Mental Health, Resilience, and Physical Activity in Civilians Affected by Conflict-Related Trauma: A Cross-Sectional Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I am satisfied that this study is rigorous, significant and original. The originality comes not from the survey instruments of the methodology, all of which have well established histories and uses. The originality relates to the chosen research population. The author does not 'over claim' the significance of the results, and the measured way in which the results are presented adds to their credibility. While all scholars have their own preferences (biases?) when it comes to approaches and schools of thought, I do not wish to impose changes to the paper in order to be closer to my own approach. 

The topic has a degree of originality and extends its application in the field. It focuses on associations between mental health and physical exercise among civilians exposed to conflict-related trauma in Israel. It addresses a practical gap in the field since such a study has not been carried out on the population that it studies in this work. However, the research instruments are well known and have been applied in numerous different contexts with different populations.

This study provides insights that are specific to the communities being researched in Israel. To that extent, the significance and originality are geographically situated. However, given the extent of war-related trauma being experienced in different parts of the world, there may be transferable insights and practices. However, the paper Yli-Panula, E., et al., 2024. The mediating effect of resilience between physical activity and mental health: a meta-analysis and pathway analysis. Frontiers in Public Health, [online] 17, p.1434624 conducts a meta analysis of papers that research the link between mental wellbeing and physical activity, indicating that a wide range of related studies have already been carried out.  

In Section 2: Materials and Methods it states, ‘Most participants were evacuated from "Community A,"’ while others were recruited from Community B. Specific numbers of participants from each Community should be set out here. If there is a disparity between them, the author should justify the numbers of participants and acknowledge the benefits and weaknesses of this approach.

Questionnaire and Variables

The author sets out the questionnaires that were used. Provide a justification for the four that were selected. There are other survey tools available – why choose these ones? What are the benefits and limitations?

Statistical Analysis

The description of the statistical analysis is vague. Set out the process more clearly: Step 1…Step 2… and so on. A diagram might work well here. Justify the choice of analytical processes: what are the benefits vs the limitations.

The conclusions are consistent with the evidence that has been presented. The author appropriately interprets the data and does not overstate causal relationships. Consequently, the findings are clearly based on the recommendations offered. However, the Conclusions section is short and generic. Including some data – numbers, percentages, degree of significance of findings – will strengthen the conclusions overall.

• Any additional comments on the tables and figures.
To make it easier for the reader to easily pick out the significant factors, highlight the key numbers in Bold.

Author Response

Healthcare- 3689498

 

 

Dear Reviewer

Many thanks to your thoughtful and constructive feedback. The detailed comments contributed significantly to enhancing the clarity, depth, and academic rigor of the manuscript. All suggestions were carefully considered, and appropriate revisions have been made accordingly. Responses to each comment appear below.

1. 
   In Section 2: Materials and Methods it states, ‘Most participants were evacuated from "Community A,"’ while others were recruited from Community B. Specific numbers of participants from each Community should be set out here. If there is a disparity between them, the author should justify the numbers of participants and acknowledge the benefits and weaknesses of this approach.

 

Response to Comment 1:

This study included individuals aged 25 years and older who were residents of rural communities in a conflict-affected region of southern Israel. Of the total sample, 43 participants (40.2%) were evacuated from "Community A," which was directly exposed to severe trauma and violence during the conflict. An additional 19 participants (17.8%) were from "Community B," a nearby locality within the same region that was not directly impacted by the traumatic events. The remaining 45 participants (42.1%) were evacuees from nine other rural communities within the broader conflict zone.

Participants were recruited while temporarily residing in designated hotels for evacuees. As a result, the distribution of participants across communities was not uniform, primarily due to logistical constraints and voluntary participation rates. Specifically, individuals were approached on-site, and not all those contacted agreed to participate. This natural variation in community representation reflects the context of real-time data collection under emergency conditions. The implications of this sampling approach—including its strengths in ecological validity and limitations in terms of representativeness—are acknowledged and discussed in the study’s limitations section.

 

This is the new version of the section:

This study included individuals aged 25 years and older who were residents of rural communities in a conflict-affected region of southern Israel. The mean age of participants was 46.2 years (SD = 12.9). Of the 107 participants, 77 (72.0%) were female and 30 (28.0%) were male. In terms of community origin, 43 participants (40.2%) were evacuated from "Community A," which was directly exposed to severe trauma and violence during the conflict. An additional 19 participants (17.8%) were from "Community B," a nearby locality within the same region that was not directly impacted by the traumatic events. The remaining 45 participants (42.1%) were evacuees from nine other rural communities within the broader conflict zone. Marital status was reported as follows: 85 participants (79.4%) were married or in a committed relationship, 8 (7.5%) were single, 9 (8.4%) were divorced, 4 (3.7%) were widowed, and 1 (0.9%) reported another status. Regarding education level, 22 participants (20.6%) had completed high school, 55 (51.3%) held a bachelor’s degree, 29 (27.1%) had a master’s degree or higher, and 1 participant (0.9%) reported no formal education. A detailed breakdown of demographic characteristics is presented in Table 1.

 

 

Questionnaire and Variables

2. The author sets out the questionnaires that were used. Provide a justification for the four that were selected. There are other survey tools available – why choose these ones? What are the benefits and limitations?

 

Response to Comment 2:

The first questionnaire was a demographic questionnaire designed by the researchers to collect background information (e.g., age, gender, academic status, and physical activity habits); as it did not assess psychological constructs, no psychometric validation was required.

The other four questionnaires selected for this study were chosen based on their extensive use in prior research, their validated translations into Hebrew, and their established psychometric properties—including reliability and validity—in both international and Israeli contexts.

The GAD-7 anxiety scale (Löwe et al., 2008) was selected due to its brevity, clinical relevance, and widespread use in both research and healthcare settings. Importantly, it has been recommended by the Israeli Ministry of Health’s Professional Committee for Clinical Psychology and has been validated in Hebrew, (Cronbach α=.931), (Reference 22).

The Connor-Davidson Resilience Scale (CD-RISC) (Connor & Davidson, 2003) was chosen for its strong theoretical grounding, robust psychometric properties, and extensive use in resilience research especially during the last two years in Israeli studies. The Hebrew version of the scale was translated and validated by the original developers, ensuring linguistic and conceptual equivalence (Cronbach α=.923)  (Reference 23).

For assessing well-being, the Mental Health Inventory (MHI) (Veit & Ware, 1983) (Reference 24), was selected. This measure is widely cited in the literature and has been validated in its Hebrew version by Florian and Drori (1990), (Reference 25), who confirmed its psychometric adequacy in an Israeli population (Cronbach α=.960)  (Reference 24). This questionnaire was also recommended by the Israeli Ministry of Health’s Professional Committee for Clinical Psychology.

Physical activity was assessed using the International Physical Activity Questionnaire – Short Form (IPAQ-SF), a widely used and validated self-report instrument developed for international comparison of physical activity levels (Reference 26). The questionnaire was administered in its Hebrew-translated version, which has previously demonstrated acceptable validity and reliability in a study among Israeli undergraduate students (Reference 27).

All selected instruments have been previously used in Israeli studies, and their Hebrew versions have undergone systematic processes of translation, adaptation, and psychometric validation, which ensures cultural relevance and measurement accuracy in the local context. Their selection was guided by both methodological rigor and practical applicability, balancing comprehensiveness with feasibility.

While alternative instruments exist, the selected tools were preferred due to their established validity in the target population, their frequent use in the literature, and their suitability for the study’s objectives.

(22) Lowe, B.; Decker, O.; Müller, S.; Brähler, E.; Schellberg, D.; Herzog, W.; Herzberg, P. Y. Validation and standardization of the Generalized Anxiety Disorder Screener (GAD-7) in the general population. Medical Care 2008, 46 (3), 266–274. https://doi.org/10.1097/MLR.0b013e318160d093

(23) Connor, K. M.; Davidson, J. R. T.  Development of a new resilience scale: The Connor-Davidson Resilience Scale (CD-RISC). Depression and Anxiety 2003, 18(2), 76–82. https://doi.org/10.1002/da.10113

 

(24) Veit, C.T.; Ware, J.E. Jr. The structure of psychological distress and wellbeing in general populations. J Consult and Clin Psychol 1983, 51(5), 730–742.

 

(25) Florian V and Drori Y. Mental Health Inventory (MHI): psychometric characteristics and normative data regarding the Israeli population. Psychology 1990; 1: 26–35.

(26) Booth M. Assessment of physical activity: an international perspective. Res Q Exerc Sport 2000 Jun;71 Suppl 2:114-120. [doi: 10.1080/02701367.2000.11082794] [Medline: 25680021]

(27) Joseph, G.; Schori, H. The Beneficial Effect of the First COVID-19 Lockdown on Undergraduate Students of Education: Prospective Cohort Study. JMIR Formative Research 2022 Feb 23,6(2), e27286. doi: 10.2196/27286

 

This is the new version of the Questionnaire and Variables section:

The questionnaire comprised several sections.
The demographic section collected data on age, gender, marital status, education level, place of residence during the conflict, current residence, employment status prior to the conflict, whether family members had been drafted or affected by the conflict, and whether the current residence included a protected shelter.

Anxiety symptoms were measured using the Generalized Anxiety Disorder Screener (GAD-7) [22], a validated 7-item instrument assessing symptom severity over the past two weeks. Items were rated on a 4-point Likert scale (1 = never to 5 = nearly every day). The GAD-7 is widely used in clinical and research settings and has been validated in Hebrew (Cronbach α=.931.); it is also recommended by the Israeli Ministry of Health’s Professional Committee for Clinical Psychology.

Resilience was assessed with the Connor-Davidson Resilience Scale (CD-RISC) [23], which evaluates an individual's ability to adapt to adversity. Responses were rated on a 6-point Likert scale (1 = never to 6 = nearly all the time). The Hebrew version of the CD-RISC was translated and validated by the original developers (Cronbach α=.923) and has been widely used in Israeli research, demonstrating strong psychometric properties.

Well-being was measured using a 10-item well-being questionnaire [24] assessing emotional, psychological, and social aspects of well-being. Participants responded on a 5-point Likert scale (1 = agree to 5 = agree very much). The Hebrew version, adapted and validated by Florian and Drori (Cronbach α=.96) [25], has been shown to have good reliability and validity in the Israeli population.

Physical activity was assessed using the International Physical Activity Questionnaire – Short Form (IPAQ-SF), a widely used and validated self-report instrument developed for international comparison of physical activity levels [26]. The questionnaire was administered in its Hebrew-translated version, which has previously demonstrated acceptable validity and reliability in a study among Israeli undergraduate students [27].

 Participants reported their average weekly duration of physical activity (in minutes) and the amount of time spent in vigorous-intensity exercise, defined as activities inducing heavy breathing and a noticeably elevated heart rate, as recommended by the World Health Organization (WHO),[28].

 

For analysis, physical activity was categorized in two ways:
Total weekly exercise duration: Non-exercisers: 0–30 minutes/week, Moderate exercisers: 31–180 minutes/week, High exercisers: >180 minutes/week.
Vigorous-intensity activity duration: Non-exercisers: 0–30 minutes/week, Moderate exercisers: 31–180 minutes/week, High exercisers: >180 minutes/week.

 

 

 

Statistical Analysis

3. The description of the statistical analysis is vague. Set out the process more clearly: Step 1…Step 2… and so on. A diagram might work well here. Justify the choice of analytical processes: what are the benefits vs the limitations.

 

Response to comment 3:

Thank you for this important and constructive comment. The paragraph has been revised and is now clearer and more structured.

This is the revised section: “Descriptive statistics (means, standard deviations, and frequencies) were calculated to summarize demographic characteristics and key study variables. Normality of the continuous variables (resilience, anxiety, and well-being) was assessed using skewness and kurtosis values. All values fell within acceptable ranges (|skewness| < 1, |kurtosis| < 3), indicating that the assumption of approximate normality was met across the three physical activity groups (non-exercisers, moderate exercisers, and high exercisers). Homogeneity of variances was tested using Levene’s test, which indicated that the assumption was met for resilience (p = .787) and well-being (p = .526), but not for anxiety (p = .002). Therefore, in addition to the one-way ANOVA, a non-parametric Kruskal–Wallis test was conducted for anxiety.

To examine group differences in psychological variables across levels of physical activity (in terms of total duration and intensity), one-way analyses of variance (ANOVA) were conducted for resilience and well-being, as the assumptions of normality and homogeneity of variances were met. For anxiety, the Kruskal–Wallis test was used due to violation of the homogeneity of variances assumption. Pearson correlation coefficients were computed to assess associations among anxiety, resilience, well-being, and physical activity variables. To evaluate the predictive value of physical activity and psychosocial factors on mental well-being, a multiple linear regression analysis was performed. Predictor variables included weekly physical activity duration, frequency of physical activity, duration of vigorous-intensity activity, anxiety levels, resilience levels, perceived level of security in the current residence, and place of residence. Place of residence was categorized as “Community A” (high exposure to trauma and violence) and “Community B” (a nearby area not directly impacted by the conflict).

A post hoc power analysis was conducted to evaluate whether the sample size (N = 107; group sizes n = 22, 45, and 40) provided sufficient statistical power for the primary analyses. Based on the observed group means and standard deviations, effect sizes were estimated as follows: anxiety (Cohen’s d = 0.32, small to medium), resilience (d = 0.71, medium to large), and well-being (d = 0.92, large). These values indicate that the study had adequate statistical power (≥ 0.80) to detect medium to large effects using both ANOVA and regression models. All statistical tests were two-tailed, and statistical significance was set at p < .05.

All analyses were conducted using IBM SPSS Statistics, version 29.0.1.0.”

 

4. The conclusions are consistent with the evidence that has been presented. The author appropriately interprets the data and does not overstate causal relationships. Consequently, the findings are clearly based on the recommendations offered. However, the Conclusions section is short and generic. Including some data – numbers, percentages, degree of significance of findings – will strengthen the conclusions overall.

 

Response to Comment 4:

 

Thank you for your thoughtful feedback. We appreciate your recognition of the study’s alignment between findings and conclusions, and we agree that providing more specific data would strengthen the final section. In response, we have revised the Conclusions section to include key statistical findings—such as p-values and effect sizes—to more clearly support the interpretation of results. This revised paragraph offers a more data-informed and nuanced summary, while maintaining appropriate caution given the study’s limitations.

This is the revised conclusion:

This study highlights the potential psychological benefits of physical activity among civilians affected by traumatic events. A high volume of vigorous-intensity physical activity was associated with enhanced resilience (p = 0.035, d = -0.68) and well-being (p = 0.007, d = -0.89), with individuals engaging in higher levels of activity reporting better psychological outcomes. Although the association between physical activity and anxiety did not reach conventional statistical significance (p = 0.075), a trend toward lower anxiety was observed among the most active participants (d = 0.31). These findings suggest that promoting physical activity may contribute meaningfully to psychological recovery following trauma. Integrating physical activity into post-trauma intervention programs, alongside psychosocial support, could help address the complex needs of trauma-exposed communities and support improved long-term mental health outcomes.

 

 

5. To make it easier for the reader to easily pick out the significant factors, highlight the key numbers in Bold.

 

Response to Comment 5:

Thank you for this valuable insight. To make it easier for the reader to easily pick out the significant factors the significant factors were highlighted in Bold in table 2 and 3.

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

This study examines the mental health effects of physical activity on individuals displaced by conflict. While mass casualty events in conflict zones often result in profound psychological distress, physical activity is increasingly recognized as a potential protective factor. The authors examine how physical activity relates to anxiety, resilience, and well-being in a sample of evacuees. Using a cross-sectional survey design with 107 participants, the study explores these associations and identifies physical activity, particularly at higher intensities, as a significant contributor to improved mental health outcomes.

The study addresses an important and under-researched population, and the topic is timely and of clear relevance within both public health and humanitarian contexts. However, the manuscript presents several issues that must be addressed before it can be considered for publication. I therefore recommend a major revision at this stage.

The abstract clearly outlines the study's background, objectives, methods, and key findings. However, it includes specific statistical values (e.g., means, p-values, regression coefficients), which are generally unnecessary in an abstract. I recommend focusing on the overall trends and key conclusions rather than detailed numerical results. This will improve readability.

Lines 66-70: The author states: "Overcoming physical obstacles strengthens the belief that one can overcome barriers in other areas of life. Specifically, for populations that have experienced traumatic events, such as civilians evacuated from their homes, physical activity can serve as an important anchor for maintaining both physical and mental stability." This is a strong and general claim presented without any supporting citation. In the context of the introduction, where theoretical grounding is crucial, such statements should be substantiated with relevant literature. If empirical support exists, it should be cited. If this is the author’s conceptual framing or hypothesis, that should be made explicit to maintain academic transparency and avoid misleading the reader.

Section 2.1 – Study Population and Design. This subsection lacks essential descriptive information about the participants. Specifically, there is no mention of participants’ age (mean, standard deviation, and range), gender distribution, or the number of individuals from each group (Community A vs. Community B). These demographic details are critical for assessing the representativeness of the sample and for contextualizing the study’s findings. I recommend including a clear breakdown of these variables to enhance the transparency and interpretability of the study.

On this regard (lines 164-175), the description of the study sample presented in the Results section belongs in the methodology section, not under results. Participant demographics—such as age, gender distribution, education level, and pre-conflict employment status—are not study findings, but rather characteristics of the sample. I recommend relocating this paragraph to the appropriate subsection under Study Population and Design or creating a distinct Participants subsection within Methods.

Additionally, the term “predominantly women” is vague. Please clarify what proportion of participants were women by providing a specific percentage or count.

Lines 127–129 – the description of the GAD-7 used to assess anxiety symptoms is incomplete. Specifically, the manuscript should clearly state the response scale (e.g., Likert points, range of scores).

Line 143: It appears that a comment from a previous review was inadvertently included in the body of the manuscript: "Your Statistical Analysis section is solid and clear… Here's an enhanced version." This language is clearly intended as reviewer feedback rather than part of the article itself. Please revise the text to remove any remnants of prior peer review comments and ensure that the section reflects only the finalized content intended for publication. Careful proofreading is advised before resubmission.

Lines 180–184: The manuscript reports the use of a one-way ANOVA to test for differences in resilience, anxiety, and well-being across three physical activity groups. While ANOVA is appropriate for comparing means across multiple groups, the relatively small sample sizes in each group (n=22, n=45, n=40) raise concerns about the statistical power and robustness of the test. The authors should clarify whether assumptions for ANOVA—such as normality and homogeneity of variances—were tested and met. If these assumptions were not verified, or if sample sizes are deemed too small, non-parametric alternatives (e.g., Kruskal-Wallis test) might be more suitable.

Additionally, the authors should elaborate on the rationale for grouping participants in this way and justify the selected cutoffs for “moderate” and “high” physical activity. Clarifying these methodological choices would strengthen the validity of the results. Importantly, the results section does not report that no significant differences were found when comparing the “no physical activity” group to others.

Lines 212–213: The manuscript describes the Pearson correlation coefficient (r = 0.296, p = 0.002) between resilience levels and weekly duration of intense physical activity as a "weak" relationship. While the correlation coefficient indicates a modest association, given the sample size, this statistically significant correlation may hold meaningful practical relevance. In psychological and behavioral studies, correlations of this magnitude can be important. Therefore, I recommend revising the description to reflect this nuance—for example, referring to the relationship as “a small to moderate but statistically significant positive association”—to better capture its significance in context.

Furthermore, there is an inconsistency in how correlation strengths are characterized. The correlation between resilience and intense physical activity (r = 0.296) is described as “weak,” whereas the lower negative correlation between physical activity duration and anxiety (r = –0.199, p = 0.04) is presented without any qualifier regarding its strength (lines 223-224).

Table 2 currently includes both p-values and asterisks (*) indicating statistical significance. To improve clarity and reduce redundancy, it is advisable to present only one of these indicators. Typically, the use of asterisks to denote significance levels is sufficient, provided that the table legend or footnote clearly explains what each number of asterisks represents (e.g., *p < 0.05, **p < 0.01). Please include such a description in the table caption or footnote to guide readers.

Lines 254-255: The statement, “Our findings emphasize the significant role of physical activity in promoting psychological health during times of crisis,” overstates the conclusions given the relatively small sample size. It would be more appropriate to use more cautious language, such as “Our findings suggest a potential role…” or “These results indicate that physical activity may contribute to psychological health…” to better reflect the exploratory nature of the study and its limitations.

The limitations section does not address the unequal distribution of participants, not only across physical activity groups (notably the smaller “no physical activity” group) but also between Community A and Community B. These imbalances may impact the statistical power and generalizability of the findings. It is important to explicitly acknowledge these sample distribution issues to provide a more thorough discussion of the study’s limitations.

Author Response

Healthcare- 3689498

 

 

Dear Reviewer

 

Many thanks to your thoughtful and constructive feedback. The detailed comments contributed significantly to enhancing the clarity, depth, and academic rigor of the manuscript. All suggestions were carefully considered, and appropriate revisions have been made accordingly. Responses to each comment appear below.

 

1.The abstract clearly outlines the study's background, objectives, methods, and key findings. However, it includes specific statistical values (e.g., means, p-values, regression coefficients), which are generally unnecessary in an abstract. I recommend focusing on the overall trends and key conclusions rather than detailed numerical results. This will improve readability.

 

Response to Comment 1:

Thank you for your helpful comment. We agree that reducing the amount of detailed statistical information in the abstract would improve readability and accessibility for a broader audience. In response, we have revised the Results section of the abstract to remove specific numerical values such as means, coefficients, and exact p-values where not essential, and instead emphasized overall trends and key findings. The updated version offers a clearer and more reader-friendly summary of the results while maintaining scientific accuracy.

This is the revised section:

“Results: Evacuees engaging in more than three hours of vigorous-intensity physical activity exhibited significantly higher resilience and better well-being compared to those with lower activity levels. Although not statistically significant, the data suggested a possible pattern of lower anxiety among evacuees engaging in higher levels of physical activity. Regression analysis identified higher resilience and lower anxiety as significant predictors of greater mental well-being. Additionally, residing in a community exposed to a higher number of traumatic events was associated with reduced well-being. The overall model explained a substantial portion of the variance in mental well-being.”

 

2.Lines 66-70: The author states: "Overcoming physical obstacles strengthens the belief that one can overcome barriers in other areas of life. Specifically, for populations that have experienced traumatic events, such as civilians evacuated from their homes, physical activity can serve as an important anchor for maintaining both physical and mental stability." This is a strong and general claim presented without any supporting citation. In the context of the introduction, where theoretical grounding is crucial, such statements should be substantiated with relevant literature. If empirical support exists, it should be cited. If this is the author’s conceptual framing or hypothesis, that should be made explicit to maintain academic transparency and avoid misleading the reader.

 

Response to Comment 2:

 

Thank you for this valuable comment. We agree that the sentence in question was a strong conceptual statement not directly supported by empirical evidence. As we aim to ensure that all claims in the introduction are evidence-based, we have removed the sentence to maintain academic rigor and avoid introducing unsupported assertions. The paragraph has been revised accordingly to preserve clarity and coherence without altering the referenced content.

This is the revised version of the paragraph:

Mental resilience is defined as an individual's ability to cope with stress, adapt to it, and grow from it [8]. In this context, physical activity plays a key role in strengthening personal resilience. First, it provides a sense of control, which is critical in traumatic life situations where individuals experience a loss of control over their circumstances. Regaining a sense of control through purposeful and planned activities, such as strength training or aerobic exercise, helps trauma survivors restore confidence in their ability to handle challenges [9]. Second, physical activity contributes to the improvement of coping abilities. Studies suggest that individuals who engage in regular physical activity develop higher mental resilience because they practice dealing with physical challenges, which translates into better psychological coping skills [10,11].

 

3.Section 2.1 – Study Population and Design. This subsection lacks essential descriptive information about the participants. Specifically, there is no mention of participants’ age (mean, standard deviation, and range), gender distribution, or the number of individuals from each group (Community A vs. Community B). These demographic details are critical for assessing the representativeness of the sample and for contextualizing the study’s findings. I recommend including a clear breakdown of these variables to enhance the transparency and interpretability of the study.

 

Response to Comment 3:

Thank you for this valuable comment. The demographics data was added.

This is the new version of the section:

This study included individuals aged 25 years and older who were residents of rural communities in a conflict-affected region of southern Israel. The mean age of participants was 46.2 years (SD = 12.9). Of the 107 participants, 77 (72.0%) were female and 30 (28.0%) were male. In terms of community origin, 43 participants (40.2%) were evacuated from "Community A," which was directly exposed to severe trauma and violence during the conflict. An additional 19 participants (17.8%) were from "Community B," a nearby locality within the same region that was not directly impacted by the traumatic events. The remaining 45 participants (42.1%) were evacuees from nine other rural communities within the broader conflict zone. Marital status was reported as follows: 85 participants (79.4%) were married or in a committed relationship, 8 (7.5%) were single, 9 (8.4%) were divorced, 4 (3.7%) were widowed, and 1 (0.9%) reported another status. Regarding education level, 22 participants (20.6%) had completed high school, 55 (51.3%) held a bachelor’s degree, 29 (27.1%) had a master’s degree or higher, and 1 participant (0.9%) reported no formal education. A detailed breakdown of demographic characteristics is presented in Table 1.

 

4.On this regard (lines 164-175), the description of the study sample presented in the Results section belongs in the methodology section, not under results. Participant demographics—such as age, gender distribution, education level, and pre-conflict employment status—are not study findings, but rather characteristics of the sample. I recommend relocating this paragraph to the appropriate subsection under Study Population and Design or creating a distinct Participants subsection within Methods.

 

Response to Comment 4:

Thank you for this valuable observation. We agree that a general description of the sample (e.g., age range, gender distribution, education level, and marital status) belongs in the Methods section and will relocate the corresponding narrative to a dedicated Participants subsection.

However, we respectfully suggest retaining the demographic table in the Results section, as several of the demographic variables presented (e.g., current security level, place of residence etc.) were later included in the regression analyses as independent variables. Presenting the full table within the Results section allows readers to clearly see the descriptive statistics of the variables that were used in the statistical analysis, thereby enhancing clarity and understanding of the findings. This approach is consistent with common academic reporting practices, particularly when demographic variables serve both as background characteristics and as predictors in the analysis.

 

5.Additionally, the term “predominantly women” is vague. Please clarify what proportion of participants were women by providing a specific percentage or count.

 

Response to Comment 5:

Thank you for the comment. The exact percentage of women was added as exhibited in the response to comment 3.

 

6.Lines 127–129 – the description of the GAD-7 used to assess anxiety symptoms is incomplete. Specifically, the manuscript should clearly state the response scale (e.g., Likert points, range of scores).

 

Response to Comment 6:

Thank you for this comment. The response scale for all questionnaires was added.

This is the new version of the paragraph:

The questionnaire comprised several sections.
The demographic section collected data on age, gender, marital status, education level, place of residence during the conflict, current residence, employment status prior to the conflict, whether family members had been drafted or affected by the conflict, and whether the current residence included a protected shelter.

The questionnaire comprised several sections.
The demographic section collected data on age, gender, marital status, education level, place of residence during the conflict, current residence, employment status prior to the conflict, whether family members had been drafted or affected by the conflict, and whether the current residence included a protected shelter.

Anxiety symptoms were measured using the Generalized Anxiety Disorder Screener (GAD-7) [22], a validated 7-item instrument assessing symptom severity over the past two weeks. Items were rated on a 4-point Likert scale (1 = never to 5 = nearly every day). The GAD-7 is widely used in clinical and research settings and has been validated in Hebrew (Cronbach α=.931.); it is also recommended by the Israeli Ministry of Health’s Professional Committee for Clinical Psychology.

Resilience was assessed with the Connor-Davidson Resilience Scale (CD-RISC) [23], which evaluates an individual's ability to adapt to adversity. Responses were rated on a 6-point Likert scale (1 = never to 6 = nearly all the time). The Hebrew version of the CD-RISC was translated and validated by the original developers (Cronbach α=.923) and has been widely used in Israeli research, demonstrating strong psychometric properties.

Well-being was measured using a 10-item well-being questionnaire [24] assessing emotional, psychological, and social aspects of well-being. Participants responded on a 5-point Likert scale (1 = agree to 5 = agree very much). The Hebrew version, adapted and validated by Florian and Drori (Cronbach α=.96) [25], has been shown to have good reliability and validity in the Israeli population.

 

 

7.Line 143: It appears that a comment from a previous review was inadvertently included in the body of the manuscript: "Your Statistical Analysis section is solid and clear… Here's an enhanced version." This language is clearly intended as reviewer feedback rather than part of the article itself. Please revise the text to remove any remnants of prior peer review comments and ensure that the section reflects only the finalized content intended for publication. Careful proofreading is advised before resubmission.

 

Response to Comment 7:

Thank you so much for this comment, this is an embarrassing typo and was deleted.

 

8.Lines 180–184: The manuscript reports the use of a one-way ANOVA to test for differences in resilience, anxiety, and well-being across three physical activity groups. While ANOVA is appropriate for comparing means across multiple groups, the relatively small sample sizes in each group (n=22, n=45, n=40) raise concerns about the statistical power and robustness of the test. The authors should clarify whether assumptions for ANOVA—such as normality and homogeneity of variances—were tested and met. If these assumptions were not verified, or if sample sizes are deemed too small, non-parametric alternatives (e.g., Kruskal-Wallis test) might be more suitable.

 

Response to Comment 8:

We thank the reviewer for this important and constructive comment.

In response, we carefully examined the distributional assumptions underlying the use of ANOVA.

Normality of the continuous variables (resilience, anxiety, and well-being) was assessed using skewness and kurtosis values within each physical activity group (non-exercisers, moderate exercisers, and high exercisers). Skewness values ranged between –0.164 and 1.143, and kurtosis values ranged between –1.38 and 1.059 across all variables and groups. These values fall within commonly accepted thresholds (|skewness| < 1, |kurtosis| < 3), indicating that the assumption of approximate normality was reasonably met.

Homogeneity of variances was tested using Levene’s test. The assumption was met for resilience (p = .787) and well-being (p = .526), but was violated for anxiety (p = .002). As a result, we adjusted our analytical approach accordingly: one-way ANOVAs were used to compare group means for resilience and well-being, whereas a non-parametric Kruskal–Wallis test was conducted to examine group differences in anxiety.

The Kruskal–Wallis test for anxiety did not reach statistical significance (H = 5.026, p = .081). Pairwise comparisons further revealed no statistically significant group differences after Bonferroni correction, though the comparison between the high and moderate activity groups approached significance (unadjusted p = .025; Bonferroni-adjusted p = .075), suggesting a potential trend.

Finally, to address concerns about sample size and statistical power, a post hoc power analysis was conducted. Results indicated that the study had adequate power (≥ 0.80) to detect medium to large effect sizes. Specifically, the observed effects for resilience (Cohen’s d = 0.71) and well-being (d = 0.92) were within the detectable range, and the effect for anxiety was small to medium (d = 0.32). These findings support the appropriateness and robustness of the statistical procedures employed, given the nature of the data and sample size.

Accordingly, we have revised the relevant paragraph in the Statistical Analysis section of the Methods to reflect these clarifications:

"Descriptive statistics (means, standard deviations, and frequencies) were calculated to summarize demographic characteristics and key study variables. Normality of the continuous variables (resilience, anxiety, and well-being) was assessed using skewness and kurtosis values. All values fell within acceptable ranges (|skewness| < 1, |kurtosis| < 3), indicating that the assumption of approximate normality was met across the three physical activity groups (non-exercisers, moderate exercisers, and high exercisers). Homogeneity of variances was tested using Levene’s test, which indicated that the assumption was met for resilience (p = .787) and well-being (p = .526), but not for anxiety (p = .002). Therefore, in addition to the one-way ANOVA, a non-parametric Kruskal–Wallis test was conducted for anxiety.

To examine group differences in psychological variables across levels of physical activity (in terms of total duration and intensity), one-way analyses of variance (ANOVA) were conducted for resilience and well-being, as the assumptions of normality and homogeneity of variances were met. For anxiety, the Kruskal–Wallis test was used due to violation of the homogeneity of variances assumption. Pearson correlation coefficients were computed to assess associations among anxiety, resilience, well-being, and physical activity variables. To evaluate the predictive value of physical activity and psychosocial factors on mental well-being, a multiple linear regression analysis was performed. Predictor variables included weekly physical activity duration, frequency of physical activity, duration of vigorous-intensity activity, anxiety levels, resilience levels, perceived level of security in the current residence, and place of residence. Place of residence was categorized as “Community A” (high exposure to trauma and violence) and “Community B” (a nearby area not directly impacted by the conflict).

A post hoc power analysis was conducted to evaluate whether the sample size (N = 107; group sizes n = 22, 45, and 40) provided sufficient statistical power for the primary analyses. Based on the observed group means and standard deviations, effect sizes were estimated as follows: anxiety (Cohen’s d = 0.32, small to medium), resilience (d = 0.71, medium to large), and well-being (d = 0.92, large). These values indicate that the study had adequate statistical power (≥ 0.80) to detect medium to large effects using both ANOVA and regression models. All statistical tests were two-tailed, and statistical significance was set at p < .05."

 

 

 

9.Additionally, the authors should elaborate on the rationale for grouping participants in this way and justify the selected cutoffs for “moderate” and “high” physical activity. Clarifying these methodological choices would strengthen the validity of the results. Importantly, the results section does not report that no significant differences were found when comparing the “no physical activity” group to others.

 

Response to Comment 9:

Thank you for this valuable comment.

The classification of physical activity levels into "moderate" and "high" was based on the World Health Organization (WHO) guidelines for weekly physical activity duration. This clarification has been added to the Methods section as follows:

“Participants reported their average weekly duration of physical activity (in minutes) and the amount of time spent in vigorous-intensity exercise, defined as activities inducing heavy breathing and a noticeably elevated heart rate, as recommended by the World Health Organization (WHO), [28].”

 

In addition, to address the reviewer’s concern regarding comparisons involving the “non-exercisers” group, we have added the following sentence to the Results section:

“No significant differences were found when comparing the non-exercisers group to the other activity groups in certain outcome variables.”

These clarifications aim to enhance the transparency and rigor of the methodology and the interpretation of findings.

 

 

10.Lines 212–213: The manuscript describes the Pearson correlation coefficient (r = 0.296, p = 0.002) between resilience levels and weekly duration of intense physical activity as a "weak" relationship. While the correlation coefficient indicates a modest association, given the sample size, this statistically significant correlation may hold meaningful practical relevance. In psychological and behavioral studies, correlations of this magnitude can be important. Therefore, I recommend revising the description to reflect this nuance—for example, referring to the relationship as “a small to moderate but statistically significant positive association”—to better capture its significance in context.

 

Response to Comment 10:

Thank you very much for this insightful and constructive comment, which highlights the relevance of effect sizes in psychological research.

 The description in the manuscript has been revised accordingly to better reflect the magnitude and significance of the observed association. The revised sentence now refers to the correlation between resilience and vigorous physical activity as “a small to moderate but statistically significant positive association,” aligning with conventions in psychological and behavioral sciences.

This is the revised section:

“The Pearson's correlation coefficient measurement indicated a small to moderate but statistically significant positive association between resilience levels and the total weekly physical activity level (r=0.300, P=0.002), as well as between resilience levels and weekly exercise duration involving vigorous-intense physical activity causing heavy breathing (r=0.296, P=0.002), suggesting that as the total exercise time and time spent on vigorous-intense physical activities causing heavy breathing increased, resilience levels also rose (Table 2).

Similarly, a small to moderate but statistically significant positive association was found between well-being levels and weekly exercise duration (r=0.264, P=0.006), as well as between well-being and time spent on vigorous-intense physical activity (r=0.312, P=0.001), indicating that longer exercise durations and higher-intensity exercise were associated with higher levels of well-being and resilience.

Regarding anxiety, a small but statistically significant negative correlation was observed between the time spent on physical activity per week and anxiety levels (r=-0.199, P=0.04), indicating that as the duration of physical activity increased, anxiety levels decreased. We also found a small to moderate negative correlation between anxiety levels and resilience (r=-0.33, P<0.001), as well as between anxiety levels and well-being (r=-0.473, P<0.001). In other words, as resilience and well-being increased, anxiety levels decreased. Finally, a moderate positive association was found between resilience and well-being (r=0.553, P<0.001), indicating that as resilience levels increased, so did well-being levels.”

 

 

11.Furthermore, there is an inconsistency in how correlation strengths are characterized. The correlation between resilience and intense physical activity (r = 0.296) is described as “weak,” whereas the lower negative correlation between physical activity duration and anxiety (r = –0.199, p = 0.04) is presented without any qualifier regarding its strength (lines 223-224).

 

Response to Comment 11:

Thank you very much for this helpful comment. To ensure consistency in describing the strength of correlations, the manuscript has been revised accordingly. A description of the correlation’s strength has been added to the relevant sentence in the Results section, and the correlation between physical activity and anxiety is now described as a “small but statistically significant negative association.”

Regarding anxiety, a small but statistically significant negative correlation was observed between the time spent on physical activity per week and anxiety levels (r=-0.199, P=0.04), indicating that as the duration of physical activity increased, anxiety levels decreased. We also found a small to moderate negative correlation between anxiety levels and resilience (r=-0.33, P<0.001), as well as between anxiety levels and well-being (r=-0.473, P<0.001). In other words, as resilience and well-being increased, anxiety levels decreased. Finally, a moderate positive association was found between resilience and well-being (r=0.553, P<0.001), indicating that as resilience levels increased, so did well-being levels.”

 

 

 

 

12. Table 2 currently includes both p-values and asterisks (*) indicating statistical significance. To improve clarity and reduce redundancy, it is advisable to present only one of these indicators. Typically, the use of asterisks to denote significance levels is sufficient, provided that the table legend or footnote clearly explains what each number of asterisks represents (e.g., *p< 0.05, **p < 0.01). Please include such a description in the table caption or footnote to guide readers.

 

Response to Comment 12:

Thank you for this valuable insight. The significance values were removed from Table 2, and only the asterisks were retained to indicate statistical significance. A corresponding footnote has been added below the table to clarify the meaning of the asterisks: **p < 0.01.

 

13.Lines 254-255: The statement, “Our findings emphasize the significant role of physical activity in promoting psychological health during times of crisis,” overstates the conclusions given the relatively small sample size. It would be more appropriate to use more cautious language, such as “Our findings suggest a potential role…” or “These results indicate that physical activity may contribute to psychological health…” to better reflect the exploratory nature of the study and its limitations.

 

Response to Comment 13:

Thank you very much for this comment and for reminding us to be more cautious and humbler in interpreting our findings. We agree that the original phrasing may have overstated the conclusions, given the exploratory nature of the study and its sample size. The sentence has been revised to read:
“Our findings indicate that physical activity may contribute to psychological health during times of crisis.”

 

14.The limitations section does not address the unequal distribution of participants, not only across physical activity groups (notably the smaller “no physical activity” group) but also between Community A and Community B. These imbalances may impact the statistical power and generalizability of the findings. It is important to explicitly acknowledge these sample distribution issues to provide a more thorough discussion of the study’s limitations.

 

Response to Comment 14:

Thank you for this important observation. We agree that the unequal distribution of participants—both across physical activity groups and between Community A and Community B—may influence statistical power and the generalizability of the findings. We have now revised the Limitations section to explicitly acknowledge and discuss these imbalances, their implications, and the contextual factors that contributed to them. Please see the updated limitation section:

“Several limitations should be considered. First, the cross-sectional design precludes causal inference and captures psychological states at a single point—approximately two months post-event—while mental health trajectories may evolve over time. Second, convenience sampling—necessitated by the dispersed nature of evacuees and the urgent post-crisis context—may introduce selection bias and limit generalizability. Recruitment occurred in temporary evacuee hotels under emotionally charged and logistically constrained conditions, which influenced participation willingness and contributed to a relatively small overall sample size.

In addition, uneven distribution across key subgroups should be acknowledged. Specifically, participant representation from different communities was imbalanced, with a higher proportion from Community A (directly affected) and fewer from Community B and other rural areas. Likewise, the distribution across physical activity levels was unequal, with notably fewer participants in the "no physical activity" group. These disparities may reduce statistical power and affect the generalizability of subgroup comparisons.

Furthermore, unmeasured confounding variables—such as individual coping strategies, social support, and pre-existing mental health conditions—could influence the associations observed. Ethical and emotional considerations prevented the collection of detailed psychological history, as participant well-being and minimizing retraumatization were prioritized. As a result, potential confounders could not be controlled for, which may have affected internal validity.

Despite these limitations, several strengths enhance the value of this study. The sample included a relatively diverse demographic profile in terms of gender, marital status, education, and occupational backgrounds, allowing for a broader understanding of physical activity’s role across different population segments. Additionally, this is one of the first studies to explore the interplay of physical activity, resilience, and mental health specifically among civilians displaced by conflict in this region, providing novel insights with potential international relevance.
Future research should employ longitudinal designs to monitor changes in physical activity, resilience, and psychological health over time, helping to establish causal pathways. Additionally, examining potential mediators such as social support, coping mechanisms, and personality traits could clarify how physical activity exerts its effects. Intervention studies integrating physical activity with psychosocial therapies are warranted to evaluate efficacy in improving mental health outcomes in trauma-affected populations. Confirmatory studies should also explore optimal types, durations, and intensities of physical activity to maximize resilience and well-being benefits. These efforts will help inform tailored intervention strategies and public health policies in conflict and disaster settings.”

 

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript "Mental Health, Resilience, and Physical Activity in Civilians Affected by Conflict-Related Trauma: A Cross-Sectional Study" explores the relationship between physical activity, anxiety, resilience, and general well-being among evacuated civilians from a conflict-affected region in Israel. It’s a relevant and scientifically timely topic. The manuscript certainly shows potential but requires substantial revision before publication can be justified.

In the abstract, I miss the inclusion of clearly formulated, explicit hypotheses.

The aim of the study is clearly stated in the introduction, namely the exploration of the relationship between physical activity and mental health outcomes among evacuees. However, explicit research questions or testable hypotheses are missing, at least as far as I could see. Nevertheless, it would be quite straightforward to formulate a hypothesis, for example: We hypothesize that higher levels of physical activity are associated with lower anxiety, higher resilience, and greater well-being. This is merely a suggestion for the author.

The literature used is up to date. However, it is noticeable that some citations remain general or are selectively chosen. For instance, there is little attention given to mixed or null findings in the literature on sport-based interventions for trauma. Could the authors take another critical look at this? As inspiration, I would like to draw the author’s attention to the following recent study:
Janković et al. (2024). Patterns of sports participation across the lifespan and their links to psychopathology, resilience, and childhood trauma: a cross-sectional approach. International Journal of Sport and Exercise Psychology, 1–22. https://doi-org.tilburguniversity.idm.oclc.org/10.1080/1612197X.2024.2431252

The methods section is clearly structured and describes the participants, instruments, and statistical analyses. However, I miss a power analysis; without it, it’s unclear whether the sample of 107 participants is sufficient to detect the presumed effects. This is especially important given the relatively complex regression analyses used.

Participants are described as evacuees from rural communities in southern Israel. The sample was obtained using convenience sampling, which increases the risk of selection bias. Although this is acknowledged in the discussion, no attempt was made to implement a methodological correction, such as weighting demographic variables. Additionally, there is no information provided about participants’ psychological history, which is important in trauma research. This deserves attention.

The description of the instruments is mostly adequate, but the operationalization of physical activity requires improvement. A “brief self-report measure adapted from a previous study” is used, but — if I interpreted this correctly, it refers to an internal publication by the author regarding students during the COVID lockdown, without clear validation. Could the author address this issue?

The statistical analysis techniques, ANOVA, Pearson correlations, and regression, are appropriate for the study’s objectives. However, the assumptions of these analyses (such as normality, linearity, or homoscedasticity) are not tested or reported anywhere. This step is essential in both regression and variance analyses. Moreover, no interaction effects (e.g., between gender and activity) are investigated, although such effects can be highly relevant in stress-related research.

The results are well presented with clear tables and figures. A clear distinction is made between total physical activity and vigorous activity (defined as causing “heavy breathing”). The correlations are statistically significant but weak (e.g., r = 0.30 for activity and resilience) and are not critically interpreted as such. Additionally, the order in which the findings are presented is inconsistent, sometimes vigorous activity is discussed first, and other times total activity, which can be confusing for the reader.

The discussion links the findings back to the existing literature and does a good job explaining how physical activity may yield both physiological and psychosocial benefits for trauma-exposed populations. However, there is still room for improvement. First, the discussion hardly addresses the non-significant results, such as the absence of an effect of vigorous activity on anxiety. Interpreting such null findings is crucial. Second, the manuscript frequently uses overly causal language, despite being based on a cross-sectional design. Terms such as “physical activity leads to...” should be avoided; instead, one should say “is associated with...”.

The language used in the manuscript is generally clear.

Returning briefly to the methods: how was missing data handled? Could this be clarified?

Author Response

Healthcare- 3689498

 

 Dear Reviewer

Many thanks to your thoughtful and constructive feedback. The detailed comments contributed significantly to enhancing the clarity, depth, and academic rigor of the manuscript. All suggestions were carefully considered, and appropriate revisions have been made accordingly. Responses to each comment appear below.

 

1.In the abstract, I miss the inclusion of clearly formulated, explicit hypotheses.

 

Response to Comment 1:

Thank you for this important comment. The following sentence has been added to the abstract to clearly state the study’s hypothesis: “We hypothesized that higher levels of  intense physical activity would be associated with higher levels of resilience and well-being and lower levels of anxiety.

 

2.The aim of the study is clearly stated in the introduction, namely the exploration of the relationship between physical activity and mental health outcomes among evacuees. However, explicit research questions or testable hypotheses are missing, at least as far as I could see. Nevertheless, it would be quite straightforward to formulate a hypothesis, for example: We hypothesize that higher levels of physical activity are associated with lower anxiety, higher resilience, and greater well-being. This is merely a suggestion for the author.

 

Response to Comment 2:

Thank you for this important comment. The following sentence has also been added to the end of the Introduction to clearly state the study’s hypothesis “We hypothesized that higher levels of physical activity would be associated with higher levels of resilience and well-being and lower levels of anxiety.”

 

3.The literature used is up to date. However, it is noticeable that some citations remain general or are selectively chosen. For instance, there is little attention given to mixed or null findings in the literature on sport-based interventions for trauma. Could the authors take another critical look at this? As inspiration, I would like to draw the author’s attention to the following recent study:
Janković et al. (2024). Patterns of sports participation across the lifespan and their links to psychopathology, resilience, and childhood trauma: a cross-sectional approach. International Journal of Sport and Exercise Psychology, 1–22. https://doi-org.tilburguniversity.idm.oclc.org/10.1080/1612197X.2024.2431252

 

Response to Comment 3:

We greatly appreciate this insightful and constructive comment. Following your recommendation, we conducted a more critical review of the literature on sport-based interventions, with particular attention to studies reporting mixed or null findings. This process led to the incorporation of several additional peer-reviewed sources, including the valuable study by Janković et al. (2024), which you kindly suggested. Their work provided a nuanced perspective on patterns of sports participation across the lifespan and their variable associations with psychological outcomes. Furthermore, we integrated evidence from Eime et al. (2010) and Cekin (2015), which highlight differential effects depending on the type and context of physical activity. These studies were incorporated at relevant points in the revised discussion to better reflect the complexity and variability in the literature. This revision has strengthened the discussion by offering a more balanced and evidence-based interpretation of our findings.

Please see the updated discussion.

 

4.The methods section is clearly structured and describes the participants, instruments, and statistical analyses. However, I miss a power analysis; without it, it’s unclear whether the sample of 107 participants is sufficient to detect the presumed effects. This is especially important given the relatively complex regression analyses used.

 

Response to Comment 4:

Thank you for this important observation. The study employed ANOVA, Pearson correlations, and linear regression analyses to examine differences and associations among psychological variables across physical activity groups. A total sample of N = 107 was included, with group sizes of n = 22, 45, and 40, respectively. While some variation existed between group sizes, a post hoc power analysis indicated that the study had adequate statistical power (≥ 0.80) to detect medium to large effect sizes—levels that align with the effects observed in the data.

Specifically, based on the group means and standard deviations:

• The difference in anxiety levels reflected a small to medium effect (Cohen’s d = 0.32).
• The difference in resilience reflected a medium to large effect (d = 0.71).
• The difference in well-being reflected a large effect (d = 0.92).

These results suggest that the study was sufficiently powered to detect the observed effects, particularly for resilience and well-being.

This is the revised paragraph in the Statistical Analysis section of the Methods:

A post hoc power analysis was conducted to evaluate whether the sample size (N = 107; group sizes n = 22, 45, and 40) provided sufficient statistical power for the primary analyses. Based on the observed group means and standard deviations, effect sizes were estimated as follows: anxiety (Cohen’s d = 0.32, small to medium), resilience (d = 0.71, medium to large), and well-being (d = 0.92, large). These values indicate that the study had adequate statistical power (≥ 0.80) to detect medium to large effects using both ANOVA and regression models. All statistical tests were two-tailed, and statistical significance was set at p < .05.

 

5.Participants are described as evacuees from rural communities in southern Israel. The sample was obtained using convenience sampling, which increases the risk of selection bias. Although this is acknowledged in the discussion, no attempt was made to implement a methodological correction, such as weighting demographic variables. Additionally, there is no information provided about participants’ psychological history, which is important in trauma research. This deserves attention.

 

Response to Comment 5:

 

We appreciate the reviewer’s comment regarding sampling and the absence of data on participants' psychological history. As noted, the sample was obtained through convenience sampling due to the urgent and unstable conditions following the events of October 7. Participants were recruited in evacuee hotels during a highly sensitive period, only a few weeks after the traumatic incident, which imposed ethical and logistical constraints on data collection.

While we acknowledge that this approach increases the risk of selection bias, our ability to apply corrective statistical methods such as weighting demographic variables was limited. First, reliable population-level demographic data for the evacuees was not available, and second, the primary goal was to reduce participant burden and prioritize emotional safety. Collecting detailed psychological history was intentionally avoided out of concern for retraumatization and ethical responsibility, given the acute nature of the crisis and participants’ psychological vulnerability.

These limitations are acknowledged in the discussion section and framed within the context of conducting research in emergency and post-traumatic environments. We believe this study nevertheless provides valuable insight into the psychological effects of displacement during a large-scale security crisis.

 

6.The description of the instruments is mostly adequate, but the operationalization of physical activity requires improvement. A “brief self-report measure adapted from a previous study” is used, but — if I interpreted this correctly, it refers to an internal publication by the author regarding students during the COVID lockdown, without clear validation. Could the author address this issue?

 

Response to Comment 6:

 

Thank you for this important and helpful comment. The physical activity questionnaire used in this study was the International Physical Activity Questionnaire – Short Form (IPAQ-SF), originally developed by Booth (2000) for international use in physical activity surveillance [reference added]. The questionnaire was used in its Hebrew-translated version, which demonstrated acceptable validity and reliability in a previous peer-reviewed study conducted among undergraduate students in Israel (Joseph & Schori, 2022). The revised manuscript now includes a clearer and more accurate description of the measure, its origin, and psychometric properties.

This is the revised section:

“Physical activity was assessed using the International Physical Activity Questionnaire – Short Form (IPAQ-SF), a widely used and validated self-report instrument developed for international comparison of physical activity levels [26]. The questionnaire was administered in its Hebrew-translated version, which has previously demonstrated acceptable validity and reliability in a study among Israeli undergraduate students [27].

This reference was added:

26. Booth M. Assessment of physical activity: an international perspective. Res Q Exerc Sport 2000 Jun;71 Suppl 2:114-120. [doi: 10.1080/02701367.2000.11082794] [Medline: 25680021]

 

 

7.The statistical analysis techniques, ANOVA, Pearson correlations, and regression, are appropriate for the study’s objectives. However, the assumptions of these analyses (such as normality, linearity, or homoscedasticity) are not tested or reported anywhere. This step is essential in both regression and variance analyses. Moreover, no interaction effects (e.g., between gender and activity) are investigated, although such effects can be highly relevant in stress-related research.

 

Response to Comment 7:

We thank the reviewer for this important and constructive comment.

In response, we carefully examined the distributional assumptions underlying the use of ANOVA.

Normality of the continuous variables (resilience, anxiety, and well-being) was assessed using skewness and kurtosis values within each physical activity group (non-exercisers, moderate exercisers, and high exercisers). Skewness values ranged between –0.164 and 1.143, and kurtosis values ranged between –1.38 and 1.059 across all variables and groups. These values fall within commonly accepted thresholds (|skewness| < 1, |kurtosis| < 3), indicating that the assumption of approximate normality was reasonably met.

Homogeneity of variances was tested using Levene’s test. The assumption was met for resilience (p = .787) and well-being (p = .526) but was violated for anxiety (p = .002). As a result, we adjusted our analytical approach accordingly: one-way ANOVAs were used to compare group means for resilience and well-being, whereas a non-parametric Kruskal–Wallis test was conducted to examine group differences in anxiety.

The Kruskal–Wallis test for anxiety did not reach statistical significance (H = 5.026, p = .081). Pairwise comparisons further revealed no statistically significant group differences after Bonferroni correction, though the comparison between the high and moderate activity groups approached significance (unadjusted p = .025; Bonferroni-adjusted p = .075), suggesting a potential trend.

Finally, to address concerns about sample size and statistical power, a post hoc power analysis was conducted. Results indicated that the study had adequate power (≥ 0.80) to detect medium to large effect sizes. Specifically, the observed effects for resilience (Cohen’s d = 0.71) and well-being (d = 0.92) were within the detectable range, and the effect for anxiety was small to medium (d = 0.32). These findings support the appropriateness and robustness of the statistical procedures employed, given the nature of the data and sample size.

Accordingly, we have revised the relevant paragraph in the Statistical Analysis section of the Methods to reflect these clarifications:

"Descriptive statistics (means, standard deviations, and frequencies) were calculated to summarize demographic characteristics and key study variables. Normality of the continuous variables (resilience, anxiety, and well-being) was assessed using skewness and kurtosis values. All values fell within acceptable ranges (|skewness| < 1, |kurtosis| < 3), indicating that the assumption of approximate normality was met across the three physical activity groups (non-exercisers, moderate exercisers, and high exercisers). Homogeneity of variances was tested using Levene’s test, which indicated that the assumption was met for resilience (p = .787) and well-being (p = .526), but not for anxiety (p = .002). Therefore, in addition to the one-way ANOVA, a non-parametric Kruskal–Wallis test was conducted for anxiety.

To examine group differences in psychological variables across levels of physical activity (in terms of total duration and intensity), one-way analyses of variance (ANOVA) were conducted for resilience and well-being, as the assumptions of normality and homogeneity of variances were met. For anxiety, the Kruskal–Wallis test was used due to violation of the homogeneity of variances assumption. Pearson correlation coefficients were computed to assess associations among anxiety, resilience, well-being, and physical activity variables. To evaluate the predictive value of physical activity and psychosocial factors on mental well-being, a multiple linear regression analysis was performed. Predictor variables included weekly physical activity duration, frequency of physical activity, duration of vigorous-intensity activity, anxiety levels, resilience levels, perceived level of security in the current residence, and place of residence. Place of residence was categorized as “Community A” (high exposure to trauma and violence) and “Community B” (a nearby area not directly impacted by the conflict).

A post hoc power analysis was conducted to evaluate whether the sample size (N = 107; group sizes n = 22, 45, and 40) provided sufficient statistical power for the primary analyses. Based on the observed group means and standard deviations, effect sizes were estimated as follows: anxiety (Cohen’s d = 0.32, small to medium), resilience (d = 0.71, medium to large), and well-being (d = 0.92, large). These values indicate that the study had adequate statistical power (≥ 0.80) to detect medium to large effects using both ANOVA and regression models. All statistical tests were two-tailed, and statistical significance was set at p < .05."

In addition, Interaction effects between gender and the main study variables were tested; however, no statistically significant interactions were found. As a result, these analyses were not included in the manuscript in order to maintain clarity and focus. We appreciate the reviewer’s point and will consider presenting interaction effects in future studies where relevant findings emerge.

 

8.The results are well presented with clear tables and figures. A clear distinction is made between total physical activity and vigorous activity (defined as causing “heavy breathing”). The correlations are statistically significant but weak (e.g., r = 0.30 for activity and resilience) and are not critically interpreted as such. Additionally, the order in which the findings are presented is inconsistent, sometimes vigorous activity is discussed first, and other times total activity, which can be confusing for the reader.

 

Response to Comment 8:

Thank you for your thoughtful feedback. The Results section has been revised to ensure a consistent structure: total weekly physical activity is now presented before vigorous activity throughout the text. In addition, all correlations are now described with appropriate interpretation of their magnitude, including the statistically significance.

The Pearson's correlation coefficient measurement indicated a small to moderate but statistically significant positive association between resilience levels and the total weekly physical activity level (r=0.300, P=0.002), as well as between resilience levels and weekly exercise duration involving vigorous-intense physical activity causing heavy breathing (r=0.296, P=0.002), suggesting that as the total exercise time and time spent on vigorous-intense physical activities causing heavy breathing increased, resilience levels also rose (Table 2).

Similarly, a small to moderate but statistically significant positive association was found between well-being levels and weekly exercise duration (r=0.264, P=0.006), as well as between well-being and time spent on vigorous-intense physical activity (r=0.312, P=0.001), indicating that longer exercise durations and higher-intensity exercise were associated with higher levels of well-being and resilience.

Regarding anxiety, a small but statistically significant negative correlation was observed between the time spent on physical activity per week and anxiety levels (r=-0.199, P=0.04), indicating that as the duration of physical activity increased, anxiety levels decreased. We also found a small to moderate negative correlation between anxiety levels and resilience (r=-0.33, P<0.001), as well as between anxiety levels and well-being (r=-0.473, P<0.001). In other words, as resilience and well-being increased, anxiety levels decreased. Finally, a moderate positive association was found between resilience and well-being (r=0.553, P<0.001), indicating that as resilience levels increased, so did well-being levels.

 

 

9.The discussion links the findings back to the existing literature and does a good job explaining how physical activity may yield both physiological and psychosocial benefits for trauma-exposed populations. However, there is still room for improvement. First, the discussion hardly addresses the non-significant results, such as the absence of an effect of vigorous activity on anxiety. Interpreting such null findings is crucial. Second, the manuscript frequently uses overly causal language, despite being based on a cross-sectional design. Terms such as “physical activity leads to...” should be avoided; instead, one should say “is associated with...”.

The language used in the manuscript is generally clear.

 

Response to Comment 9:

We thank the reviewer for this thoughtful and constructive comment. In response, we have strengthened our discussion of the non-significant association between vigorous physical activity and anxiety, emphasizing that this result should be interpreted with caution and highlighting potential explanations for the lack of statistical significance, such as individual differences and possible nonlinear effects.

We also carefully reviewed the manuscript for any instances of causal language. Where needed, we revised the wording to ensure consistency with the cross-sectional nature of the study. Expressions such as “leads to” were replaced with “is associated with,” and causal interpretations were avoided throughout the discussion. We appreciate the reviewer’s helpful guidance in ensuring the clarity and accuracy of our interpretations

Please see the updated discussion.

 

10.Returning briefly to the methods: how was missing data handled? Could this be clarified?

 

Response to Comment 10:

Thank you for this helpful comment. We confirm that there were no missing data in the dataset. The online questionnaire was designed so that it could not be submitted unless all items were completed. As an additional measure, responses were reviewed for completeness at the time of collection, and only fully completed questionnaires were included in the analysis.

 

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I am happy with the author's detailed response to the review comments. I am content for the article to go forward for publication.

Author Response

I would like to express my sincere gratitude to the reviewer for the thoughtful, insightful, and constructive feedback. The valuable comments and suggestions significantly contributed to improving the clarity, depth, and overall quality of the manuscript. I truly appreciate the time and expertise invested in reviewing this work.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have made a commendable effort to address all the comments from the previous review. Because of these revisions, I believe the paper should be accepted for publication.

Author Response

I would like to express my sincere gratitude to the reviewer for the thoughtful, insightful, and constructive feedback. The valuable comments and suggestions significantly contributed to improving the clarity, depth, and overall quality of the manuscript. I truly appreciate the time and expertise invested in reviewing this work.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors made many changes, and I compliment them for that! They have included a hypothesis in the abstract, the RQ is now clearer in the introduction, with a reference to hypotheses. Well done! The literature section has been updated, and it’s good to read that the authors now discuss mixed/negative findings about physical activity and trauma in a more balanced way. This means the literature no longer gives only a one-sided positive view.

Great that a post-hoc power analysis was carried out, this way, readers know that the sample size is large enough to detect medium/large effects.

The authors again recognize the possible selection bias because they used convenience sampling. It’s a pity there is still no information about the psychological history of the participants, but for me, this is not a reason to reject the manuscript. My concern abaut the measure to assess physical activity has been resolved. The statistical assumptions are now reported. The results are presented clearly. The discussion now also pays attention to non-significant results. The wording has been adjusted as well. 

However, one point remains important for me, namewly how missing data were handled. In the revised manuscript, I can't see if there were missing data or not. It would be enough if the authors add just one sentence in the section ‘Statistical Analysis’ section saying that there were no missing data, and, If there are missing data, they should clarify how they dealt with it. This is the only issue that remains. 

Author Response

I would like to express my sincere gratitude to the reviewer for the thoughtful, insightful, and constructive feedback. The valuable comments and suggestions significantly contributed to improving the clarity, depth, and overall quality of the manuscript. I truly appreciate the time and expertise invested in reviewing this work.

 

Response to comment 1:

Thank you for this important comment. As requested, a clarifying sentence was added to the beginning of the Statistical Analysis section, stating that only fully completed questionnaires were included in the study, and therefore, there were no missing data to address.

This is the sentence added: “Only fully completed questionnaires were included in the analysis; therefore, there were no missing data to address.”