Structured PREreview of Association of COVID-19 risk factors with systemic fungal infections in hospitalized patients

Does the introduction explain the objective of the research presented in the preprint?

Yes

It outlines the context of the COVID-19 pandemic and the emergence of associated fungal infections, highlighting the clinical significance of these complications. The introduction identifies specific risk factors and therapeutic interventions that may influence the incidence of systemic fungal infections among COVID-19 patients.

Are the methods well-suited for this research?

Somewhat inappropriate

The authors examine the degree to which a number of clinical variables are associated with fungal infections. We would recommend that they carefully review their methods and approach with a clinical epidemiologist. Throughout the paper they implicitly suggest causal relationships among the variables, most notably by selectively suggesting that some variables predict fungal infection while fungal infection predicts other variables. As an example, fungal infection predicts mortality but diabetes predicts fungal infection. It is important in these types of analyses to distinguish between characteristics that clearly pre-date the fungal infection (e.g. age, sex) and those that occur concurrently with the fungal infection. Are fungal infections more likely to occur among patients who are seriously ill or are patients who get fungal infections more likely to become seriously ill? The current study isn’t able to answer that question as it is presented. Is being in the ICU a risk factor for acquiring a fungal infection or is a fungal infection a risk factor for being sent to the ICU? Or, are they unrelated, but both associated with a third factor, such as having severe COVID? To take another example, there is an association between steroid use and fungal infections. That could be because use of the steroids predisposes you to fungal infection or it could be that fungal infections are associated with severe COVID and people with severe COVID are treated with steroids. It would be useful to include a directed acyclic graph (DAG) in the paper to represent potential causal relationships and the relationships among the variables analyzed. It would be useful to do an analysis of the degree of correlation among all of the variables, not just each variable with fungal infection as that may help to better understand the nature of the relationships. Depending on the results of those examinations, it may also be useful to do a multivariable regression analysis rather than just the bivariate analyses that were conducted.

Are the conclusions supported by the data?

Somewhat unsupported

The authors state that the predictors of severe COVID predict fungal infections but they provide no evidence of that. It could be that only severe COVID predicts having a fungal infection and that, corrected for the severity of the COVID, none of the other risk factors provide any additional predictive power. Alternatively it could be that they each predict fungal infections, regardless of whether or not the patient has severe COVID (though that seems less likely). It could also be that none of them predict severe COVID but that they just predict fungal infections – and fungal infections predict severe COVID (super unlikely given the high number of severely ill patients without fungal infections). They also state: Steroid therapy may need to be modified to account for a risk or a presence of a fungal infection in vulnerable patients. We don’t dispute the statement, but it was true prior to this study. Steroid therapy has been previously well described as a risk factor for fungal infections and steroid therapy of patients with fungal infections must consider the possibility that it may exacerbate the infection. However, this study didn’t provide additional evidence to suggest that, at comparable levels of COVID severity, that the addition of steroids increased the risk of fungal infections. Because steroid use is associated with COVID severity and COVID severity appears to be related to fungal infections, it could be that the association between steroid use and fungal infections may just reflect the COVID severity relationship. As a result, we don’t see that the paper provides additional evidence of the potential harms associated with steroid use beyond those that are already well described.

Are the data presentations, including visualizations, well-suited to represent the data?

Highly appropriate and clear

The data presentations adequately summarize the information of the article. It would be informative to include the length of treatment in days in table S3 and S4, as well as including the results table from the supplemental data into the main text.

How clearly do the authors discuss, explain, and interpret their findings and potential next steps for the research?

Somewhat clearly

- Discussion goes into detail about their findings and provides clear interpretations for numerical data found, even describing statistical significance - Describes association between risk factors, COVID-19 and secondary fungal infections and talks about independence with respect to association - Confounders were not discussed nor efforts to address bias.

Is the preprint likely to advance academic knowledge?

Moderately likely

The paper contributes useful descriptive data on the relationship between fungal infections and patient characteristics. It may lead clinicians treating patients at high risk for fungal infections to be more attentive to the possibility of a fungal infection.

Would it benefit from language editing?

No

There may be minor language issues, but they do not impact clarity or understanding. Nevertheless, the statistical analysis results are currently presented in the supplemental information section, a recommendation would be to incorporate the results of both descriptive statistics (particularly percentages) and inferential statistics (specifically, confidence intervals and p-values) into the main text, this would enhance the understanding and provide a more easy way to analyze the data.

Would you recommend this preprint to others?

Yes, but it needs to be improved

It provides useful basic information and provides the basis for a more nuanced analysis of the relationship between risk factors and fungal infections among patients hospitalized with COVID.

Is it ready for attention from an editor, publisher or broader audience?

No, it needs a major revision

It is recommended that the methods in particular be reviewed, as there is confusion regarding variables and research questions that are not answered. We encourage modification of these previously described problems by a clinical epidemiologist to establish clarity with the variables and prediction of fungal infections.

Competing interests

The authors declare that they have no competing interests.