Cherry-picking of vaping research is reminiscent of tobacco industry days

Organisation: Position: Deadline Date: Location:

VapingAn article in the American Journal of Medicine reports no association between current vaping and self-reported cardiovascular disease among never smokers. But among dual users there were increased odds of having ever been told one has cardiovascular disease, writes Professor Michael Siegel of Boston University. The response of a prominent tobacco control researcher echoed what the tobacco industry used to do – dismiss findings one doesn’t like.

Siegel, of Boston University’s School of Public Health, probed the issue of research ‘cherry-picking’ in a 9 April article for his blog titled “The Rest of the Story: Tobacco and Alcohol News Analysis and Commentary…Providing the whole story behind tobacco and alcohol news”.

Like previous studies of its kind, this was a cross-sectional study that examined the association between current vaping-smoking status and ever having been told that one has cardiovascular disease (including heart attack, coronary artery disease, or stroke).

Previous studies used data from the National Health Interview Survey (NHIS) or the Population Assessment of Tobacco and Health (PATH) study; this paper used data from the Behavioral Risk Factor Surveillance Survey (BRFSS).

The authors reported no association between current vaping and self-reported cardiovascular disease among never smokers. However, they found that among current dual users there was an increased odds of having ever been told that one has cardiovascular disease.

The rest of the story

In response to the study, one prominent tobacco control researcher concluded that dual use is causally associated with cardiovascular disease, while dismissing the negative finding that e-cigarette use was not associated with vaping among never smokers.

He wrote: “The fact that the authors did not find an effect of e-cigarettes alone may be because they stratified the sample on e-cig and cigarette use, which reduces the sample size for each comparison, and so the power to detect an effect.”
This reminds me of what the tobacco industry used to do. If they saw a finding that they liked, they would emphasise that finding, but if they saw a finding that they didn't like, they would just dismiss it.


This is sometimes called ‘cherry-picking’. I've never picked cherries, but I assume that when doing so, one only picks the cherries that you like and disregards the ones that you don't.

As objective scientists, we can't cherry-pick. It allows one to have a pre-conceived conclusion and then to simply publicise findings that support the conclusion while dismissing those that do not. It appears that this is what is going on here.
Cherry-picking is becoming more and more common among tobacco control researchers and advocacy groups.

Recently I spoke at a conference on vaping, and one of the other speakers on the panel told the audience that there was no evidence vaping can help people quit smoking. The basis of that conclusion was that “there is no clinical trial that shows vaping to be effective… we need a clinical trial”.

When I then pointed out that a randomised clinical trial published in March in the prestigious New England Journal of Medicine found that vaping was twice as effective as nicotine replacement therapy for smoking cessation, they simply dismissed it, saying: “Well I still don't think it's effective.” (The same person also did not think there is enough evidence to conclude that vaping is any safer than smoking.)

The reality is that many tobacco control researchers and advocates will not be convinced by any amount of data. You could have two clinical trials, both finding that vaping is effective for some smokers, and they would still dismiss the findings. (In fact, we do have two clinical trials – it is amazing to see how many tobacco control advocates continue to insist that there have not been any clinical trials on the use of vaping for smoking cessation.)

Here, a positive finding is accepted and touted, while a negative finding is just dismissed. The reasoning given – that the study didn't have the power to detect an effect – doesn't hold water because the sample size of never smoking vapers in the study (15,863) exceeded the number of dual users (12,908).

But even the conclusion that dual use is causally associated with cardiovascular disease is unsupported by the evidence presented in the paper. This is a cross-sectional study, so it is entirely possible that the onset of cardiovascular disease preceded the vaping.

In fact, this is almost certainly the case for most of the study subjects because e-cigarettes have been popular for only about eight years, and it takes decades for cardiovascular disease to develop.

It may actually be that the cardiovascular disease ‘caused’ the vaping because having a heart attack or stroke is a strong stimulus for a smoker to try to quit, and many smokers try to quit by using e-cigarettes.

Moreover, dual users are almost certainly a different population from exclusive vapers and one systematic difference between the groups is likely that dual users have a heavier or more intense smoking history, making it more difficult for them to get off of e-cigarettes. If this were the case, it would explain the observed finding that dual use was associated with a higher risk of reporting cardiovascular disease.

The bottom line
The bottom line is that we can't draw causal conclusions from a cross-sectional study like this one, especially one in which it is impossible to determine which came first: the heart attack or the vaping.

So to tout the association between dual use and heart disease as a causal finding is bad enough. But cherry-picking findings that support a pre-determined conclusion, while dismissing those which do not support that conclusion, is sinking to the level of the tobacco industry which we once criticised for doing the very same thing.

* Dr Michael Siegel is a professor in the Department of Community Health Sciences at Boston University School of Public Health. He has 32 years of experience in the field of tobacco control. He previously spent two years working at the Office on Smoking and Health at CDC, where he conducted research on secondhand smoke and cigarette advertising. He has published nearly 70 papers related to tobacco. He testified in the landmark Engle lawsuit against tobacco companies, which resulted in an unprecedented $145 billion verdict against the industry. He teaches social and behavioral sciences, mass communication and public health, and public health advocacy in the masters of public health programme.

The association between e-cigarette use and cardiovascular disease among never and current combustible cigarette smokers: BRFSS 2016 & 2017


Albert D Osei, Mohammadhassan Mirbolouk, Olusola A Orimoloye, OmarDzaye, SM Iftekhar Uddin, Emelia J Benjamin, Michael E Hall, Andrew P DeFilippis, Andrew Stokes, Aruni Bhatnagar, KhurramNasir, and Michael J Blaha.


The prevalence of e-cigarette use in the United States has increased rapidly. However, the association between e-cigarette use and cardiovascular disease remains virtually unknown. Therefore, we aimed to examine the association between e-cigarette use and cardiovascular disease among never and current combustible cigarette smokers.


We pooled 2016 and 2017 data from the Behavioral Risk Factor Surveillance System (BRFSS), a large, nationally representative, cross-sectional telephone survey. We included 449,092 participants with complete self-reported information on all key variables.

The main exposure, e-cigarette use, was further divided into daily or occasional use, and stratified by combustible cigarette use (never and current). Cardiovascular disease, the main outcome, was defined as a composite of self-reported coronary heart disease, myocardial infarction, or stroke.


Of 449,092 participants, there were 15,863 (3.5%) current e-cigarette users, 12,908 (2.9%) dual users of e-cigarettes + combustible cigarettes, and 44,852 (10.0%) with cardiovascular disease. We found no significant association between e-cigarette use and cardiovascular disease among never combustible cigarette smokers.

Compared to current combustible cigarette smokers who never used e-cigarettes, dual use of e-cigarettes + combustible cigarettes was associated with 36% higher odds of cardiovascular disease (Odds Ratio [OR], 1.36; 95% CI, 1.18–1.56); with consistent results in subgroup analyses of premature cardiovascular disease in women less than 65 years and men less than 55 years old.


Our results suggest significantly higher odds of cardiovascular disease among dual users of e-cigarettes + combustible cigarettes compared to smoking alone. These data, though preliminary, support the critical need to conduct longitudinal studies exploring cardiovascular disease risk associated with e-cigarette use, particularly among dual users.

New study finds vaping is not associated with cardiovascular disease among never smokers; But tobacco control researcher dismisses findings


The association between e-cigarette use and cardiovascular disease among never and current combustible cigarette smokers:


Receive Medical Brief's free weekly e-newsletter

Related Posts

Thank you for subscribing to MedicalBrief

MedicalBrief is Africa’s premier medical news and research weekly newsletter. MedicalBrief is published every Thursday and delivered free of charge by email to over 33 000 health professionals.

Please consider completing the form below. The information you supply is optional and will only be used to compile a demographic profile of our subscribers. Your personal details will never be shared with a third party.

Thank you for taking the time to complete the form.