Fear of Dying and Catastrophic Thinking Are Associated with More Severe Post-Traumatic Stress Symptoms Following COVID-19 Infection

Round 1

Reviewer 1 Report

The manuscript is well done and makes an important contribution to our literature. But it may overstate the results. See my comments on the title.

see my comments regarding the title

Author Response

Reviewer 1

 

1. The reviewer suggested changing the title so that the text did not imply that causal relations were found among study variables.

 

In the revised manuscript, the title has been changed to, “Fear of Dying and Catastrophic Thinking are Associated with more Severe Post-Traumatic Stress Symptoms Following COVID-19 Infection.

 

2. The reviewer recommended editing the manuscript to ensure that the text described ‘relations’ or ‘associations’ among variables without implying support for causal relations.

 

In the revised manuscript, the text has been edited to more accurately reflect that the results describe ‘relations’ among variables, not causal or temporal relations. The terminology of ‘prediction’ is retained only for describing the results of the regression analysis and in conceptual speculation in the Discussion section.

Reviewer 2 Report

The study examines the role of fear of death and catastrophic thinking as predictors of post-traumatic stress symptoms (PTSS) severity in individuals previously infected with COVID-19, with the aim of identifying modifiable psychological factors relevant to clinical prevention. Overall, the study is conceptually and clinically interesting, as it addresses timely and relevant psychological mechanisms in the context of post-COVID-19 mental health.

The authors could further elaborate on the theoretical rationale underlying the link between fear of death and catastrophic thinking. In particular, it would be helpful to reference integrative theoretical models that could provide a clearer conceptual framework for understanding the interaction between these two predictive variables (see, for example: https://doi.org/10.1016/j.paid.2025.113251; https://doi.org/10.31083/j.jin2305105).

From a methodological perspective, the cross-sectional design employed limits the ability to draw causal inferences. It is advisable that this limitation be explicitly stated in the abstract and that a longitudinal study be proposed as a potential future direction to confirm the directionality of the observed effects.

Regarding the measurement of the variable “fear of death,” the use of a dichotomous scale (yes/no) appears overly simplistic, especially considering the theoretical complexity of this construct. For future studies, the use of validated psychometric instruments capable of capturing the intensity and qualitative nuances of death anxiety—such as the Death Anxiety Scales—is recommended.

Statistical analyses were conducted appropriately, and the use of hierarchical regression proved effective for assessing the incremental contribution of psychological variables. However, the interpretation of results could be further enriched by incorporating mediation or moderation analyses. A relevant example would be to explore whether catastrophic thinking mediates the relationship between physical symptoms and post-traumatic stress symptoms.

The sample consists exclusively of Canadian participants with a high level of education, which may limit the generalizability of the findings to other cultural and socioeconomic contexts. It would be appropriate to include a reflection on how such factors might influence illness perception and emotional responses to the disease.

Finally, it is suggested to include specific examples of brief, validated interventions targeting the reduction of catastrophic thinking or death-related anxiety, which could be adapted to the specific context of the COVID-19 pandemic.

Minore issues

The reference list does not adhere to APA 7 style; it is recommended to adjust citation and reference formatting accordingly.
Statistical indicators such as p, r, β, and R² should be italicized in accordance with APA conventions.

Author Response

Reviewer 2

 

1. The reviewer recommended referencing an integrative theoretical model that could provide a clearer conceptual framework for understanding the potential role of fear of dying and catastrophic thinking the development of PTSD/PTSS.

 

In the revised manuscript, Ehlers and Clark’s (2000) cognitive model of PTSD is referenced as an integrative model of the potential role of fear of dying and catastrophic thinking in the development of PTSD/PTSS (lines 81-86).

 

2. The reviewer recommended addressing the limitations of the cross-sectional design in the Abstract and that a longitudinal study be proposed as a potential future direction to confirm the directionality of the observed effects.

 

In the revised manuscript, the following sentence was added to the Abstract, “The cross-sectional design of the study precludes statements about causality and conclusions about temporal relations among variables must await replication in a longitudinal design.” (lines 25-27).

 

3. In future studies, the reviewer recommended the use of validated psychometric instruments capable of capturing the intensity and qualitative nuances of death anxiety, such as the Death Anxiety Scales.

 

The reviewer’s recommendation has been noted.

 

4. The reviewer recommended exploring whether catastrophic thinking mediates the relationship between physical symptoms and post-traumatic stress symptoms.

 

Examination of the changes in beta weights through successive steps of the hierarchical regression reveals that the beta weight for initial physical symptoms is no longer significant when ongoing physical symptoms enters the equation. There is considerable variance overlap between initial and ongoing physical symptoms since ongoing physical symptoms is a subset of initial physical symptoms. In a direct regression analysis, the beta values for ongoing physical symptoms and catastrophic thinking remain significant, thus providing no support for mediation. In the revised manuscript, semi-partial correlations for the final regression equation are provided to permit examination of the unique contributions of the variables in the regression equation.

 

5. The reviewer noted that participants consisted of Canadians with a high level of education, which may limit the generalizability of the findings to other cultural and socioeconomic contexts.

 

In the revised manuscript, this limitation is noted (lines 400-402)

 

6. The reviewer suggested noting brief validated interventions that could be used to reduce catastrophic thinking or fear of death that could be adapted to the context of COVID.

 

In the revised manuscript, we address the question of brief interventions to reduce catastrophic thinking and fear of death that could be adapted to the context of COVID. (lines 388-395)

 

7. The reviewer noted that in APA style, statistical indicators such as p, r, β, and R²should be italicized.

 

In the revised manuscript, statistical indicators such as p, r, β, and R² have been italicized.

 

Reviewer 3 Report

1. The authors should briefly describe the psychometric soundness of their single item indicator of the “fear of dying.”
2. The authors should justify their decision to measure fear of death due to COVID-19 rather than a more general measure of “death anxiety,” “fear of death,” etc. (In the introduction, findings pertaining to the fear of death seem to reflect a more general concern rather than one related to a specific illness.)
3. The authors appear to be using sex and gender interchangeably (see lines 170 and 193). If participants were asked whether they identified as a man, woman, etc., then the authors should use “gender” throughout. If participants were asked whether they were male, female, etc., then “sex” should be used.
4. A scale score reliability coefficient (e.g., McDonald’s omega) should be provided for the measure of catastrophic thinking.
5. Given that fear of dying is a dichotomous variable, when examining its association with continuous variables, point biserial correlations should be computed.
6. The authors should provide effect sizes for all t-tests (e.g., Cohen’s d), chi-square tests (e.g., Cramer’s V), etc.
7. For the chi-square analyses, degrees of freedom also should be provided.
8. For some of the chi-square analyses, cell size violations may be of concern. For example, the chi-square between gender and marital status.
9. The authors should review the text in lines 205-214 to ensure it matches the values outlined in Table 2. For example, number of vaccines and infection burden is -.27 (line 210) and -.28 (Table 2).
10. No gender differences were reported for PCL-5 scores. Thus, it is unclear why the authors treated gender as a covariate in the hierarchical regression analysis. Similarly, Table 2 reveals that education and scores on the PCL-5 did not correlate significantly. Thus, it is unclear why the authors would elect to treat education as a covariate.
11. The authors need to provide a rationale for the order in which they enter predictor variables. Why, for example, is number of vaccinations entered in Step 2? Why are initial infection burden and ongoing infection burden entered in separate steps?
12. The authors should provide a squared semi-partial for each predictor variable (i.e., this detail allows the reader to determine the proportion of unique variance accounted for).
13. One possible direction for future research could involve examining whether certain symptoms are more strongly associated with PTSS in men versus women. For instance, due to its less normative nature, is the occurrence of hair loss more stressful for women than for men?
14. In the Discussion section, the authors should avoid using causal language. For instance, it is possible that persons evidencing greater levels of PTSS are, in turn, more likely to engage in catastrophic thinking.
15. An important limitation is the authors’ use of a COVID-19 specific indicator of “fear of dying.” Table 2 suggests that those with more severe initial and ongoing symptoms were (justifiably) more likely to fear dying from COVID-19. It would be interesting to determine whether a generalized fear of death/death anxiety would emerge as a statistically significant predictor of PTSS.
16. Line 44: a word seems to be missing after “following”
17. Line 66: add “has” before “yet”
18. Line 89, suggested edit: “…as researchers begin to address…”
19. Line 100, suggested edit: “…in predicting the severity of PTSS.”
20. Line 206: the correlation between age and the severity of PTSS is -.21 (not -.26).

See above for detailed comments. Points 16-20 reflect minor typographic issues. 

Author Response

Reviewer 3

 

1. The reviewer queried the psychometric soundness of the single-item measure of fear of dying.

 

In the revised manuscript, issues related to the psychometric soundness of single-item measures are addressed in the Methods section (lines 152-159).

 

2. The reviewer requested justification for the use of a COVID-specific measure of fear of dying, as opposed to a more general measure of fear of death.

 

In the revised manuscript, the justification for the use of a COVID-specific measure of fear of dying is provided in the Method section (lines 152-159). In the Introduction, the review of studies that have demonstrated an association between fear of dying and PTSD used situation- or illness-specific measures of fear of dying.

 

3. The reviewer recommended using the term ‘gender’ throughout the paper as opposed to alternating between ‘gender’ and ‘sex’.

 

In the revised manuscript, the term ‘gender’ is used throughout.

 

4. The reviewer suggested computing McDonald’s Omega for the Symptom Catastrophizing Scale.

 

In the revised manuscript, McDonald’s Omega is provided for the Symptom Catastrophizing Scale.

 

5. The reviewer recommended computing point-biserial correlations for correlations with the fear of dying measure.

 

The values reported in Table 2 involving the fear of dying variable are point-biserial correlations. In the revised manuscript, this is indicated in the footnote for the Table.

 

6. The reviewer suggested providing effect sizes for t-test and chi-square tests.

 

In the revised manuscript, effect sizes are provided.

 

7. The reviewer suggested providing the degrees of freedom for the chi-square test.

 

In the revised manuscript, the degrees of freedom for chi-square tests are provided.

 

8. The reviewer noted that, for some χ ² analyses, low cell sizes might be a concern.

 

It is our understanding that χ² is robust to minor violations of cell size lower thresholds (i.e., no more than 20% of cell sizes below 5).  This criterion was not met for the χ² examining the relation between gender and ethic background. The ‘other’ category for ethnic background was deleted and the analysis re-run. In the analysis reported in the revised manuscript, only 1 cell has less than 5 observations.

 

9. The reviewer noted that there were discrepancies between values reported in Table 2 and those reported in the text.

 

In the revised manuscript, the errors have been corrected.

 

10. The reviewer questioned why gender, and education would be included as covariates in the regression analysis since they were not significantly correlated with the PCL-5.

 

In the revised manuscript, the regression analysis has been re-run without gender and education as covariates. The pattern of findings remains unchanged.

 

11. The reviewer suggested providing a rationale for the order of entry of variables in the regression analysis.

 

In the revised manuscript, the rationale for the order of entry of variables is provided in the Data Analytic Approach section of the Methods (lines 196-207). Issues related to temporal order and conceptual primacy were considered in decisions about the order of entry of variables. Age and number of vaccinations were entered first as they would have occurred prior to infection. Infection burden and ongoing symptom burden were entered next as indices of the ‘objective severity’ of the infection. Both variables were computed as symptom counts as opposed to severity ratings to minimize confounding from psychological processes. Initial symptom severity and ongoing symptom burden were entered in separate steps given that the persistence of symptoms of likely to be psychologically experienced as qualitatively different from the experience of the symptoms of acute infection. Fear of dying was entered next as an appraisal that would have been made only after symptoms emerged. Catastrophic thinking was entered last to determine whether an alarmist appraisal of symptom severity contributed to the prediction of PTSS beyond the variables already entered in the analysis.

 

12. The reviewer suggested providing semi-partial correlations in the regression table to permit examination of the unique contribution of each variable to the prediction of the severity of PTSS.

 

In the revised manuscript, semi-partial correlations are provided in the regression table.

 

13. The reviewer suggested that future examination of the data could explore gender differences in the relation between COVID specific symptoms and the severity of PTSS.

 

We thank the reviewer for this suggestion, and we plan to conduct the recommended analyses.

 

14. The reviewer suggested avoiding causal language in the discussion.

 

In the revised manuscript, care is taken to highlight the interpretive constraints posed by the cross-sectional design of the study.

 

15. The reviewer suggested that future research could address how generalized fear of death might emerge as a significant predictor of PTSS.

 

We thank the reviewer for this recommendation and will include a measure of generalized fear of death in future studies.

 

16. The reviewer noted that a word might be missing in line 44 after the word ‘following’.

 

In the revised manuscript, the words ‘COVID-19 infection’ have been added.

 

17. The reviewer suggested adding the word ‘had’ before ‘yet’ in line 66.

 

The text of the manuscript has been modified as the reviewer suggested.

 

18. The reviewer suggested an edit to line 89 of the manuscript.

 

The text of the manuscript has been modified as the reviewer suggested.

 

19. The reviewer suggested an edit to line 100 of the manuscript.

 

The text of the manuscript has been modified as the reviewer suggested.

 

20. The reviewer noted an error in reporting the correlation between age and PTSS.

 

In the revised manuscript, the error has been corrected.

Round 2

Reviewer 2 Report

The reviewers’ comments have been addressed by the authors. However, some minor issues remain concerning the revision of the reference list. 

It is recommended to include the following sources: https://doi.org/10.1016/j.paid.2025.113251 and https://doi.org/10.31083/j.jin2305105, and to reformat the entire bibliography according to APA style guidelines (7th edition).

Author Response

1. The reviewer suggested adding references https://doi.org/10.1016/j.paid.2025.113251 and https://doi.org/10.31083/j.jin2305105.

I reviewed the suggested references but could not see how they could be referenced without making significant alterations to the text.

2. The reviewer suggested formatting the manuscript according to APA 7th edition.

The revised manuscript is referenced in APA 7th edition style.

 

Reviewer 3 Report

The authors have been responsive to the comments I provided in my first review.  A few minor issues remain (see detailed comments); however, I believe these points can be addressed expeditiously. 

1. Line 85, missing bracket (…evaluative judgments).
2. Line 156: add “the” before “conceptualization”
3. Line 202: … given that the persistence of symptoms of likely to be…
4. Line 214: use “N = 381”
5. Line 233: degrees of freedom are missing from the chi-square test.
6. Table 1: N, n, and SD should be italicized.
7. Table 3: Step 3 is missing
8. The percentages listed on lines 275 to 278 do not total 100.
9. Line 275: Nine (2.4%) of participants reported… “of” should be removed.

Author Response

1. The reviewer indicated that on Line 85, there was a missing bracket (...evaluative judgments).

In the revised manuscript, the text has been corrected.

2. The reviewer suggested that on Line 156: the word "the" should be added before "conceptualization"

In the revised manuscript, the text has been modified as suggested by the reviewer.

3.  The reviewer highlighted awkward wording on Line 202.

The text on Line 202 has been modified.

4. The reviewer suggested that on Line 214: "N = 381" should be used.

In the revised manuscript, the text has been modified as suggested by the reviewer.

5. The reviewer noted that on Line 233: degrees of freedom were missing from the chi-square test.

In the revised manuscript, the degrees of freedom have been added.

6.  The reviewer suggested that in Table 1: N, n, and SD should be italicized.

The revised manuscript has been modified as suggested by the reviewer.

7. The reviewer noted that in Table 3: Step 3 was missing.

In the revised manuscript, Table 3 has been corrected.

8.  The reviewer highlighted that the percentages listed on lines 275 to 278 did not total 100.

In the revised manuscript, the values in lines 275 to 278 have been corrected.

9. The reviewer suggested that on Line 275 the word "of" should be removed.

The text has been modified as suggested by the reviewer.