?:abstract
|
-
BACKGROUND: Data is conflicting regarding the impact of tobacco smoking in people with pneumonia due to SARS-CoV-2 infection (COVID-19). METHODS: We performed a retrospective multicentre cohort study of 9,991 consecutive patients hospitalized in a major New York academic centre between March 7th and June 5th, 2020 with laboratory confirmed COVID-19. The clinical outcomes assessed included risk of hospitalization, in-hospital mortality, risk of intensive care unit (ICU) admission, and need for mechanical ventilation among smokers (current and former). Multivariable logistic regression and propensity score models were built to adjust for potential confounders. FINDINGS: Among 9,991 consecutive patients diagnosed with COVID-19, 2,212 (22.1%) patients were self-reported smokers (406 current and 1806 former). Current smoking was not associated with an increased risk of hospitalization (propensity score [PS]-adjusted OR 0.91; p=0.46), in-hospital mortality (PS-OR 0.77; p=0.12), ICU admission (PS-OR 1.18; p=0.37), or intubation (PS-OR 1.04; p=0.85). Similarly, former smoking was not associated with an increased risk of hospitalization (PS-OR 0.88; p=0.11), in-hospital mortality (PS-OR 1.03; p=0.78), ICU admission (PS-OR 1.03; p=0.95), or intubation (PS-OR 0.93; p=0.57). Furthermore, smoking (current or former) was not associated with an increased risk of hospitalization (PS-OR 0.85; p=0.05), in-hospital mortality (PS-OR 0.94; p=0.49), ICU admission (PS-OR 0.86; p=0.17), or intubation (PS-OR 0.79; p=0.06). INTERPRETATION: Smoking is a well-known risk factor associated with greater susceptibility and subsequent increased severity of respiratory infections. In the current COVID-19 pandemic, smokers may have increased risk and severe pneumonia. In the current COVID-19 pandemic, smokers are believed to have an increased risk of mortality as well as severe pneumonia. However, in our analysis of real-world clinical data, smoking was not associated with increased in-patient mortality in COVID-19 pneumonia, in accordance with prior reports. This article is protected by copyright. All rights reserved.
|