PREreview of The incubation periods of Monkeypox virus clade Ib

Peer Review of: The Incubation Periods of Monkeypox Virus Clade Ib Preprint Reviewer: Mam Roheya Jack 

Co-reviewer: Leen Arnaout

Summary and Overall Impression:

This manuscript explores an epidemiologic assessment  estimating Monkeypox virus (MPXV) clade Ib's incubation period in Uvira, Democratic  Republic of Congo (DRC). The authors  incorporate clinical,  laboratory, and exposure information to create a Bayesian model to estimate the virus’s  incubation period based on symptom onset criteria, and stratify incubation period estimates according to  demographic factors and transmission modes. The findings are that the median  incubation duration is 18.6 days (95% CrI: 15.2–22.6) with extensive heterogeneity  between non-sexual (18.9 days) and sexual (9.5 days) transmission. The findings offer  new insights on MPXV clade Ib, necessitating revised global post-exposure monitoring  recommendations.

The study is timely and highly relevant, particularly given recent global transmission of  MPXV  as well as the WHO's emergency declaration. The study is methodologically sound, using robust Bayesian inference techniques to generate realistic incubation period estimates while considering variations in exposure, along with methods to correct for  data collection bias. The conclusions that have been made are well-evidenced in the  reported results and have powerful public health policy implications. The study contributes  valuable new information to the epidemiology of MPXV clade Ib and is important in the long run for outbreak readiness. The paper is scientifically  robust, unbiased in a commercial sense, and adheres to ethical research. The figures are  easily interpretable, although legends require additional  clarification. But a few of the issues would benefit from some clarifications and further  discussion mainly regarding potential underreporting of exposure types.

Recommendation: Minor Revisions Required

Major Issue:

1. Potential Underreporting of Exposure Types:

The study acknowledges that, for every patient surveyed, only the most recent exposure event was  recorded. However, some cases may have had multiple exposures,  potentially affecting incubation period estimates. Although the authors acknowledge addressing these  uncertainties through sensitivity analyses, I recommend that the authors provide more details on how  they were modeled to enhance clarity. In particular, elaborating on the statistical methods used to adjust for misreported exposure dates or  multiple exposures would strengthen the results section. Perhaps moving those details from the Supplementary Methods to the paper’s main body would be helpful.

Section-by-Section Review:

A. Abstract

The abstract accurately summarizes the study but could be more concise. There  was a repetitive mention of Incubation Period Analysis that could be condensed  into one sentence to reduce redundancy. When summarizing the findings, providing multiple  quantiles in the abstract makes it overly technical. A more concise summary of  the incubation period findings would improve readability.

B. Background

The background and rationale for the study are well articulated, but the discussion on MPXV clade  comparisons could be expanded. The first paragraph introduces MPXV clade Ib but does not compare it adequately to clade Ia and clade II, such as in this sentence,"Data on how this clade differs from  others, including the incubation period which both shapes outbreak dynamics and  informs epidemic response strategies, remain limited." Therefore, I recommend the authors Provide a brief historical context on the known incubation  periods of clade Ia and II, highlighting what is expected to be different in clade Ib.

Also, in the statement about transmission dynamics (line 3-4), a major epidemiological  distinction from other clades was stated, yet there is no in-depth discussion on  how transmission pathways differ between clades. This needs further elaboration.

C. Methods

While this paper employs Bayesian modeling to take into account variations in patient exposure to MPXV, the methodology for exposure tracking is a potential limitation, as only the most  recent exposure event was considered. The paper’s body provides limited details on:

• Prior distributions used in the Bayesian model: they mention “Log-normal, Gamma and Weibull” distributional assumptions of incubation periods, yet do not elaborate on why these three distributions were chosen, how they were evaluated, and what was the final choice in the model.

• How posterior distributions were validated.

• Model comparison metrics were not elaborated upon, beyond mentioning they used “estimated leave-one-out cross-validation log-predictive density” as a metric to validate model performance..

I recommend the authors:

Include a discussion on the implications of exposure misclassification and conduct  a sensitivity analysis.

• Specify which priors were used (e.g., weakly informative, uninformative, or  empirical).

• Explain how model convergence was assessed (e.g., R-hat values,  posterior predictive checks).

• Clarify why the Weibull distribution was chosen for incubation periods over  alternatives like Gamma or Log-normal.

• Provide model performance comparisons (e.g., include WAIC/DIC scores in  supplementary material).

The official MPXV case definition in DRC differs from WHO’s standard. The study  uses a simplified case definition that may introduce misclassification bias. "At the Uvira MTC, following outbreak confirmation, a simplified case definition was  applied: anyone presenting with a skin rash or papulo-vesicular or pustular  eruptions was considered a suspected mpox case."

I recommend the authors:

• Justify why a simplified case definition was used instead of the full DRC or  WHO criteria.

• Acknowledge the potential for false positives, particularly in febrile rash  illnesses.

• Discuss whether case misclassification could bias incubation period  estimates, and how the authors plan to address this in their models.

The manuscript also does not describe how missing data were handled for cases  without PCR confirmation, Incomplete exposure history, Lost follow-up patients  or otherwise missing critical data. I recommend the authors add an explicit section on handling missing data in the Methods section.

D. Results: In general, the  findings are clearly presented. Some tables and figures lack clarity, particularly in the legends.

I recommend the authors provide more detailed figure captions to clarify statistical  methodology (e.g., model selection, assumptions).

E. Discussion

The discussion effectively links findings to public health policy interventions to limit MPXV spread. The comparison of incubation periods of clade Ib with clade IIa is limited and does not fully address why clade Ib  has a longer incubation period. Therefore, I recommend the authors expand on biological or epidemiological explanations for  longer incubation periods in clade Ib, or recommend future explorations into possible causes for such biological differences..

F. Conclusion

The conclusion appropriately summarizes the findings and their implications for  surveillance and outbreak control. The recommendation to revise WHO guidelines is impactful but needs a stronger  justification based on existing literature on the incubation periods and epidemiology of clade Ib, in relation to other clades. Therefore I recommend the authors strengthen the argument for guideline revisions with supporting  evidence from similar epidemiological studies.

Competing interests

The authors declare that they have no competing interests.