Monday, November 12, 2018

Tobacco Control Special Issue Yields Smoke But No Hazard for IQOS Aerosol


Recently a special issue of the journal Tobacco Control offered 22 articles on Philip Morris International’s (PMI) IQOS heat-not-burn tobacco product (here), aiming to dissuade the FDA from allowing IQOS sales as reduced risk products for smokers.  In the words of editor Stanton Glantz: “Policy makers should give greater weight to the advice provided by public health scientists than to submissions from industry when it comes to regulating tobacco products such as heated tobacco products.”



Dr. Glantz’s advice has been inaccurate in the past. A year ago, when he argued that “PMI’s own data fails to support a modified risk claim in people who are actually using [IQOS],”  I detailed the errors in his analysis (here).  This new collection of anti-IQOS articles deserves the same scrutiny.



One treatise, by UCSF’s Gideon St. Helen and colleagues, acknowledges that “All substances in PMI’s list of 58 constituents (PMI-58) were lower in IQOS emissions compared with mainstream smoke of reference cigarettes.”  That accurate statement is followed by the suggestion that PMI ignored that “levels of 56 other constituents, which are not included in the PMI-58 list or FDA’s list of [harmful and potentially harmful constituents], were higher in IQOS emissions.”  The authors criticize PMI for not undertaking “non-targeted chemical analysis” to identify these compounds.



Dr. St. Helen et al. are mistaken.  The chart above, taken from PMI’s presentation to the FDA advisory committee last January (here), documents that the company performed that analysis.  PMI’s extensive investigation of “other” substances is confirmed on the FDA website, where a chart labeled “non-targeted screening” appears in a document submitted by PMI on December 8, 2017, and published by FDA on January 12, 2018 (here). 



The December filing acknowledged that “80 compounds were found to be of higher concentration or new in [IQOS] aerosols…compared to cigarette smoke.  A toxicological assessment was performed and identified 68 of those that do not present specific toxicological concern. 8 compounds present potential genotoxic concerns based on structure-activity computational alerts and 4 compounds are classified mutagens/carcinogens.” 



PMI scientists did not dismiss the findings: “Although a minor subset of compounds identified in this study show intrinsic (potential) toxicological hazards, overall toxicity of the THS generated aerosol was lower compared to cigarette [smoke]. In vitro studies demonstrated a marked decreased biological activity of [IQOS] aerosol compared to smoke. An in vivo 90 days inhalation study showed in general a lower biological activity of the mainstream aerosol from [IQOS] when compared to mainstream smoke.”



Importantly, PMI is undertaking more research: “Additional investigations are ongoing to further characterize the impact of the [mutagens/carcinogens] on the overall toxicity of [IQOS] products.”        

Wednesday, November 7, 2018

Truth Initiative Stumbles in JUUL Study


It’s surprising how many fundamentally flawed e-cigarette studies are aggressively promoted by their authors and allied organizations as grounds for FDA regulatory action.  It is also concerning when authors refuse to acknowledge or respond to honest scientific inquiries about their research.  Here is a representative case involving researchers associated with Truth Initiative, a non-profit anti-tobacco organization.

Researchers led by Donna Vallone, Ph.D., recently published a study in Tobacco Control on the “prevalence and correlates of JUUL [e-cigarette] use among a national sample of youth and young adults (here).”  While they reported that the overall prevalence of ever and current (past 30-day) JUUL use was 6.0% and 3.3% respectively, they failed to disclose information about the most important correlate of JUUL use – other e-cigarette use.

The authors noted that among underage children (15-17 years), current JUUL use was 6% and current use of combustible tobacco was 7%.  They connected these, finding that children who were currently smoking were five times more likely to use JUUL than non-smokers.  However, 11% of children in that age group currently used e-cigarettes.  They ignored this important correlate in their analyses.  Instead, they inexplicably included e-cig use among other members of the youths’ households.

There are other significant problems with this study.  First, youths and young adults were asked: “…on how many days did you smoke a Juul vape?” (emphasis added)  That wording likely confused participants. 

Second, the authors didn’t define current use of e-cigarettes, nor did they even give any description of the question in their survey. 

Third, the survey flow for JUUL and e-cigarette questions was not provided.  Were separate questions about these products asked of all participants, or did researchers ask first about e-cigarettes, and then only ask current e-cig users if they used JUULs? 

Fourth, Vallone et. al. defined JUUL “regular use” as 10-30 days in the past month, and they reported that 25% of youth fell into this category.  That percentage is grossly inflated.  The CDC and other authorities use a more credible “frequent” category of 20+ days (here and here), which would generate a lower percentage of users at risk.    

In summary, Vallone et. al. produced an error-ridden study focused on JUUL “smoking,” while ignoring the effect of other e-cigarette use.  The obvious problems ought to have been resolved in peer review.  Additional questions remain, owing to the fact that the authors used a private dataset.  When Truth Initiative posted the study on the Society for Nicotine and Tobacco Research listserv, I asked the authors to resolve some of these problems; they did not acknowledge my listserv post.


Thursday, October 25, 2018

CDC Data Reveal Many Far More Dangerous Teen Behaviors Than Vaping


In the hierarchy of teenage risk behaviors, government data shows that vaping pales compared to drinking-, driving- and weapon-related activities. It is remarkable that public health officials and the media focus so greatly on the former, to the detriment of teen safety and health.

A report from the CDC earlier this year (here), based on the 2017 Youth Risk Behavior Survey (YRBS), allows us to put the data in context.

The prevalence of past-month e-cigarette use in the 2017 YRBS was 13%.  That rate is higher than those for cigarettes (8.8%), cigars (8.0%) and smokeless tobacco (5.5%).  However, as I discussed recently (here), the vaping rate pales next to those for marijuana (19.8%) and alcohol (29.8%).  In fact, the e-cigarette rate is nearly identical to the rate for binge drinking (4 or 5 drinks within a couple hours).

These rates of drug use are troubling, but there is much worse in the federal report for parents and policymakers to be concerned about.  Following is a list of other risky behaviors by high school students in the past 30 days.


Prevalence (%) of Risky Behaviors Among American High School Students (YRBS, 2017)


Past 30 Days


Rarely/never wore a seatbelt (as an occupant)5.9%
Rode with driver who had been drinking16.5%
Drove after drinking5.5%
Drove after marijuana use13.0%
Texted or emailed while driving39.2%
Carried a weapon (e.g. gun, knife, club)15.7%


Past 90 Days
Had sexual intercourse28.7%
..…and used condom, 53.8% of previous


Past Year


Involved in physical fight23.6%
Physically bullied on school property19.0%
Electronically bullied14.9%


Felt sad or hopeless31.5%
Considered suicide17.2%
Made suicide plan13.6%
Attempted suicide7.4%



Media coverage of the CDC YRBS report (here) was largely confined to the sensational -- “Fewer teens having sex and using drugs, CDC says.”

The absence of focus on the more prevalent and dangerous behaviors in part reflects the FDA-led public health community fixation on vaping.  This may stem from the fact that, of the many greater-risk categories outlined by the CDC, the FDA’s regulatory authority only extends to tobacco. To a federal agency with a hammer, everything is a nail.