Longitudinal Changes in Maternal Depressive and Anxiety Symptoms Following COVID-19 During Pregnancy: A Cohort Study from Slovakia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study examines longitudinal trends in depressive and anxiety symptoms among pregnant women diagnosed with SARS-CoV-2 infection. The subject is both timely and clinically significant, particularly within the broader context of perinatal mental health during the COVID-19 era. Key strengths include the longitudinal study design and the use of well-validated assessment tools (EPDS, EPDS-3A, and VAS). Nevertheless, several methodological, analytical, and reporting concerns should be addressed to enhance the manuscript’s overall rigor and clarity.

The Comments

1- Regarding Study Design and Potential Bias

• The retrospective evaluation of psychological symptoms during the infection phase introduces the possibility of recall bias, as participants reported their prior mental states after recovery.
• The reliance on complete-case analysis (170 out of 1,184 participants) raises concerns about selection bias, which may affect the generalizability of the findings.
• Including a comparison between participants retained in the analysis and those lost to follow-up would help assess potential attrition bias.

2- Regarding Sample Size and Representativeness

• The final sample represents a considerable reduction from the original cohort, with only 170 participants included in the analysis.
• The small number of hospitalized cases (n = 9) limits the capacity to draw conclusions regarding severe COVID-19 outcomes.
• A formal sample size justification or power calculation should be provided.

3. Regarding Measurement Issues

• The use of retrospective assessment for infection-period symptoms, contrasted with prospective postpartum measurements, introduces inconsistency in data collection.
• Dependence on self-reported obstetric complications without verification from medical records may compromise data reliability.
• It would be helpful to clarify whether validated local-language versions of the EPDS and EPDS-3A were utilized.
•  
• 4- Regarding Statistical Analysis

• The choice of repeated-measures ANCOVA requires clearer justification, particularly in terms of handling missing data and covariates.
• Although some statistical assumptions (e.g., homogeneity of variance) were only partially met, their potential impact on the results is not adequately discussed.
• While effect sizes are reported, their clinical interpretation (e.g., small, moderate, or large) is missing.
• Alternative approaches, such as mixed-effects models, could be considered to better accommodate longitudinal data and missing values.

5- Interpretation of Findings

• The assertion that psychological distress represents a “time-limited stress response” may be somewhat overstated, given the observational design and absence of a control group.
• The lack of association between infection severity and psychological outcomes should be interpreted cautiously due to limited statistical power.
• The discussion would benefit from stronger integration with existing longitudinal research, rather than focusing primarily on cross-sectional studies.

6- Confounding Factors

• Several relevant confounders were not fully accounted for, including:
  • Socioeconomic status
  • Pre-existing mental health conditions
  • Levels of social support
• These factors may substantially influence mental health outcomes and should be acknowledged as limitations.

Other Comments

1. Clarity and Writing

• Certain sections, particularly in the discussion, contain repetitive statements (e.g., regarding the lack of association with infection severity).
• Minor grammatical errors and formatting inconsistencies should be corrected.

2. Introduction

• Although informative, the introduction could be more concise and better focused on the specific research gap.
• The novelty of the study relative to prior longitudinal research should be highlighted earlier.

3. Results Presentation

• Tables and figures should be more effectively integrated into the narrative, with clearer interpretation provided.
• Confidence intervals should be reported alongside p-values where appropriate.

4. Discussion

• The clinical implications, particularly regarding screening and timing of interventions, should be further elaborated.
• A more balanced discussion of strengths and limitations is recommended.

5. Terminology and Consistency

• Consistent terminology should be maintained throughout the manuscript, particularly for terms such as:
  • “SARS-CoV-2 infection” vs. “COVID-19 infection”
  • “postpartum” vs. “postnatal”

Author Response

Thank you very much for taking the time to review our manuscript. We have addressed each comment in the following section and updated the manuscript accordingly (with yellow highlights and English improvements are grey).

 

Comment 1

1- Regarding Study Design and Potential Bias

1.1 The retrospective evaluation of psychological symptoms during the infection phase introduces the possibility of recall bias, as participants reported their prior mental states after recovery.

RESPONSE 1
We thank the reviewer for this important comment. The possibility of recall bias related to the retrospective assessment of psychological symptoms during the infection period had already been acknowledged in the original manuscript. To further clarify this limitation, we expanded the relevant statement in the revised version of the manuscript.

Changes in manuscript:

"Several limitations should be considered when interpreting these findings. First, the retrospective assessment of psychological symptoms during the infection and postpartum periods may be subject to recall bias. The extended recall interval between the initial emotional experiences across assessment periods and the follow up data collection may have influenced the accuracy of participants reports. Furthermore, psychological responses may also have been influenced by changing pandemic conditions across the study period."

 

Comment 2

1.2 The reliance on complete-case analysis (170 out of 1,184 participants) raises concerns about selection bias, which may affect the generalizability of the findings. Including a comparison between participants retained in the analysis and those lost to follow-up would help assess potential attrition bias.

RESPONSE 2

We thank the reviewer for this important methodological observation. We acknowledge that the use of complete-case analysis and the substantial reduction in the final analytical sample may have introduced selection and attrition bias, potentially limiting the generalizability of the findings. We agree that a comparison between participants included in the final analysis and those lost to follow-up could provide additional insight into potential attrition bias. However, complete demographic and clinical data were not available for all non-completing participants, which limited our ability to perform a reliable comparative analysis. These limitations have now been further acknowledged in the revised manuscript.

Changes in manuscript:

The following statement was added to the Limitations section:
“The use of complete-case analysis and the relatively high attrition rate may have introduced selection bias, potentially limiting the generalizability of the findings to the broader population of pregnant women affected by SARS-CoV-2 infection. In addition, incomplete demographic and clinical data from participants lost to follow-up limited the possibility of conducting a detailed attrition bias analysis.”

 

Comment 3

2- Regarding Sample Size and Representativeness

The final sample represents a considerable reduction from the original cohort, with only 170 participants included in the analysis.

The small number of hospitalized cases (n = 9) limits the capacity to draw conclusions regarding severe COVID-19 outcomes.

RESPONSE 3

We thank the reviewer for this observation. We acknowledge that the final analytical sample represented a substantial reduction from the original cohort. This limitation had already been acknowledged and was further clarified.

 

Comment 5

A formal sample size justification or power calculation should be provided.

RESPONSE 5

We thank the reviewer for this suggestion. A formal a priori sample size calculation was not performed, as the study was designed as an exploratory retrospective longitudinal cohort study based on all available eligible participants with complete repeated measurements. This clarification has now been added to the Methods section of the revised manuscript.

Changes in manuscript:
The following statement was added to the Methods section:
“ A formal a priori sample size calculation was not performed, as the study was exploratory in nature and based on all available eligible participants with complete repeated measurements across all three assessment periods.”

 

Comment 6

3- Regarding Measurement Issues

We thank the reviewer for this important observations.

The use of retrospective assessment for infection-period symptoms, contrasted with prospective postpartum measurements, introduces inconsistency in data collection.

RESPONSE 6

We would like to clarify that the postpartum assessments were not conducted strictly prospectively in real time. Rather, all three assessment periods were evaluated retrospectively during follow-up data collection. Nevertheless, we acknowledge that differences in temporal distance from the assessed periods may have introduced variability in recall accuracy between assessment timepoints. This clarification has now been added to the revised manuscript.

Changes in manuscript:

"Several limitations should be considered when interpreting these findings. First, the retrospective assessment of psychological symptoms during the infection and postpartum periods may be subject to recall bias. The extended recall interval between the initial emotional experiences across assessment periods and the follow up data collection may have influenced the accuracy of participants reports. Furthermore, psychological responses may also have been influenced by changing pandemic conditions across the study period."

 

Comment 7

Dependence on self-reported obstetric complications without verification from medical records may compromise data reliability.

RESPONSE 7

This limitation had already been acknowledged in the original manuscript and is discussed in the Limitations section.

 

Comment 8

It would be helpful to clarify whether validated local-language versions of the EPDS and EPDS-3A were utilized.

RESPONSE 8

The Methods section has been revised to clarify that a validated Slovak-language version of the Edinburgh Postnatal Depression Scale (EPDS) was utilized in the study.

Changes in manuscript:
The following sentence was added to the Methods section:
“A validated Slovak-language version of the Edinburgh Postnatal Depression Scale (EPDS), demonstrating good psychometric properties in postpartum Slovak women, was used.”

 

Comment 9

4- Regarding Statistical Analysis

The choice of repeated-measures ANCOVA requires clearer justification, particularly in terms of handling missing data and covariates.

RESPONSE 9

We thank the reviewer for this valuable methodological suggestion. Repeated-measures ANCOVA was selected to evaluate within-subject changes across the three assessment periods while adjusting for clinically relevant covariates in participants with complete repeated measurements. The rationale for this approach has now been clarified in the Methods section.

Changes in manuscript:

“To evaluate changes over time while adjusting for covariates, a repeated-measures ANCOVA was conducted. This approach was selected to evaluate within-subject changes across the three assessment periods (infection, six weeks postpartum, one year postpartum) while adjusting for clinically relevant covariates in participants with complete repeated measurements".

 

Comment 10

Although some statistical assumptions (e.g., homogeneity of variance) were only partially met, their potential impact on the results is not adequately discussed.

RESPONSE 10

We thank the reviewer for this important observation. The potential impact of partially unmet assumptions for between-subject comparisons has now been further addressed in the revised manuscript.

Changes in manuscript:

“The assumption of homogeneity of variances was only partially satisfied for the between-subject factor (COVID-19 severity), which may have affected the robustness of between-subject comparisons and should be considered when interpreting severity-related findings.“

 

Comment 11

While effect sizes are reported, their clinical interpretation (e.g., small, moderate, or large) is missing.

In addition, the interpretation of effect sizes and the potential impact of partially unmet assumptions for between-subject comparisons have been further clarified in the revised manuscript.

RESPONSE 11

Changes in manuscript:

“Effect sizes were expressed as partial eta squared (ηp²), with values of 0.01, 0.06, and 0.14 interpreted as small, medium, and large, respectively." And further included in the Results section.

 

Comment 12

Alternative approaches, such as mixed-effects models, could be considered to better accommodate longitudinal data and missing values.

RESPONSE 12

We acknowledge that mixed-effects models may offer advantages in handling incomplete longitudinal data and accommodating missing observations. However, the present analyses were based on a complete-case approach including only participants with complete repeated measurements across all assessment periods. The potential advantages of mixed-effects approaches have now been acknowledged in the revised manuscript as a consideration for future research.

Changes in manuscript:

“Future studies may benefit from the use of mixed-effects models, which may provide greater flexibility in handling incomplete longitudinal data.”

 

Comment 13

5- Interpretation of Findings

The assertion that psychological distress represents a “time-limited stress response” may be somewhat overstated, given the observational design and absence of a control group.

The lack of association between infection severity and psychological outcomes should be interpreted cautiously due to limited statistical power.

The discussion would benefit from stronger integration with existing longitudinal research, rather than focusing primarily on cross-sectional studies.

RESPONSE 13

We thank the reviewer for this valuable observation. In response to these comments, the Discussion section has been revised to adopt a more cautious interpretation of the observed psychological trajectories, particularly regarding the characterization of distress as a potentially time-limited response. Statements referring to the temporal nature of psychological distress and the lack of association with infection severity were softened to better reflect the observational design of the study and the limited statistical power related to severe COVID-19 cases. In addition, the Discussion section has been revised to strengthen the integration of available longitudinal studies examining maternal mental health during and after the COVID-19 pandemic. At the same time, we acknowledge that longitudinal evidence in this specific area remains relatively limited compared to the predominantly cross-sectional literature.

 

Comment 14

6- Confounding Factors

Several relevant confounders were not fully accounted for, including: Socioeconomic status, Pre-existing mental health conditions, Levels of social support. These factors may substantially influence mental health outcomes and should be acknowledged as limitations.

RESPONSE 14

We thank the reviewer for this important observation. We acknowledge that several potentially relevant confounding factors, including socioeconomic status, pre-existing mental health conditions, and levels of social support, were not systematically assessed in the present study and may have influenced psychological outcomes. This limitation has now been acknowledged in the revised manuscript.

 

Other comments:

Comment 15

2. Introduction

• Although informative, the introduction could be more concise and better focused on the specific research gap.
• The novelty of the study relative to prior longitudinal research should be highlighted earlier.

RESPONSE 15

We thank the reviewer for this helpful suggestion. The Introduction section has been revised to place greater emphasis on the specific research gap and the longitudinal contribution of the present study earlier in the text. In addition, several sections of the Introduction were streamlined to improve conciseness and focus.

 

Comment 16

3. Results Presentation

Tables and figures should be more effectively integrated into the narrative, with clearer interpretation provided.

Confidence intervals should be reported alongside p-values where appropriate.

RESPONSE 16

We thank the reviewer for this valuable suggestion. The Results section has been revised to improve the integration and interpretation of tables and figures within the narrative text. Confidence intervals were reported where appropriate and available within the applied statistical framework.

 

Comment 17

4. Discussion

The clinical implications, particularly regarding screening and timing of interventions, should be further elaborated.

A more balanced discussion of strengths and limitations is recommended.

RESPONSE 17

We thank the reviewer for these valuable suggestions. The Discussion section has been revised to further elaborate on the clinical implications of the findings, particularly regarding the importance of timely mental health screening and psychological support during acute SARS-CoV-2 infection in pregnancy. In addition, the Discussion and Limitations sections were expanded to provide a more balanced consideration of the strengths and limitations of the study.

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

 

Congratulations on the great work done. However, consider revising the manuscript in line with the following comments.

 

Line 47: Pregnant women represent a particularly vulnerable population,

© Kindly rephrase

 

Line 54: The prevalence of unipolar depression in individual trimesters of pregnancy is re-54 ported as..

© Kindly rephrase

 

Line 60: Depression in the perinatal period, including the postnatal form, is diag-56 nosed using, for example, the Edinburgh Postnatal Depression Scale (EPDS)

© Not clear on what you are trying to say?

 

Line 69: Moreover, a meta-analysis by Ghazanfarpour et al. [13] provided further evidence 69 and points to differences in the prevalence of depression and anxiety between continents. Their continent-based analysis showed that the prevalence of anxiety was higher in Western countries (38%) than in Asian countries (7.8%). The authors report the prevalence of anxiety in individual countries as follows: Italy (38%), Canada (56%), Pakistan (14%), Greece (53%), Sri Lanka (17.5%), China (0.3–29%), and Iran (3.8%). The prevalence of depression was recorded in Canada (37%), Belgium (25%), Turkey (35.4%), Sri Lanka 75 (19.5%), and China (11–29%).

© Short it up, not the country-based data, but only continental-wise data supports the problem statement, I believe.

 

Line 100: upon providing informed consent

© Oral or written consent?

 

Line 113: Data collection took place continuously throughout the pandemic, and data collection 113 ended on May 5, 2023

© Mentioned the exact three points of data collection. 

 

Line 118: visual analogue scales (VAS)

© through telephonic contact, how is VAS feasible? Need further clarification.

 

Line 91-148: © Kindly use standard reporting guidelines

 

Line 150: Between March 2020 and May 2023, the…

© The levels of fear and anxiety are varied throughout this timeline, right? How did you address?

 

© Is the Bassline infection timeline the same for all the participants? Kindly clarify

 

© All the 3 measures among the participants done in the same timeline? Kindly clarify

 

© Because the data were collected between March 2020 and May 2023, but it was not clearly mentioned when all 3 data sets were collected.

 

© Causal relationship must be interpreted safely.

 

 

 

 

 

 

 

 

 

 

 

Author Response

We sincerely appreciate the time and effort reviewer dedicated to our manuscript. Our detailed responses are provided below (and changes with green highlights, English improvements with grey highlights in manuscript):

Comments 1:

Line 47: Pregnant women represent a particularly vulnerable population,

© Kindly rephrase

Line 54: The prevalence of unipolar depression in individual trimesters of pregnancy is re-54 ported as..

© Kindly rephrase

Line 60: Depression in the perinatal period, including the postnatal form, is diag-56 nosed using, for example, the Edinburgh Postnatal Depression Scale (EPDS)

© Not clear on what you are trying to say?

Line 69: Moreover, a meta-analysis by Ghazanfarpour et al. [13] provided further evidence 69 and points to differences in the prevalence of depression and anxiety between continents. Their continent-based analysis showed that the prevalence of anxiety was higher in Western countries (38%) than in Asian countries (7.8%). The authors report the prevalence of anxiety in individual countries as follows: Italy (38%), Canada (56%), Pakistan (14%), Greece (53%), Sri Lanka (17.5%), China (0.3–29%), and Iran (3.8%). The prevalence of depression was recorded in Canada (37%), Belgium (25%), Turkey (35.4%), Sri Lanka 75 (19.5%), and China (11–29%).

© Short it up, not the country-based data, but only continental-wise data supports the problem statement, I believe.

Line 100: upon providing informed consent

© Oral or written consent?

RESPONSE 1

We thank the reviewer for these helpful comments and suggestions. The Introduction section was revised to improve clarity and conciseness, including rephrasing several sentences and simplifying the presentation of epidemiological data. The description of depression assessment in the perinatal period was clarified, and country-specific prevalence data were removed to improve readability and focus. In addition, the Methods section was revised to clarify that written informed consent was obtained from all participants prior to questionnaire distribution.

 

Comment 2:

Line 113: Data collection took place continuously throughout the pandemic, and data collection 113 ended on May 5, 2023

© Mentioned the exact three points of data collection. 

RESPONSE 2

We thank the reviewer for this important observation. The Methods section has been revised to further clarify the timing of data collection and the assessment periods. Specifically, we clarified that participants entered the study at different time points during the pandemic period between March 2020 and May 2023, and that psychological symptoms were retrospectively assessed for three predefined periods: during acute SARS-CoV-2 infection, six weeks postpartum, and one year postpartum.

Change in manuscript

„Data collection was initiated after ethical approval in 2021 and continued until May 2023. The study retrospectively covered the pandemic period from March 2020 to May 2023. Participants entered the study at different time points during the pandemic, while psychological outcomes were retrospectively assessed across the three predefined study periods.“

 

Comment 3

Line 118: visual analogue scales (VAS)

© through telephonic contact, how is VAS feasible? Need further clarification.

RESPONSE 3

We thank the reviewer for this observation. Participants were contacted by telephone only for recruitment purposes. The questionnaires, including the visual analogue scales (VAS), were self-administered and distributed electronically via e-mail or by post following written informed consent. This clarification has been added to the revised manuscript.

 

Comment 4

Line 91-148: © Kindly use standard reporting guidelines

RESPONSE 4

We thank the reviewer for this valuable suggestion. The manuscript was revised with consideration of the STROBE reporting recommendations for observational studies, and a corresponding statement has been added to the Methods section.

 

Comment 5

Line 150: Between March 2020 and May 2023, the…

© The levels of fear and anxiety are varied throughout this timeline, right? How did you address?

RESPONSE 5

We thank the reviewer for this important observation. A limitation acknowledging that psychological responses may have been influenced by changing pandemic conditions across the study period has been added to the revised manuscript.

Change in manuscript

„Furthermore, psychological responses may also have been influenced by changing pandemic conditions across the study period.“

 

Comments 6

© Is the Bassline infection timeline the same for all the participants? Kindly clarify

© All the 3 measures among the participants done in the same timeline? Kindly clarify

RESPONSE 6

We thank the reviewer for this important clarification request. The Methods section was revised to explicitly state that participants entered the study at different time points during the pandemic period, and therefore baseline SARS-CoV-2 infection timing was not identical across all participants.

 

Comment 7

© Because the data were collected between March 2020 and May 2023, but it was not clearly mentioned when all 3 data sets were collected.

RESPONSE 7

We thank the reviewer for this important clarification request. Participants retrospectively evaluated their psychological symptoms for three predefined assessment periods: during SARS-CoV-2 infection, six weeks postpartum, and one year postpartum. These assessments were completed during follow-up questionnaire administration.

 

Comment 8

 © Causal relationship must be interpreted safely.

RESPONSE 8

We thank the reviewer for this important observation. The interpretation of the findings was revised throughout the Discussion section to ensure a more cautious interpretation of potential causal relationships, in line with the observational nature of the study.

Reviewer 3 Report

Comments and Suggestions for Authors

Almost the entire manuscript is well written, with appropriate descriptions, analyses, and strengths/limitations. Thank you for this contribution to women's health.

Figure 1A, however, should have the word "mean" added to the Y-axis.

Also, re the 27% incidence of meeting the criterion for anxiety and 27% incidence of meeting criterion for depression: was there complete overlap between these two forms of psychological distress, no overlap, or how much overlap?  Others have found about 90% overlap between depression and anxiety in non-pregnant samples. It would be valuable to give the overall % with anxiety or depression to show the impact of COVID on mental health.

Author Response

We are grateful to the reviewer for constructive feedback on our work. We have addressed each comment in the following section and updated the manuscript accordingly (Changes are highlighted in blue in the manuscript; Improvements of English are highlighted in grey).

 

Comment 1

Figure 1A, however, should have the word "mean" added to the Y-axis.

RESPONSE 1

We thank the reviewer for this helpful suggestion. The Y-axis label in Figure 1A has been revised to clarify that the presented values represent mean EPDS scores.

 

Comment 2

Also, re the 27% incidence of meeting the criterion for anxiety and 27% incidence of meeting criterion for depression: was there complete overlap between these two forms of psychological distress, no overlap, or how much overlap?  Others have found about 90% overlap between depression and anxiety in non-pregnant samples. It would be valuable to give the overall % with anxiety or depression to show the impact of COVID on mental health.

RESPONSE 2

We thank the reviewer for this valuable observation. Additional analyses regarding the overlap between clinically relevant depressive symptoms (EPDS ≥ 11) and anxiety symptoms (EPDS-3A ≥ 5) were added to the Results section. The revised manuscript now reports the proportion of participants meeting criteria for both depressive and anxiety symptoms across all assessment periods, thereby providing additional insight into the overall burden of psychological distress.

Change in manuscript:

„Considerable overlap between clinically relevant depressive symptoms (EPDS ≥ 11) and anxiety symptoms (EPDS-3A ≥ 5) was observed across all assessment periods. The proportion of participants meeting criteria for both depressive and anxiety symptoms was 21.18% during SARS-CoV-2 infection, 16.47% at six weeks postpartum, and 4.12% at one year postpartum.”