Jump to comment:

• Up in smoke: The reported association between e-cigarette use and wheezing in this study is probably spurious
  James D. Sargent, MD, Susanne E. Tanski, MD, Mary F. Brunette, MD, Jennifer A. Emond, MS, PhD and Steven Woloshin, MD
  Published on: 7 March 2019
• A different interpretation
  Stewart N Brock
  Published on: 6 March 2019

• Published on: 7 March 2019

  Up in smoke: The reported association between e-cigarette use and wheezing in this study is probably spurious

  • James D. Sargent, MD, Director/Professor of Pediatrics and Biomedical Data Science The C. Everette Koop Institute at Dartmouth/Geisel School of Medicine
  • Other Contributors:
    • Susanne E. Tanski, MD, Associate Director/Associate Professor of Pediatrics
    • Mary F. Brunette, MD, Associate Professor of Psychiatry
    • Jennifer A. Emond, MS, PhD, Assistant Professor of Biomedical Data Science
    • Steven Woloshin, MD, Co-director of the Center for Medicine and Media at The Dartmouth Institute/Professor of Medicine

  NOT PEER REVIEWED

  We read with interest a recent publication by Li, et. al, entitled Association of smoking and electronic cigarette use with wheezing and related respiratory problems in adults: cross-sectional results from the Population Assessment of Tobacco and Health (PATH) study, wave 2. The primary finding, reported in the abstract, was that risk of wheezing and related respiratory symptoms was significantly increased in current exclusive e-cigarette users compared to never users, with an adjusted odds ratio of 1.67 (1.23, 2.15). We think the report is misleading for several reasons.

  First, the main multivariable analysis (Table 2) did not adequately adjust for important confounders that impact wheeze, most importantly, cigarette smoking history. In most analyses of medical outcomes in adults, pack-years of smoking has a strong relation to smoking-related diseases, over-and-above current smoking status. Since three quarters of vapers in the main model were ex-smokers, cigarette smoking history is almost certainly contributing to the size and significance of the main reported finding. Other combustible tobacco use and current marijuana smoking would also be expected to exacerbate cough and wheeze. Our bet is that large numbers of e-cigarette users also use marijuana.

  The authors partially addressed smoking history with a secondary analysis (Table 3), in which they stratified by former smoking status. In that analysis, vaping was not significantl...

  Show More

  NOT PEER REVIEWED

  We read with interest a recent publication by Li, et. al, entitled Association of smoking and electronic cigarette use with wheezing and related respiratory problems in adults: cross-sectional results from the Population Assessment of Tobacco and Health (PATH) study, wave 2. The primary finding, reported in the abstract, was that risk of wheezing and related respiratory symptoms was significantly increased in current exclusive e-cigarette users compared to never users, with an adjusted odds ratio of 1.67 (1.23, 2.15). We think the report is misleading for several reasons.

  First, the main multivariable analysis (Table 2) did not adequately adjust for important confounders that impact wheeze, most importantly, cigarette smoking history. In most analyses of medical outcomes in adults, pack-years of smoking has a strong relation to smoking-related diseases, over-and-above current smoking status. Since three quarters of vapers in the main model were ex-smokers, cigarette smoking history is almost certainly contributing to the size and significance of the main reported finding. Other combustible tobacco use and current marijuana smoking would also be expected to exacerbate cough and wheeze. Our bet is that large numbers of e-cigarette users also use marijuana.

  The authors partially addressed smoking history with a secondary analysis (Table 3), in which they stratified by former smoking status. In that analysis, vaping was not significantly associated with any respiratory symptom among persons with no previous cigarette smoking history. The secondary analysis strongly suggests that if a measure capturing extent of former smoking was entered as a covariate in the main model, the primary reported finding of an association between e-cigarette use and pulmonary symptoms would have been null.

  Second, the analyses did not adjust for conditions that exacerbate wheeze (chronic obstructive pulmonary disease, congestive heart failure and certain cardiac medications), another potential source of residual confounding. These conditions may also cause declining health, prompting the person to quit smoking or switch from smoking to vaping, adding to a cross sectional association through reverse causality.

  While addressing this last issue requires longitudinal data, we believe adding former smoking status and marijuana smoking to the main analysis should be done. The inclusion of these covariates would provide a less biased estimate of the cross-sectional association between vaping and respiratory symptoms and perhaps reach entirely different conclusions. We hope the authors will publish updated results in their response to this comment.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 6 March 2019

  A different interpretation

  • Stewart N Brock, Public Health Specialist Somerset County council

  NOT PEER REVIEWED
  A key finding of this paper does not find its way into the abstract namely, "no significant differences in wheezing and related respiratory symptoms was found when comparing current vapers who never smoked with never smokers. " This tends to suggest, unsurprisingly, that it is the prior smoking history that is the critical factor in current wheeze. The paper supports the harm reduction hypothesis for switching to vaping completely from smoking as per the conclusion. But the conclusion also states in the first line, "Vaping was associated with increased risk of wheezing and related respiratory symptoms, " which is incorrect without adding "in current smokers." Vapers who previously smoked have lower risk than those who continue to smoke, including dual users, and those who never smoked have no increased risk.

  Conflict of Interest:
  None declared.

  • Back to top