Impact of Smoking on Household SARS-CoV-2 Transmission

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Please see the attached file.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 1 
We thank the reviewer for their comments, which have contributed to improving 
the clarity and methodological rigor of the manuscript. Below, we respond to each 
comment. All changes have been incorporated into the revised version. 
Comment 1 
The manuscript provides novel evidence by demonstrating that smoking among 
household contacts, rather than among index cases, is associated with an 
increased risk of SARS-CoV-2 infection. This distinction is important and 
relatively underexplored in the literature. The prospective design and the use of 
active follow-up with diagnostic testing strengthen the validity of the findings. 
However, the manuscript would benefit from a clearer articulation of its novelty in 
the Introduction and Discussion. Although the authors mention that data on intra
household transmission are scarce, they should more explicitly clarify how their 
study advances previous household transmission studies, particularly those cited 
in references [11–15]. It is recommended to explicitly highlight what is new 
compared with previous research on intra-household transmission, especially 
regarding smoking as a susceptibility factor rather than a transmission factor. 
Response: We agree that the distinction between the role of the index case and 
that of the household contact is fundamental. We have revised the Introduction 
to clarify that, although most studies focus on clinical severity, our main 
contribution is the analysis of how smoking influences transmission dynamics 
within households. We have highlighted that our study allows smoking to be 
examined both as a factor that may facilitate viral transmission by the index case 
(transmissibility) and, above all, as a factor that increases the susceptibility of 
household contacts to infection, even in a context of high vaccination coverage. 
We believe that this perspective differentiates our study from previously published 
research. 
Changes made: A new paragraph has been added to the Introduction, 
immediately before the statement of the study aim. 
Comment 2 
The prospective cohort design is appropriate and well justified. Inclusion and 
exclusion criteria are clearly defined, and follow-up procedures are adequately 
described. However, the use of convenience sampling for primary care centres 
may introduce selection bias and limit the generalisability of the results. Although 
this aspect is briefly acknowledged, its potential impact on external validity 
deserves 
more 
explicit 
discussion. 
In addition, grouping current smokers and former smokers may obscure dose
related, response-related, or time-since-cessation effects, which are biologically 
plausible and epidemiologically relevant. It is recommended to expand the 
Limitations section to better address the implications of convenience sampling 
and to provide a stronger justification for combining smokers and ex-smokers, 
explicitly discussing how this methodological decision may bias results toward 
the null or overestimate the observed effects. 
Response: We have expanded the Strengths and Limitations section to discuss 
the potential impact of convenience sampling on the external validity of the study. 
We have also included a more detailed justification for grouping current smokers 
and former smokers, based on the need to ensure sufficient statistical power, 
acknowledging that this decision may have diluted differences related to smoking 
intensity or time since cessation. 
Changes made: The Strengths and Limitations section has been expanded to 
include a detailed discussion of convenience sampling and the grouping of 
smokers and former smokers, including the potential direction of bias. 
Comment 3 
Smoking status is self-reported and broadly categorized. While this is common in 
epidemiological studies, it introduces potential misclassification bias. 
The manuscript lacks information on: 
● Cigarettes per day 
● Duration of smoking 
● Passive smoke exposure within households 
These factors may significantly influence infection susceptibility and 
transmission. 
It is recommended to more explicitly acknowledge that smoking intensity and 
duration were not measured and to clarify whether passive smoking among non
smokers was assessed or considered. 
Response: We agree with the reviewer that having detailed information on 
smoking consumption, intensity, and exposure to environmental tobacco smoke 
would have enriched the analysis. 
Changes made: Additional text has been incorporated into the Strengths and 
Limitations section addressing the lack of detailed data on tobacco exposure 
and its implications. 
Comment 4 
The statistical methods are appropriate and clearly described. The use of logistic 
regression models adjusted for key confounders (age, vaccination status, prior 
infection) is a strength. However, the rationale for variable selection in the 
backward regression model should be described in more detail. In addition, 
potential household-level clustering effects are not discussed. Since multiple 
contacts may belong to the same household, observations may not be fully 
independent. It is recommended to clarify the criteria used for backward selection 
(e.g., p-value thresholds) and to discuss whether household clustering was 
considered or, if not, to justify this choice or acknowledge it as a limitation. 
Response: We have clarified in the Statistical Analysis section that the 
significance criterion used in the backward regression model was p < 0.10. 
Regarding household-level clustering, we have addressed this issue in the 
Limitations section, acknowledging that some degree of correlation between 
observations cannot be completely ruled out and that this may have affected the 
precision of the estimates. 
Changes made: The Statistical Analysis section now specifies the criterion 
used in the backward selection procedure (p < 0.10). 
The potential impact of non-independence between observations has been 
conceptually acknowledged in the Strengths and Limitations section. 
Comment 5 
The interpretation of the findings is generally balanced and cautious. The authors 
appropriately avoid causal claims and contextualize their results within the 
existing literature. However, some interpretations could be refined: 
● The lack of association between index case smoking and transmission should 
be interpreted cautiously, given the limited power to detect effects of moderate 
magnitude. 
● The protective effect of index case vaccination is consistent with previous 
studies, but the discussion could better integrate this finding with mechanisms 
related to viral load and preventive behaviors. 
It is recommended to emphasize that null findings regarding index case smoking 
may reflect limited statistical power and to expand the discussion on biological 
and behavioral mechanisms linking smoking to increased susceptibility rather 
than transmissibility. 
Response: We have revised the Discussion accordingly. We now explain that 
the absence of a statistically significant association between index case smoking 
and intra-household transmission should be interpreted with caution, as the study 
may have had limited statistical power to detect effects of smaller magnitude. In 
addition, we have expanded the discussion of biological and behavioral 
mechanisms to emphasize that smoking may primarily increase host 
susceptibility rather than viral transmissibility. 
Changes made: A sentence has been added to the Discussion to qualify the 
interpretation of the non-significant result related to index case smoking, and the 
text has been expanded to strengthen the explanation of the potential biological 
mechanisms. 
Comment 6 
The public health relevance of the findings is well articulated, particularly 
regarding smoking cessation, vaccination, and the protection of older adults. 
However, the manuscript could benefit from more concrete policy implications, 
such as: 
● Targeted smoking cessation interventions for households with vulnerable 
individuals. 
● Integration of smoking status into risk assessments within contact tracing 
processes. 
Response: We have expanded the final part of the Discussion to provide more 
specific public health implications, considering the potential inclusion of smoking 
status as a risk variable in household contact assessments, as well as the 
importance of targeted smoking cessation interventions and reinforced 
preventive measures in households with vulnerable individuals. 
Changes made: Additional text has been added to the final part of the 
Discussion to clarify the public health implications. 
We believe that these revisions have substantially strengthened the manuscript 
and have allowed us to address all of the reviewer’s comments.

Reviewer 2 Report

Comments and Suggestions for Authors

 The topic of the manuscript is important for public health because it links a modifiable exposure (smoking) with household transmission dynamics, and the prospective design and systematic follow-up/testing of contacts are notable strengths. The manuscript is generally well structured and the main results are clearly presented; however, some clarifications in the Methods and some tightening of the Discussion/Conclusions would improve interpretability and reduce overlap with the Results.

Materials and Methods

Comment 1: Regarding close contacts and the 7-day follow-up, there is a risk that infections occurring later may not be detected (some may be detected even after day 7–10). Please provide a brief justification for the chosen follow-up period or acknowledge in the Limitations the possibility of underestimating secondary infections. According to the WHO, SARS‑CoV‑2 infectiousness can begin 1–2 days before symptom onset and may extend up to 8–10 days after symptom onset.

Comment 2: Please clarify how ‘previous SARS-CoV-2 infection’ in household contacts was defined (data source: self-report vs registry/medical records) and specify the time window relative to enrolment/index-case episode (e.g., any prior infection vs infections ≥90 days earlier). Without a defined look-back period, the protective association is difficult to interpret.

Discussion

Comment 3: Overall, the Discussion is well referenced, but it partly reads as a second Results section. Consider reducing the repetition of numerical findings (e.g., ORs/percentages and the 30-day complications) and focusing more on interpretation, plausible mechanisms, and implications.

Conclusions

Comment 4: Consider strengthening the Conclusions by adding a brief synthesis of the main take-home message (‘so what?’) and 1–2 concrete directions for future research (e.g., separating current vs former smokers, quantifying smoking intensity, longer follow-up for secondary cases), rather than reiterating the Discussion.

Author Response

Response to Reviewer 2 
We thank the reviewer for their comments, which have contributed to improving 
the clarity and methodological rigor of the manuscript. Below, we respond to each 
observation. All changes have been incorporated into the revised version of the 
manuscript. 
Comment 1 
Regarding close contacts and the 7-day follow-up, there is a risk that infections 
occurring later may not be detected (some may be detected even after days 7
10). Please provide a brief justification for the chosen follow-up period or 
acknowledge in the Limitations section the possibility of underestimating 
secondary infections. According to the WHO, SARS-CoV-2 infectiousness can 
begin 1–2 days before symptom onset and may extend up to 8–10 days after 
symptom onset. 
Response: We have clarified in the Methods section that the 7-day follow-up 
period was established in accordance with the contact tracing protocol in force at 
the time of the study, which prioritized early detection of secondary cases in the 
household setting. In addition, we have acknowledged in the Strengths and 
Limitations section that a follow-up limited to 7 days may have underestimated 
the detection of infections with longer incubation periods, and therefore some 
secondary cases may have gone undetected. 
Changes made: A justification for the 7-day follow-up period has been added to 
the Materials and Methods section. 
A new limitation regarding the potential underestimation of infections due to the 
limited duration of follow-up has been incorporated into the Strengths and 
Limitations section. 
Comment 2 
Please clarify how “previous SARS-CoV-2 infection” in household contacts was 
defined (data source: self-report vs registry/medical records) and specify the time 
window relative to enrolment or the index case episode (e.g., any prior infection 
vs infections occurring ≥90 days earlier). Without a defined look-back period, the 
protective association is difficult to interpret. 
Response: We have clarified in the manuscript that the variable “previous SARS
CoV-2 infection” was defined based on clinical records and epidemiological 
surveillance data available prior to study enrolment. No specific retrospective 
time window in relation to the index case episode was established. Additionally, 
we have acknowledged in the Strengths and Limitations section that the absence 
of a defined temporal window limits the interpretation of the observed protective 
association. 
Changes made: A clarification regarding the definition of the “previous SARS
CoV-2 infection” variable has been added. 
A limitation related to the absence of a defined temporal window for this variable 
has been incorporated into the Strengths and Limitations section. 
Comment 3 
Overall, the Discussion is well referenced, but it partly reads as a second Results 
section. Consider reducing the repetition of numerical findings (e.g., 
ORs/percentages and the 30-day complications) and focusing more on 
interpretation, plausible mechanisms, and implications. 
Response: We have revised the Discussion to reduce repetition of numerical 
results already presented and to strengthen the interpretative focus. 
Changes made: The Discussion section has been edited to minimize 
reiteration of quantitative data and to emphasize interpretation of the findings 
and their implications. 
Comment 4 
Consider strengthening the Conclusions by adding a brief synthesis of the main 
take-home message (“so what?”) and 1–2 concrete directions for future research 
(e.g., separating current vs former smokers, quantifying smoking intensity, or 
incorporating longer follow-up for secondary cases), rather than reiterating the 
Discussion. 
Response: We have strengthened the Conclusions section by incorporating a 
clear synthesis of the main message of the study and by adding concrete 
directions for future research. 
Changes made: A final paragraph has been added to the Conclusions section 
summarizing the main take-home message and outlining potential directions for 
future studies. 
We believe that these revisions have substantially strengthened the manuscript 
and have adequately addressed all of the reviewer’s comments.

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the opportunity to take part in this review.

The article is very well structured and clearly addresses the increased susceptibility to SARS-CoV-2 infection among smoker contacts; the results are clear and well presented. The discussion is well argued, including comparative aspects related to the increased susceptibility to certain infections.

However, several limitations of the study were identified. These include the small sample size and the participants’ medical history. Patients older than 45 years are also those who frequently present significant comorbidities, which are known to substantially increase susceptibility to infections, in addition to the effect of smoking itself.

The conclusions are clear and emphasize the importance of smoking cessation, as well as other preventive measures, at the population level.

Author Response

Response to Reviewer 3 
We thank the reviewer for their comments, which have contributed to improving 
the clarity and methodological rigor of the manuscript. Below, we address each 
observation. All changes have been incorporated into the revised version. 
Comment 
The manuscript is well structured and clearly addresses the increased 
susceptibility to SARS-CoV-2 infection among smoking household contacts; the 
results are clear and well presented. The Discussion is well argued, including 
comparative aspects related to increased susceptibility to certain infections. 
However, some study limitations are identified, including the relatively small 
sample size and the clinical history of participants. Individuals older than 45 years 
are also those who more frequently present significant comorbidities, which are 
known to increase susceptibility to infections, in addition to the effect of smoking 
itself. The Conclusions are clear and highlight the importance of smoking 
cessation, as well as other preventive measures, at the population level. 
Response: We appreciate the reviewer’s assessment. We have incorporated, as 
an additional limitation, the possible influence of unmeasured comorbidities, 
particularly among older participants. 
Changes made: A statement has been added to the Strengths and Limitations 
section acknowledging the potential influence of unmeasured comorbidities. 
We believe that these revisions have substantially strengthened the manuscript 
and have addressed all of the reviewer’s observations.

Reviewer 4 Report

Comments and Suggestions for Authors

Regarding the article (healthcare-4103804) by Pardos-Plaza J et al., titled "Impact of smoking on SARS-CoV-2 transmission in the home," I have some concerns.

1. The introduction mentions previous studies suggesting that nicotine might have a protective effect. However, their results show the opposite: smokers were significantly more likely to become infected (48.6% vs. 33.8%). The authors should address this issue more comprehensively. Recent literature (2024-2025) suggests that, while nicotine may interact with ACE2 receptors, the damage caused by smoking is much greater.

2. A table or diagram should be included to illustrate how smoking affects the different layers of respiratory defense.

3. The authors should explore whether smoker behavior (e.g., smoking outdoors or away from others) acts as a confounding variable that the self-report survey might have missed.

4. The study shows a small sample size for the problem it addresses. Given that people ≥65 years of age show a much higher cumulative incidence of infection (60.6% compared to 23% in younger individuals), the statistics should be powerful enough to analyze the interaction between smoking and immunosenescence.

5. The statistics should be rigorously adjusted for prior infections, as a history of SARS-CoV-2 infection provides a significant protective effect (aOR = 0.49) that may mask smoking-related risk factors.

Comments on the Quality of English Language

None

Author Response

Response to Reviewer 4 
We thank the reviewer for their comments, which have contributed to improving 
the clarity and methodological rigor of the manuscript. Below, we respond to each 
observation. All changes have been incorporated into the revised version. 
Comment 1 
The Introduction mentions previous studies suggesting that nicotine might have 
a protective effect. However, the results show the opposite: smokers presented 
a significantly higher probability of infection (48.6% versus 33.8%). The authors 
should address this issue more thoroughly. Recent literature (2024–2025) 
suggests that although nicotine may interact with ACE2 receptors, the damage 
caused by smoking is much greater. 
Response: We have expanded the Discussion to contextualize our results in 
relation to the initial hypothesis of a possible protective effect of nicotine. It is 
highlighted that the immunological and structural damage associated with 
smoking is consistent with the greater susceptibility observed among smokers. 
Changes made: A brief clarifying paragraph has been added to the Discussion. 
Comment 2 
A table or diagram illustrating how smoking affects the different layers of 
respiratory defense should be included. 
Response: We appreciate the suggestion. The mechanisms by which smoking 
affects respiratory defenses are described and referenced in the Discussion. 
Changes made: No additional changes have been made to the manuscript. 
Comment 3 
The authors should explore whether smokers’ behavior (e.g., smoking outdoors 
or away from other people) acts as a confounding variable that the self-reported 
questionnaire may not have captured. 
Response: Specific information on behaviors associated with tobacco 
consumption, such as smoking outdoors or maintaining distance from other 
household members, was not available. We acknowledge that these behaviors 
could have acted as unmeasured confounding factors. 
Changes made: A specific mention has been added to the Strengths and 
Limitations section regarding the possible influence of smoking-related 
behaviors not assessed in this study. 
Comment 4

The study presents a limited sample size for the problem addressed. Given that 
individuals aged ≥65 years show a much higher cumulative incidence of infection 
(60.6% versus 23% in younger individuals), statistical analyses should be 
sufficiently powered to analyze the interaction between smoking and 
immunosenescence. 
Response: Given the available sample size, it was not possible to robustly 
analyze the interaction between smoking and advanced age. We have 
incorporated this issue as a priority line for future research in the Conclusions 
section. 
Changes made: A mention has been added to the Conclusions section 
regarding the need for future studies to explore the interaction between smoking 
and immunosenescence. 
Comment 5 
Statistical analyses should be rigorously adjusted for previous infections, as a 
history of SARS-CoV-2 infection provides a significant protective effect (aOR = 
0.49) that could mask risk factors associated with smoking. 
Response: We appreciate this observation. Previous SARS-CoV-2 infection was 
included as an independent variable in the multivariate models, so that the 
associations between smoking and infection risk were estimated adjusting for this 
protective effect. 
Changes made: No additional changes have been made to the manuscript. 
We consider that these revisions have substantially strengthened the manuscript 
and have allowed us to address all of the reviewer’s observations.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed all my comments and significantly improved the paper. 

Reviewer 4 Report

Comments and Suggestions for Authors

The authors must correct the format of the tables, in the style of the journal.

Comments on the Quality of English Language

None