individuals, studies still report fall rates ranging from 15% to over 30% depending on age, activity type, and comorbidities. Our findings suggest that regular physical activity does not eliminate fall risk and that psychological factors may play a significant role even in active populations.

The manuscript provides a thoughtful discussion of the psychological predictors and aligns with prior literature. Well-cited. The discussion repeatedly refers to "predictors" despite the cross-sectional nature of the study. Consideration of potential selection bias (motivated, healthier older adults) is not addressed. The discussion on “ceiling effects” and functional reserve could benefit from more data or stratified analysis (e.g., HGS tertiles).

Thank you for your positive feedback and important observations. We have revised the manuscript to replace the term “predictors” with more appropriate terminology, such as “associated factors,” throughout the discussion. We also added a brief statement acknowledging the potential for selection bias due to the inclusion of physically active and likely more motivated participants. Regarding the “ceiling effect,” we agree that further stratified analysis (e.g., by HGS tertiles) could provide additional insight; however, such analyses were beyond the scope of this study. We have noted this in the limitations section as a suggestion for future research.

The conclusion appropriately summarizes key findings and their implications. Reframe causal language to better reflect cross-sectional design (e.g., replace “predict” with “are associated with”).

Thank you. We did it.

Comments on the Quality of English Language

The language quality is generally good, but some grammatical and typographic errors are present (e.g., repeated use of comma instead of decimal point in statistical notation). Also, ensure consistent use of terminology for “fall efficacy” and “fear of falling” — clarify that “fall efficacy” refers to perceived capability despite the risk, and is conceptually the opposite of “fear of falling”. Figures/Tables: Tables are informative, but one graphical summary (e.g., a forest plot or path diagram) could aid interpretation.

Thank you for these helpful observations. We have carefully reviewed the manuscript for grammatical and typographic errors and corrected all instances where commas were incorrectly used instead of decimal points in statistical notation. We have also revised the text to ensure consistent use of the term “fall efficacy,” and clarified its conceptual distinction from “fear of falling” in the Introduction. Regarding the graphical summary, we acknowledge the suggestion but have opted to maintain the current presentation format due to the scope and focus of this manuscript.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This is a potentially useful study about an important problem: falls of older adults. Sample, methods and results are clearly described. However, there are two main problems that must be addressed before this paper can be published. 

• As is mentioned in the Discussion (ll. 305-6), the cross-sectional design implies that only associations can be inferred. Yet, the paper abounds with statements that implicitly assume a causal relationship. This is the case for the frequent use of the word 'predictor', even in the title. Also, the recommendations about the importance of psychological factors for the assessment and understanding of fall risk are valid only if there is a causal relationship. In fact, in this study the causal relationship is likely to go the other way. Fall risk was measured with retrospective questions, while the psychological variables refer to the moment of study. It is quite possible that people who experienced a fall in the recent past became less confident and more apprehensive as a consequence. The paper should acknowledge this possibility, and the statements should be suitably qualified.
• Only people who participated in a community exercise program and who could perform the physical tests were included in the study. This means that people who had bad falls in the recent past with really severe consequences were excluded. This is a source of bias in the sample, which should be acknowledged. 

Other issues:

• l. 132. GmBH means Gesellschaft mit beschränkter Haftung, or private limited company (PLC) in English. So this name seems incomplete; it probably should be Seca GmBH. (I was reminded of a similar mistake that is featured in a novel by Amélie Nothomb, "Fear and Trembling", translated into Portuguese as  "Temor e Tremor".)
• Table 2, the percentage with falls among females should be 28.9%, not 8.9%
• l. 274, "ceiling effect". A comparison with results from the SHARE survey for Portugal shows that while the average grip strength in this study is only a  little lower than it is in the population, the standard deviations and the minimum values suggest that the sample includes some people with quite weak grip strength. This puts doubt on the hypothesis of a ceiling effect.
• ll. 276-8. You write that factors such as polypharmacy and comorbidities play a prominent role. These were measured in your study, so why do you not include them in the analysis of fall risk?

Author Response

Review Report 3

 

This is a potentially useful study about an important problem: falls of older adults. Sample, methods and results are clearly described. However, there are two main problems that must be addressed before this paper can be published. 

• As is mentioned in the Discussion (ll. 305-6), the cross-sectional design implies that only associations can be inferred. Yet, the paper abounds with statements that implicitly assume a causal relationship. This is the case for the frequent use of the word 'predictor', even in the title. Also, the recommendations about the importance of psychological factors for the assessment and understanding of fall risk are valid only if there is a causal relationship. In fact, in this study the causal relationship is likely to go the other way. Fall risk was measured with retrospective questions, while the psychological variables refer to the moment of study. It is quite possible that people who experienced a fall in the recent past became less confident and more apprehensive as a consequence. The paper should acknowledge this possibility, and the statements should be suitably qualified.

Thank you for this insightful and important comment. We agree that, given the cross-sectional nature of the study, causal inferences cannot be made, and the directionality of the observed associations remains uncertain. We have now replaced the term “predictor” throughout the manuscript, including in the title, and adjusted our interpretations to reflect the associative—not causal—nature of the findings. We also acknowledge the possibility of reverse causality, particularly regarding the psychological variables, as fall history was assessed retrospectively while confidence and fall efficacy were measured concurrently. These clarifications have been incorporated into the Discussion and Limitations sections.

 

• Only people who participated in a community exercise program and who could perform the physical tests were included in the study. This means that people who had bad falls in the recent past with really severe consequences were excluded. This is a source of bias in the sample, which should be acknowledged. 

Thank you for this important observation. We agree with this point, and we have already included it in the Limitations section of the manuscript, acknowledging that the exclusion of individuals unable to participate in physical tests may have introduced a bias toward healthier and less functionally impaired participants.

Other issues:

• l. 132. GmBH means Gesellschaft mit beschränkter Haftung, or private limited company (PLC) in English. So this name seems incomplete; it probably should be Seca GmBH. (I was reminded of a similar mistake that is featured in a novel by Amélie Nothomb, "Fear and Trembling", translated into Portuguese as  "Temor e Tremor".)

Thank you for this observation and the thoughtful reference. You are correct—the name was incomplete. We have revised the manuscript to correctly refer to the company as Seca GmbH.

 

• Table 2, the percentage with falls among females should be 28.9%, not 8.9%

Thank you. We correct it.

 

• l. 274, "ceiling effect". A comparison with results from the SHARE survey for Portugal shows that while the average grip strength in this study is only a  little lower than it is in the population, the standard deviations and the minimum values suggest that the sample includes some people with quite weak grip strength. This puts doubt on the hypothesis of a ceiling effect.

Thank you for this valuable observation. We agree that, although the sample consists of physically active individuals, the variability observed in grip strength—particularly the presence of lower values—does not fully support the assumption of a ceiling effect. We have revised the discussion to soften this interpretation and acknowledge that other factors may better explain the lack of association between grip strength and fall risk.

 

• ll. 276-8. You write that factors such as polypharmacy and comorbidities play a prominent role. These were measured in your study, so why do you not include them in the analysis of fall risk?

Thank you for your comment. While we fully agree that polypharmacy and comorbidities are important contributors to fall risk, these variables were not included in the present study's dataset or analysis. They have been explored in a separate, related project with a different focus. In the current manuscript, they are discussed based on findings from the broader literature, but we have now clarified this to avoid the impression that they were part of the data analyzed here.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Article

Grip Strength, Fall Efficacy, and Balance Confidence as Predictors of Fall Risk in middle-aged and older adults living in the Community

 

Dear authors,

I would like to congratulate you on your remarkable effort. Your study focuses on a topic of particular public health importance and provides important evidence on the role of perceived self-efficacy and balance confidence on fall risk in active older adults. Through constructive comments, my aim was to further strengthen your evidence, interpretation of the findings and their practical applications. I hope these comments will prove useful in the final refinement of your manuscript.

 

Introduction

The introduction thoroughly delineates the theoretical underpinnings of the issue. The physiological changes affecting balance and fall risk in the elderly have been thoroughly evaluated. The incorporation of evidence pertaining to the influence of self-perception on motor behavior is undoubtedly beneficial. The reference to self-efficacy theory was particularly noteworthy. It demonstrated a high level of expertise in the field.

Comment 1: Dear authors, an analysis of the text revealed instances of redundancy that could be streamlined through condensation. For instance, the paragraph commencing with "Beyond the physical repercussions of falls..." reiterates previously mentioned content. A consolidation of the aforementioned references pertaining to "fear of falling" is warranted.

Comment 2: In order to facilitate a more profound comprehension of the research, it would be beneficial to elucidate the rationale behind the selection of these three variables (HGS, fall efficacy, and balance confidence) and their synergy in the identification of fall risk.

Comment 3: In regard to the introduction, it would be beneficial to incorporate a concise reference to the "Stay Up - Falls Prevention Project," including its duration and general objectives, albeit in a cursory manner. This will provide an even clearer context for the study.

 

Materials and Methods

Comment 4: Dear authors, In the paragraph on Grip Strength, the description of the test is slightly vague. While the term "neutral bend with extended elbow" is mentioned, it remains ambiguous as to whether the subject's body was in a seated or standing position, although the context suggests a standing posture. It is vital to report the exact posture, as it has been demonstrated to have a significant impact on the outcomes.

Comment 5: A further point that merits consideration is the calibration status of the dynamometer utilized in the experiment. Given the critical importance of accurate measurement, it would be beneficial to include a concise reference to the calibration procedure or, at the very least, the model of the device utilized.

Comment 6: A definition please. In the "Assessment of Falls" section, the inquiries posed are deemed suitable. However, it remains ambiguous whether participants are permitted to report multiple falls, such as instances where more than one fall is documented. While the mean will be provided subsequently, the inclusion of a brief phrase indicating that "participants could report more than one fall incident in the year" would enhance the clarity of the text.

Comment 7: The utilization of the t-test for the purpose of comparing groups is appropriate. However, the text (forgive me if I don't see it) does not specify whether distributional normality was tested (beyond invoking the "central limit theorem"). Therefore, it would be beneficial to include the results of any normality tests conducted or the findings of any normal distribution analyses applied to the data.

 

Results

Comment 8: Dear Authors, as illustrated in Table 2, pertinent information is presented, including rates of falls and interventions. However, the manner in which this information is presented does not align with the standards outlined in the main text. It is recommended that a brief paragraph be incorporated to provide commentary on the percentages, such as "26.4% of participants reported a fall...".

Comment 9: As illustrated in Tables 3 and 5, the nomenclature employed for the statistical columns is somewhat unusual and may prove to be confusing to the reader. For instance, the abbreviation "x+SD" is used in lieu of the more conventional "Means ± SD," and "B, OR, 95%CI" is employed in place of the standard "B, OR, 95% CI." Conventional statistical formats would facilitate reading for an international audience accustomed to these conventional symbols. Should you believe that this will assist in the promotion of your research, please feel free to make the necessary changes.

Comment 10: The "confidence intervals" fail to provide a clear distinction between 95% confidence intervals and those of a different confidence level. While this may be a reasonable assumption, it is advisable to state it explicitly, for instance, by using the phrase "95% Confidence Interval."

Comment 11: Dear authors, a qualitative assessment of the strength of the correlations (e.g., "moderate," "weak," "strong") would facilitate comprehension of the reader and enhance the clarity of your results. While merely reporting the values of the correlations is adequate, accurate reporting will enhance the value of the study.

Comment 12: It is important to note that... While the survey indicates a higher prevalence of falls among women, it does not provide a detailed analysis of whether this difference remains significant when controlling for other variables using multivariate methods, such as logistic regression. A concise discussion outlining the potential causes of this discrepancy would be greatly appreciated.

 

Discussion

It is my pleasure to offer my congratulations on the discussion section, which is comprehensive and well-researched. Nevertheless, I would like to suggest a few improvements that I believe can enhance the scientific strength and clarity of your argument.

Comment 13: It would be beneficial to articulate more clearly why the sample may have exhibited a kind of "ceiling" in strength. That is, since the participants are physically active, they had probably already reached a functional limit that does not allow for statistical differences. Should there be a perception that this impacts the reliability of HGS as a predictor, it would be prudent to either substantiate this with pertinent literature or propose the implementation of future measurements employing more sensitive assessments of functional capacity.

Comment 14: The interpretation of FES-I and ABC is accurate; however, if further exploration is desired, a more advanced approach could be considered. It would be beneficial to explore the extent to which these effects persist irrespective of gender or other variables, a topic that remains unaddressed in the current discourse. If the statistical analysis permits, this addition would enhance the depth of the study.

Comment 15: Dear authors, I particularly liked your reference to possible interventions (dual-task training, psychological empowerment, etc.). I suggest that you strengthen this section with concrete suggestions for practical implementation so that readers (especially health professionals) can understand how to integrate these practices into falls prevention programs.

Comment 16: You have honestly expressed your limitations, but perhaps you can emphasize in a few words that your findings should not be generalized beyond populations with similar physical activity, given that your sample is selectively active.

 

Conclusions

Comment 17: The conclusions drawn therein assert that FES-I and ABC "demonstrated strong predictive value," a finding that has been validated. However, it is important to note that HGS did not demonstrate independent predictive value; rather, it exhibited a positive correlation with confidence. This observation maintains the coherence between the narrative and the statistical data.

Comment 18: Finally, a recommendation could be made regarding the necessity of longitudinal studies to ascertain whether confidence-building measures are efficacious in reducing falls over time.

Comments for author File: Comments.pdf

Author Response

Review Report 4

 

Dear authors,

I would like to congratulate you on your remarkable effort. Your study focuses on a topic of particular public health importance and provides important evidence on the role of perceived self-efficacy and balance confidence on fall risk in active older adults. Through constructive comments, my aim was to further strengthen your evidence, interpretation of the findings and their practical applications. I hope these comments will prove useful in the final refinement of your manuscript.

 Introduction

The introduction thoroughly delineates the theoretical underpinnings of the issue. The physiological changes affecting balance and fall risk in the elderly have been thoroughly evaluated. The incorporation of evidence pertaining to the influence of self-perception on motor behavior is undoubtedly beneficial. The reference to self-efficacy theory was particularly noteworthy. It demonstrated a high level of expertise in the field.

Comment 1: Dear authors, an analysis of the text revealed instances of redundancy that could be streamlined through condensation. For instance, the paragraph commencing with "Beyond the physical repercussions of falls..." reiterates previously mentioned content. A consolidation of the aforementioned references pertaining to "fear of falling" is warranted.

Thank you for this helpful suggestion. We have revised the paragraph to reduce redundancy and improve clarity. The discussion on fear of falling and related psychological constructs has been condensed, and references have been consolidated to avoid repetition while preserving the conceptual depth.

Comment 2: In order to facilitate a more profound comprehension of the research, it would be beneficial to elucidate the rationale behind the selection of these three variables (HGS, fall efficacy, and balance confidence) and their synergy in the identification of fall risk.

Thank you for your suggestion. We have expanded the final section of the Introduction to clarify the rationale for selecting these three variables. Specifically, we highlight that handgrip strength represents an objective physical marker, while fall efficacy and balance confidence capture distinct yet complementary psychological dimensions of perceived capability and fear. Together, they offer a multidimensional perspective on fall risk that is particularly relevant in active populations.

Comment 3: In regard to the introduction, it would be beneficial to incorporate a concise reference to the "Stay Up - Falls Prevention Project," including its duration and general objectives, albeit in a cursory manner. This will provide an even clearer context for the study.

 Thank you for your suggestion. While the Stay Up – Falls Prevention Project was already mentioned in the Methods section, we have now expanded the description to include its duration and general objectives. These additions aim to better contextualize the origin and framework of the present study, without overextending the scope of the Introduction.

Materials and Methods

Comment 4: Dear authors, In the paragraph on Grip Strength, the description of the test is slightly vague. While the term "neutral bend with extended elbow" is mentioned, it remains ambiguous as to whether the subject's body was in a seated or standing position, although the context suggests a standing posture. It is vital to report the exact posture, as it has been demonstrated to have a significant impact on the outcomes.

Thank you for this important observation. You are correct that posture can influence grip strength outcomes. In our protocol, participants were seated during the handgrip strength test. We have revised the Methods section to clarify this detail and improve the accuracy of the test description.

Comment 5: A further point that merits consideration is the calibration status of the dynamometer utilized in the experiment. Given the critical importance of accurate measurement, it would be beneficial to include a concise reference to the calibration procedure or, at the very least, the model of the device utilized.

Thank you for highlighting this important point. The handgrip strength test was performed using a Saehan SH5001 mechanical spring-type hand dynamometer, a widely used and validated instrument in clinical and research settings. Although the device is non-digital, it was visually inspected and tested for consistent mechanical resistance prior to data collection to ensure accurate measurement. This information has been added to the Methods section.

Comment 6: A definition please. In the "Assessment of Falls" section, the inquiries posed are deemed suitable. However, it remains ambiguous whether participants are permitted to report multiple falls, such as instances where more than one fall is documented. While the mean will be provided subsequently, the inclusion of a brief phrase indicating that "participants could report more than one fall incident in the year" would enhance the clarity of the text.

Thank you for this helpful observation. We confirm that participants were allowed to report more than one fall incident in the previous 12 months. We have now revised the text to include a brief statement clarifying this point in the “Assessment of Falls” section.

Comment 7: The utilization of the t-test for the purpose of comparing groups is appropriate. However, the text (forgive me if I don't see it) does not specify whether distributional normality was tested (beyond invoking the "central limit theorem"). Therefore, it would be beneficial to include the results of any normality tests conducted or the findings of any normal distribution analyses applied to the data.

Thank you for your observation. You are correct that this information was not originally included. We have now incorporated the results of normality testing into the Methods section.

Results

Comment 8: Dear Authors, as illustrated in Table 2, pertinent information is presented, including rates of falls and interventions. However, the manner in which this information is presented does not align with the standards outlined in the main text. It is recommended that a brief paragraph be incorporated to provide commentary on the percentages, such as "26.4% of participants reported a fall...".

Thank you for this helpful suggestion. We have now included a brief descriptive paragraph in the Results section to summarize and interpret the main findings from Table 2.

Comment 9: As illustrated in Tables 3 and 5, the nomenclature employed for the statistical columns is somewhat unusual and may prove to be confusing to the reader. For instance, the abbreviation "x+SD" is used in lieu of the more conventional "Means ± SD," and "B, OR, 95%CI" is employed in place of the standard "B, OR, 95% CI." Conventional statistical formats would facilitate reading for an international audience accustomed to these conventional symbols. Should you believe that this will assist in the promotion of your research, please feel free to make the necessary changes.

Thank you for your thoughtful suggestion. We agree that adopting conventional statistical formatting improves clarity and accessibility for an international readership. Accordingly, we have revised the column headers in Tables 3 and 5 to use standardized terminology, including "Mean ± SD" and "B, OR, 95% CI."

Comment 10: The "confidence intervals" fail to provide a clear distinction between 95% confidence intervals and those of a different confidence level. While this may be a reasonable assumption, it is advisable to state it explicitly, for instance, by using the phrase "95% Confidence Interval."

Thank you for your suggestion. We have revised the table legend to explicitly state that all confidence intervals refer to a 95% confidence level.

Comment 11: Dear authors, a qualitative assessment of the strength of the correlations (e.g., "moderate," "weak," "strong") would facilitate comprehension of the reader and enhance the clarity of your results. While merely reporting the values of the correlations is adequate, accurate reporting will enhance the value of the study.

Thank you for this suggestion. We have now added a qualitative interpretation of the strength of the correlations in the Results section. For instance, correlations between fall efficacy and balance confidence were classified as strong, while associations between grip strength and falls were considered weak, following Cohen’s conventional thresholds.

Comment 12: It is important to note that... While the survey indicates a higher prevalence of falls among women, it does not provide a detailed analysis of whether this difference remains significant when controlling for other variables using multivariate methods, such as logistic regression. A concise discussion outlining the potential causes of this discrepancy would be greatly appreciated.

 Thank you for your observation. We agree that further analysis of gender differences in fall risk would strengthen the manuscript. However, in our multivariable logistic regression model, sex was included as a covariate but did not emerge as a significant independent factor once psychological variables were accounted for. We have added a brief comment in the Discussion to acknowledge this and suggest future stratified analyses to explore potential sex-based differences in fall-related risk profiles.

Discussion

It is my pleasure to offer my congratulations on the discussion section, which is comprehensive and well-researched. Nevertheless, I would like to suggest a few improvements that I believe can enhance the scientific strength and clarity of your argument.

Comment 13: It would be beneficial to articulate more clearly why the sample may have exhibited a kind of "ceiling" in strength. That is, since the participants are physically active, they had probably already reached a functional limit that does not allow for statistical differences. Should there be a perception that this impacts the reliability of HGS as a predictor, it would be prudent to either substantiate this with pertinent literature or propose the implementation of future measurements employing more sensitive assessments of functional capacity.

Thank you for this valuable suggestion. We have revised the discussion to clarify that although the sample consisted of physically active older adults, the observed variability in grip strength—particularly the presence of lower strength values—suggests that a true ceiling effect may not have occurred. Instead, we now propose that in active populations, other factors (e.g., confidence, agility) may play a more dominant role in fall risk, as discussed in recent literature. This revision aims to more accurately represent the observed data.

Comment 14: The interpretation of FES-I and ABC is accurate; however, if further exploration is desired, a more advanced approach could be considered. It would be beneficial to explore the extent to which these effects persist irrespective of gender or other variables, a topic that remains unaddressed in the current discourse. If the statistical analysis permits, this addition would enhance the depth of the study.

Thank you for this thoughtful comment. While our current analysis did not include interaction or stratified models to examine whether the associations between FES-I, ABC, and fall risk persist across subgroups such as sex or other covariates, we agree that this is a relevant line of inquiry. Accordingly, we have added a sentence in the Discussion section suggesting that future studies should explore whether these psychological associations differ by sex or other demographic or health-related factors, using stratified or interaction analyses.

Comment 15: Dear authors, I particularly liked your reference to possible interventions (dual-task training, psychological empowerment, etc.). I suggest that you strengthen this section with concrete suggestions for practical implementation so that readers (especially health professionals) can understand how to integrate these practices into falls prevention programs.

We appreciate this excellent suggestion. We have expanded the final paragraphs of the Discussion to include more practical recommendations. For example, we now mention dual-task balance training, graded exposure to challenging tasks in safe environments, and psychological techniques like cognitive restructuring and confidence-building workshops as potential intervention strategies. These additions aim to support clinicians and public health professionals in translating our findings into practice.

Comment 16: You have honestly expressed your limitations, but perhaps you can emphasize in a few words that your findings should not be generalized beyond populations with similar physical activity, given that your sample is selectively active.

Thank you. We agree that this is an important limitation. While the Discussion already notes the selectivity of our sample, we have now emphasized more clearly that our findings should not be generalized to sedentary or frail older populations, as the participants were all physically active and voluntarily engaged in structured exercise programs.

Conclusions

Comment 17: The conclusions drawn therein assert that FES-I and ABC "demonstrated strong predictive value," a finding that has been validated. However, it is important to note that HGS did not demonstrate independent predictive value; rather, it exhibited a positive correlation with confidence. This observation maintains the coherence between the narrative and the statistical data.

Thank you for pointing this out. We have revised the Conclusion section to clarify it.

Comment 18: Finally, a recommendation could be made regarding the necessity of longitudinal studies to ascertain whether confidence-building measures are efficacious in reducing falls over time.

We appreciate this excellent recommendation. In the final part of the Discussion, we have added a recommendation for future longitudinal studies to investigate whether interventions aimed at enhancing confidence and fall efficacy effectively reduce fall incidence over time.

Author Response File: Author Response.pdf

Reviewer 5 Report

Comments and Suggestions for Authors

This study investigated the associations of grip strength, fall efficacy, and balance confidence with fall risk among middle-aged and older adults in Lisbon. While the identification of associations between psychological variables and fall risk is a meaningful finding, there are several major concerns regarding the analytical approach and the interpretation of the results. Please consider the following comments.

 

1. The term "predictor" implies a causal direction between variables. However, grip strength, fall efficacy, and balance confidence may be both causes and consequences of fall risk. Cross-sectional studies are inherently limited in their ability to disentangle such bidirectionality. Therefore, terms that imply association rather than prediction should be used. In addition, causal language, such as "predict", "effect", "impact", and "influence", must be avoided throughout the manuscript.

2. The manuscript lacks sufficient detail regarding participant recruitment. Information should be provided on the time period during which participants were recruited, the method of recruitment, whether a target sample size was pre-calculated, how many participants were intended to be enrolled, how many were actually recruited, and the final sample size included in the analysis... etc.

3. Although the authors assumed a normal distribution of their variables, the statistical analyses included both parametric (Student’s t-test) and non-parametric tests (Spearman’s correlation), which are based on contradictory assumptions about variable distribution, which is problematic. The authors can consider conducting normality tests such as the Kolmogorov-Smirnov and Shapiro-Wilk tests.

4. Binary logistic regression yields information about mere associations between variables. Multivariate logistic regression adjusting for sex, age, bmi, medications, and comorbidity should be employed.

5. Psychological factors are likely to be consequences of previous fall experiences. The possibility of such bidirectionality should be thoroughly addressed in the discussion section. The current discussion tends to overstate the influence of FES-I and ABC on fall risk. Given the study’s cross-sectional design, such directional interpretation is inappropriate. The tone should be tempered, and the limitations of the study design should be more carefully considered.

6. Line 48: There is a typo in "The During."

Author Response

Review Report 5

 

This study investigated the associations of grip strength, fall efficacy, and balance confidence with fall risk among middle-aged and older adults in Lisbon. While the identification of associations between psychological variables and fall risk is a meaningful finding, there are several major concerns regarding the analytical approach and the interpretation of the results. Please consider the following comments.

 

1. The term "predictor" implies a causal direction between variables. However, grip strength, fall efficacy, and balance confidence may be both causes and consequences of fall risk. Cross-sectional studies are inherently limited in their ability to disentangle such bidirectionality. Therefore, terms that imply association rather than prediction should be used. In addition, causal language, such as "predict", "effect", "impact", and "influence", must be avoided throughout the manuscript.

Thank you for this important observation. We have reviewed the entire manuscript and revised the language to avoid causal terms such as “predict,” “effect,” “impact,” and “influence.” All terminology has been adjusted to reflect associations, in accordance with the cross-sectional nature of the study.

2. The manuscript lacks sufficient detail regarding participant recruitment. Information should be provided on the time period during which participants were recruited, the method of recruitment, whether a target sample size was pre-calculated, how many participants were intended to be enrolled, how many were actually recruited, and the final sample size included in the analysis... etc.

Thank you for your comment. We have now added more detailed information regarding participant recruitment in the Methods section. Specifically, we clarified that participants were recruited from an existing community-based exercise program near Lisbon, Portugal. No a priori sample size calculation was conducted; instead, all individuals who were eligible and agreed to participate were included. Participants were informed about the study and invited to attend the evaluation session one week in advance. The final sample size reflects all participants from the program who met the inclusion criteria and consented to take part in the assessments.

3. Although the authors assumed a normal distribution of their variables, the statistical analyses included both parametric (Student’s t-test) and non-parametric tests (Spearman’s correlation), which are based on contradictory assumptions about variable distribution, which is problematic. The authors can consider conducting normality tests such as the Kolmogorov-Smirnov and Shapiro-Wilk tests.

Thank you for this important observation. In response, we conducted formal normality tests (Shapiro-Wilk and Kolmogorov-Smirnov) for the main continuous variables. Based on the results, we confirmed that some variables deviated from a normal distribution. As such, we used parametric tests (e.g., t-test) only where normality was not violated and applied non-parametric methods (e.g., Spearman’s correlation) where appropriate. This approach has now been clarified in the Statistical Analysis section.

4. Binary logistic regression yields information about mere associations between variables. Multivariate logistic regression adjusting for sex, age, bmi, medications, and comorbidity should be employed.

Thank you for your suggestion. We agree that adjusting for relevant covariates improves the robustness of the findings. Accordingly, we revised our analysis and conducted a multivariable binary logistic regression, including sex, age, BMI, number of medications, and comorbidities as covariates in the model. This approach allows for a more accurate interpretation of the associations between the main variables of interest and fall risk. The Methods, Results, and Table 5 have been updated to reflect this change.

5. Psychological factors are likely to be consequences of previous fall experiences. The possibility of such bidirectionality should be thoroughly addressed in the discussion section. The current discussion tends to overstate the influence of FES-I and ABC on fall risk. Given the study’s cross-sectional design, such directional interpretation is inappropriate. The tone should be tempered, and the limitations of the study design should be more carefully considered.

Thank you for this thoughtful comment. We have further revised the discussion to explicitly acknowledge the potential bidirectional relationship between psychological factors and fall risk. In particular, we emphasize that low fall efficacy and balance confidence may be both contributors to and consequences of prior falls. We also clarified that our cross-sectional design limits any interpretation regarding the directionality of these associations and have adjusted the tone accordingly.

6. Line 48: There is a typo in "The During."

Thank you for correcting this error. The typo has been corrected.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The revised manuscript demonstrates substantial improvements and successfully addresses the majority of the reviewers’ concerns. The authors have made thoughtful and well-documented revisions regarding conceptual clarity, methodological transparency, and statistical rigor. In particular, the replacement of causal terminology, clarification of psychometric properties, and expansion of the limitations section are commendable.

However, a few minor issues remain that should be addressed prior to publication to ensure clarity, consistency, and editorial quality.

Remaining Recommendations

Several sections still contain residual grammatical errors and editorial oversights. For example:

Line 50: “The During the aging process” — remove the duplicated article.

Lines 79–84: Phrases like “the physical repercussions of falls, psychological factors play a significant role...” appear twice in slightly different phrasings. One version should be removed to avoid redundancy.

A final proofread by a native English speaker or professional editor is advised to ensure fluency and clarity.

There is still some ambiguity regarding the posture during handgrip strength testing. In one part of the methods, the test is described as conducted in a seated position, while a previous sentence suggests a standing posture (line 182 vs. 184). Please clarify and ensure consistency.

While tables are informative, the addition of a visual summary (e.g., a path diagram or forest plot) could significantly improve the interpretability of associations, particularly for non-specialist readers.

Ensure uniform formatting of statistical symbols (e.g., italicize p, M, SD).

Standardize decimal notation using periods instead of commas throughout the manuscript (e.g., “88,76%” should be “88.76%”).

Review headings and subheadings for consistency with journal style.

Continue to ensure that terms such as “fall efficacy” and “fear of falling” are used consistently and that their conceptual distinction is clearly maintained throughout the manuscript.

Author Response

Dear Reviewer,

We sincerely thank you once again for your thoughtful review and constructive feedback. Below, we provide detailed responses to the remaining points raised, and we have implemented the corresponding revisions in the manuscript using Track Changes.

 

Review 2

Line 50: “The During the aging process” — remove the duplicated article.

Thank you for pointing this out. As mentioned, this duplication had already been removed in the revised version. We confirm that the phrase now reads: “During the aging process…”

 

Lines 79–84: Phrases like “the physical repercussions of falls, psychological factors play a significant role...” appear twice in slightly different phrasings. One version should be removed to avoid redundancy.

Thank you for highlighting this redundancy. We have removed the repeated phrasing and retained the most concise and relevant version to improve clarity and avoid duplication (lines 79–84).

 

A final proofread by a native English speaker or professional editor is advised to ensure fluency and clarity.

We have now conducted a final proofreading of the entire manuscript with the assistance of a professional editor fluent in academic English, ensuring fluency, clarity, and consistency in grammar and style throughout.

 

There is still some ambiguity regarding the posture during handgrip strength testing. In one part of the methods, the test is described as conducted in a seated position, while a previous sentence suggests a standing posture (line 182 vs. 184). Please clarify and ensure consistency.

Thank you for this observation. We have clarified and unified the description to indicate that all handgrip strength tests were conducted in a seated position, in line with standard protocols. The contradictory sentence has been revised for consistency (see updated methods section).

 

While tables are informative, the addition of a visual summary (e.g., a path diagram or forest plot) could significantly improve the interpretability of associations, particularly for non-specialist readers.

Thank you for this helpful suggestion. While we recognize the value of visual summaries, we chose not to include a forest plot or path diagram because the logistic regression results are already clearly and completely presented in Table 5. Given the small number of predictors and the simplicity of the model, an additional figure would largely replicate existing information without adding substantial interpretive value. To maintain clarity and avoid redundancy, we opted to retain only the table format in this case.

 

Ensure uniform formatting of statistical symbols (e.g., italicize p, M, SD).

We have reviewed and standardized the formatting of all statistical notations throughout the manuscript. All symbols such as p, M, and SD are now properly italicized in accordance with journal style guidelines.

 

Standardize decimal notation using periods instead of commas throughout the manuscript (e.g., “88,76%” should be “88.76%”).

We have corrected all decimal notations, ensuring the use of periods instead of commas throughout the manuscript.

 

Review headings and subheadings for consistency with journal style.

All headings and subheadings have been reviewed and reformatted to ensure consistency with the journal's style requirements (e.g., consistent use of sentence case or title case as appropriate).

 

Continue to ensure that terms such as “fall efficacy” and “fear of falling” are used consistently and that their conceptual distinction is clearly maintained throughout the manuscript.

We reviewed the entire manuscript to ensure conceptual consistency. The term “fall efficacy” is now used exclusively when referring to the construct measured by the FES-I, and “fear of falling” is only mentioned when discussing the broader psychological phenomenon or literature references. Their distinction has been maintained and clarified where necessary.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for addressing my comments in this revision quite adequately. No further comments.

Author Response

Thank you for your review. 

Reviewer 5 Report

Comments and Suggestions for Authors

Thank you for addressing my comments.

Good luck!

Author Response

Thank you for your review.