Study Slated for Selection Bias
Harm reduction expert Clive Bates has joined in the condemnation of the recent study from the University of California San Diego, slating it for selection bias
-
by Dave Cross
Share on:
Planet of the Vapes reported on the University of California San Diego study on Monday. It claimed to find that vapes do not help smokers quit. As part of the peer review process, tobacco harm reduction expert Clive Bates has submitted a reply to the publishing journal, stating that the research is flawed due to selection bias.
The researchers of the study concluded: “In this representative cohort study of US smokers who used ENDS, neither daily nor nondaily vaping was associated with increased smoking cessation, and each was associated with reduced tobacco abstinence, suggesting that careful adjustment of confounding is critical in studies of ENDS and smoking cessation.”
ENDS stands for electronic nicotine delivery system (vapes).
Replying to the publishing journal, Clive Bates said that the “built-in selection bias” means the authors of the study would be completely unable to link vaping to smoking cessation observations.
He wrote: “The main problem is that the study baseline (2017) includes only smokers and, therefore, only the subset of vapers who vape at that time but have not quit smoking. Secondly, it counts people who switched from smoking to vaping between baseline (2017) and follow-up (2021) as non-users, but many people who were non-users at baseline may have quit by using vapes in the interim four years.”
The sample was made up of just smokers and then looked at looks at the smoking status at follow to see if vaping led to smoking. Clive explains that there are two “fatal problems” by using this method.
- The study deliberately selects vapers who find it hard to quit smoking
- People who successfully quit with vapes between baseline and follow-up are counted as non-vapers
“The people classified as vaping (ENDS users) are necessarily also smokers at baseline. This means that vapers who had already quit smoking with vapes are excluded, and only people who haven't managed to quit with vapes are included as vapers at baseline. So, the classification as a vaper at baseline excluded anyone who found it easy to quit smoking with vapes and left only those who had found it more challenging (i.e. likely to be more cigarette-dependent).
“The period between baseline and measurement of study outcomes coincided with a substantial rise in adult vaping in the United States - this is the period in which Juul came to dominate the ENDS market. It is quite likely, therefore, that many of those classified as ENDS non-users in 2017 would have used ENDS in the intervening period and successfully quit smoking as a result. In this analysis, people whose entire process of switching from cigarettes to ENDS happened after 2017 but before 2021 are counted as ENDS non-users. This exaggerates the smoking cessation impact of non-use and understates the impact of ENDS use at follow-up.”
Clive references two studies to support his comments:
- Switching Away From Smoking and Reduction in Cigarette Consumption Among U.S. Adult Purchasers of the JUUL System Across 24 Months Including Diverse Subpopulations Disproportionately Affected by Cigarette Smoking, and
- Switching away from Cigarettes across 12 Months among Adult Smokers Purchasing the JUUL System
Photo Credit:
Author generated, Clive Bates added
Dave Cross
Journalist at POTVDave is a freelance writer; with articles on music, motorbikes, football, pop-science, vaping and tobacco harm reduction in Sounds, Melody Maker, UBG, AWoL, Bike, When Saturday Comes, Vape News Magazine, and syndicated across the Johnston Press group. He was published in an anthology of “Greatest Football Writing”, but still believes this was a mistake. Dave contributes sketches to comedy shows and used to co-host a radio sketch show. He’s worked with numerous start-ups to develop content for their websites.
