The association of smoking and influenza infection and transmission

Abstract

Background: Recent influenza outbreaks and vaccine mismatch events has highlighted the deficiency in influenza mitigation approaches. Prevention strategies targeting the current high-risk groups may not be effective as expected. Given that cigarette smoking is a well-established causal risk factor for various diseases including chronic respiratory illnesses, the unravel of true association between smoking and influenza pathogenesis may be beneficial to the mitigation of influenza, by better allocation of treatments and resources.

Objectives: To assess the relation between cigarette smoking and influenza infection in household transmission setting and verify the possibility of identifying individuals with experience of smoking as influenza risk group.

Methods: Prospective cohort study of 1978 subjects (283 index patients, 1695 household contacts) from the Hong Kong Non-pharmaceutical intervention study was recruited from 2007 to 2014. Daily symptoms were reported by the subjects, nose and throat swab (NTS) specimens were collected from each home visits and tested by quantitative reverse transcription polymerase chain reaction (RTGPCR).

Smoking status was categorized into ever smokers (current and ex-smokers) and never smokers. Three outcome measures, namely laboratory confirmed influenza, influenza-like illness and acute respiratory disease were used as outcome measures. Multivariable logistic regression and accelerated failure time model were used to assess the association of smoking and influenza transmission in household and the association of smoking and influenza disease duration respectively.

Results: The secondary infection risk of ever smokers was found to be 7.59% and 9.08% for never smokers when laboratory confirmed influenza was used as outcome. No association between exposure of smoking and influenza transmission in household was discovered when adjusted for age and sex (adjusted odds ratio: 0.84, 95% CI: 0.47G1.42). The exposure of smoking was also not significantly associated with duration of all symptoms (adjusted acceleration factor (AF): 0.82, 95% CI: 0.57G1.18), fever (adjusted AF: 0.93, 95% CI: 0.78G1.10), respiratory symptoms (adjusted AF: 0.83, 95% CI: 0.57G1.21) or viral shedding (adjusted AF: 1.10, 95% CI: 0.89G1.36), with the adjustment for age and sex. Conclusions: This study has provided evidences that exposure of smoking did not significantly affect either the risk of household transmission of influenza, or the duration of influenza illness. No evidences were available to support the inclusion of individuals with experience of smoking as influenza risk group. Nonetheless, provided the relatively small sample size of this study and the inadequate adjustment of confounders, a definitive conclusion of the association can yet be drawn. A more sophisticated study design, such as Mendelian randomization analysis with a larger sample size is warranted to provide a conclusive remark of the association.