Review 2: "Possible Roles of Phytochemicals with Bioactive Properties in the Prevention of and Recovery from COVID-19"

RR:C19 Evidence Scale rating by reviewer:

• Misleading. Serious flaws and errors in the methods and data render the study conclusions misinformative. The results and conclusions of the ideal study are at least as likely to conclude the opposite of its results and conclusions than agree. Decision-makers should not consider this evidence in any decision.

***************************************

Review: Phytochemical compounds included in the vegetables may have contributed to preventing the contraction of COVID-19, and accelerating the recovery from the disease.

One of the major concepts in epidemiology is confounding. That means we need to consider other potential variables that might explain the association. For example, if we are interested in the effect of alcohol on lung cancer, cigarette smoking is a confounding variable and needs to be taken into account. In this study there is no consideration of confounding. Introduction does state that obesity and diabetes may influence risk of COVID-19, but those are not taken into account in the study. There are many more variables that are potentially relevant, including the frequency of contacts with other people, the prevalence of virus in the community etc. Vegetables, herbs/spices etc. are correlated with numerous life style variables that are potential confounders.

• https://en.wikipedia.org/wiki/Confounding 

A second problem is calculating the chi-square test and not the relative risk with confidence interval. When the number of participants is large, trivial differences can cause small P-values. Therefore P-values are a poor way to summarize the findings. Since 1990s statisticians have emphasized the usefulness of confidence intervals, in contrast to P-values.

• https://en.wikipedia.org/wiki/Relative_risk

• https://en.wikipedia.org/wiki/Confidence_interval 

A third problem is ecological fallacy which means group-level comparisons. Authors write “in India and Japan the people who contracted COVID-19 showed significantly shorter recovery time”. However, it is impossible to draw firm conclusions from such differences. It is even possible that the limit for a person to decide that there is a disease is different in different cultures. In any case there are large numbers of differences between countries and it is misleading to imply that “herbs/spices, and fermented foods/beverages” are the most plausible explanation for country-level differences.

• https://en.wikipedia.org/wiki/Ecological_fallacy 

A fourth problem is the lack of specificity. In modern science one of our goals is specificity. We try to find eg that substance X has effect E. However, “intake of vegetables, herbs/spices, and fermented foods/beverages” is a fuzzy variable. Even if such a variable might influence the risk and recovery from COVID-19, a positive findings does not guide further directions for research. For example, if there is benefit from green tea, but not from vegetables and herbs/spices, such a finding would motivate further research to the compounds in green tea. However, a fuzzy food variable does not have such indications.

A fifth problem is ignoring vitamin C as one of the substances in vegetables. The COVID A to Z study in the USA found that vitamin C increases the recovery rate of outpatient cases of COVID-19 by 70%. Various substances in vegetables etc are interesting, but the findings of a good quality RCT should not be ignored in introduction and discussion.

• https://doi.org/10.3389/fimmu.2021.674681