Current smokers more likely to experience COVID-19 symptoms

Individuals who are smokers are more likely to report the classic symptoms suggestive of COVID-19 (fever, new persistent cough, and breathlessness) and have a greater symptom burden than non-smokers, a prospective UK-based study found.

“Our results provide compelling evidence for an association between current smoking and individual risk from COVID-19,” said the researchers.

“The present data suggest that smoking cessation should be considered as an element in strategies to address COVID-19, as smoking increases both the likelihood of symptomatic disease … and the severity of disease,” they said. In addition, reducing smoking rates in the overall population may help reduce hospitalizations due to several other conditions which also burden the healthcare system.

The results were based on analysis of information provided by 2,401,982 UK users of the Zoe COVID-19 Symptom Study app aged 16–90 years (mean age 43.6 years, 63.3 percent female, mean BMI 26.47 kg/m²) between March 24 and April 23, 2020. This included information on demographics, anthropometrics, smoking status, medical conditions, and medication use. Participants also noted their daily health status as well as SARS-CoV-2 testing status. Those who reported feeling unwell (35 percent; n=834,437) provided additional information including hospital presentation and presence of 14 potential COVID-19 symptoms*. Eleven percent of the population were smokers (n=240,873).

The participants were categorized into four groups: those who tested positive for SARS-CoV-2 (SC2P; n=7,123), those who tested negative for SARS-CoV-2 (SC2N; n=16,759), those who, based on symptoms, believed that they already had a history of COVID-19 at registration (SC2S; n=157,406), and standard users (n=2,221,088).

Standard users who were current smokers were more likely to report classic symptoms of COVID-19 –  fever, new persistent cough, and breathlessness – compared with non-smokers (adjusted odds ratio [adjOR], 1.14, 95 percent confidence interval [CI], 1.10–1.18). [Thorax 2021;doi:10.1136/thoraxjnl-2020-216422]

With a higher number of symptoms being used as a surrogate for greater disease severity, current smokers were also more likely than non-smokers to report >5 symptoms (adjOR, 1.29, 95 percent CI, 1.26–1.31) and >10 symptoms (adjOR, 1.50, 95 percent CI, 1.42–1.58).

Smoking was tied to an increased COVID-19 symptom burden in all patient categories (β=2.93, 3.02, 3.28, and 1.95 for standard users, SC2S, SC2P, and SC2N, respectively). Smoking was also associated with >10 COVID-19 symptoms among participants in the SC2S and SC2P categories (ORs, 1.33 and 1.42, respectively).

“The pattern of associations between symptoms reported did not vary between smokers and non-smokers,” said the researchers. This suggests that the results are not due to pre-existing smoking-related symptoms.

The number of symptoms in the SC2P group was higher than that of the SC2N group and standard users (mean 6.16, 3.22, and 1.12, respectively). Smoking rates were lower in the SC2P compared with the SC2N group (7.4 percent vs 9.3 percent).

Among participants who had undergone testing, smoking was associated with a lower likelihood of having a PCR-positive COVID-19 test (OR, 0.73), a result not influenced by the higher rate of healthcare workers in the study population who underwent testing (vs the whole study population) or the lower smoking prevalence among healthcare workers. However, smokers who tested positive for SARS-CoV-2 had a higher symptom burden (OR, 1.42 for >10 symptoms) and were more likely to present at a hospital for COVID-19 (OR, 2.11) than non-smokers.

According to the researchers, studies have shown that tobacco smokers have an elevated risk of developing bacterial and viral respiratory infections, raising speculation that smoking may be a risk factor for more severe COVID-19 symptoms.

Smoking may increase the risk of infection due to the repeated hand-to-mouth movement. However, the increased severity with smoking “suggests that the whole curve of disease is shifted as opposed to just the chance of infection.”

A drawback of the study was that data on ethnicity was not initially collected and was only available for 12 percent of the population. Further study is warranted to assess a potential interaction between smoking and COVID-19 symptoms by ethnicity, the researchers noted. The results also do not apply to asymptomatic COVID-19.