In March 2020, science agreed that tobacco smoking is one of the most crucial preventable risk factors of poor Covid-19 outcomes, said a June paper in the European Journal of Internal Medicine by epidemiologists in Italy. It was becoming the turning point for the recommendations to cease smoking.
Meta-analysis Errors (and More Errors Again)
A May meta-analysis of five studies by Lippi and Henry, however, halted the push. The title of this meta-analysis, “Active smoking is not associated with severity of coronavirus disease 2019 (COVID-19),” conveys its point. “This study enjoyed widespread visibility in the scientific literature, as well as the lay press and various online social networks,” the June review stated. “It has been cited, posted or tweeted, particularly by researchers or subjects financially supported by the tobacco industry.”
But “Lippi and Henry were only correct in one out of the 5 studies.”
Shortly after, Professor Fei Guo at the Department of Family Medicine of National Taiwan University published “Active smoking is associated with severity of coronavirus disease 2019 (COVID-19): An update of a meta-analysis.” In this review, she found errors in the meta-analysis of Lippi and Henry. Manually checking the studies included, the professor found four out of five studies had wrong values inputted. As she stated: “Lippi and Henry were only correct in one out of the five studies.” Moreover, she did a meta-analysis using the same five studies and found smokers had a 120% increased odds of progression into severe Covid-19.
But the damage is done. “It has been cited, posted or tweeted, particularly by researchers or subjects financially supported by the tobacco industry.”
“We double-checked the two meta-analyses and agree with all the criticism raised by [Professor] Guo,” the June review said. They further listed seven minor mistakes, some of which Professor Guo did not point out, and one fundamental methodological flaw in the meta-analysis of Lippi and Henry.
In response to the rebuttal, Lippi and Henry wrote a research letter, “Active smoking and COVID-19: a double-edged sword”, to an academic journal where they included two more studies to justify the validity of their meta-analysis. But again, their research interpretations and methods have mistakes, as the June review also exposed.
What More Robust Meta-analyses Show
Adding on, the June review cited two meta-analyses that are much more robust and competently done.
- Researchers at the University of San Francisco synthesized 19 published papers and calculated that smokers are nearly twice as likely to suffer severe Covid-19 than non-smokers — i.e., 30% vs 18%. “Smoking is a risk factor for progression of Covid-19,” the authors closed. “With smokers having higher odds of Covid-19 progression than never smokers.”
- A pre-print by researchers in London harmonized data from 28 studies, of which 22 in China, three in the US, one in France, and one mainly based in the UK. Many of these, however, were of low quality as the authors judged. Nevertheless, smokers had a 37% increased risk of more severe Covid-19 than non-smokers.
(More evidence on cigarette smoking as a risk factor for severe Covid-19 can be found in the link below.)
Meta-Analyses Reveal Who Should Be More Cautious of COVID-19
[updated 14/5] Why chronic obstructive pulmonary disease — or smoking — and COVID-19 are the worst possible duo
It Probably Wasn’t Intentional
“In conclusion, the meta-analysis by Lippi and Henry suffers from a surprising number of errors, resulting in misleading conclusions,” epidemiologists of the June review wrote. “It is the only review so far indicating no relation between smoking and Covid-19 severity.” But the epidemiologists believed that those mistakes are unintentional or “honest errors” as Lippi and Henry declared no competing interests in their publications. Lippi and Henry are also no experts in tobacco research.
“Unfortunately, this has substantially reduced the efficacy of the tobacco control community’s claims to support smoking cessation in the Covid-19 era.”
Currently, 17 more studies confirm that cigarette smoking increased the risk of progression into more severe Covid-19, the June review said. Despite that, supporters of the tobacco industry still use the meta-analysis of Lippi and Henry to deny that smoking worsens Covid-19 condition.
“Unfortunately, this has substantially reduced the efficacy of the tobacco control community’s claims to support smoking cessation in the Covid-19 era,” the June review ended. As per its title: “No double-edged sword and no doubt about the relation between smoking and COVID-19 severity.”