Thursday, January 12, 2017

CDC Omitted Important Findings in Report on 2015 National Youth Tobacco Survey





The U.S. Centers for Disease Control and Prevention released selected information from the 2015 National Youth Tobacco Survey in April 2016 (here).  The agency cherry-picked numbers from previous surveys to portray e-cigarettes as a threat to teens, while ignoring sharp declines in teen smoking (here, here and here).

Last month, the CDC released the underlying 2015 NYTS data, ending an eight-month embargo that prevented analysis by independent investigators. 

The agency traditionally reports current smoking and vaping numbers separately, ignoring dual use and wrongly suggesting that the numbers are independent.  In contrast, here I report exclusive users of these products as well as dual users.  I also add important findings that the CDC omitted.

The chart shows current use (that is, on one or more days in the past 30) of cigarettes and e-cigarettes among middle and high school students over the four-year period.  The CDC emphasized the large increase in e-cig use from 2011 to 2015.  As I noted earlier (here), the large spike from 2013 to 2014 was likely due to a change in the NYTS questions.  It is clear that the increase slowed during the last year, but prevalence of exclusive e-cig use among high schoolers still  increased from 8.2 to 10.6%; dual use declined marginally from 5.2 to 5.0%.  The prevalence of exclusive smoking among high school students was flat at 4.0%. 

Given that the CDC has relentlessly asserted that e-cigarettes are a gateway to smoking, it is unsurprising that the agency did not publicize responses to two questions in the 2015 survey that focused on which products students used first.  I report here for the first time how high school students answered these questions, comparing two groups of current smokers: those who only smoked cigarettes and dual users of cigarettes and e-cigs.

The question, “Which of the following tobacco products did you try first?” produced these results:


Table 1. Percentage of High School Current Exclusive Smokers and Dual Users Who Tried Various Tobacco Products First, NYTS 2015
ProductExclusive SmokersDual Users



Cigarettes60.2%66.8%
Cigars11.36.0
E-cigarettes6.58.7
Smokeless tobacco9.17.6
Hookah4.56.7
Other products1.12.6
Not sure2.91.3
Never tried any product4.40.3


All100%100%
Percentages in bold: Exclusive smokers significantly different than dual users.


The table shows that there were no significant differences between exclusive smokers and dual users in the product first used.  Most had tried cigarettes first (60% and 67% respectively); the rest had chosen cigars, smokeless, e-cigs and other products.  There is little evidence here for the CDC’s claim that e-cigarettes are a gateway to smoking.

A question about initiation further exposed the relationship between cigarette and e-cigarette use.  Here is how current exclusive smokers and current dual users responded:



Table 2. Relationship of Cigarette and E-Cigarette Use Among High School Current Exclusive Smokers and Dual Users, NYTS 2015
ResponseExclusive SmokersDual Users



Never tried cigs or e-cigs16.4%1.8%
Only tried cigs19.82.7
Only tried e-cigs0.51.9
Tried cigs before ever tried e-cigs48.777.2
Tried e-cigs before ever tried cigs 14.616.4


All100%100%
Percentages in bold: Exclusive smokers significantly different than dual users.


This table shows that large majorities of high school current exclusive smokers and dual users started with cigarettes.  Only 15-16% of these students are even eligible to be considered gateway cases, in which users moved from vaping to smoking.

There are troubling inconsistencies in these tables.  For example, 16% of current exclusive smokers – who only used cigarettes in the past 30 days – responded that they never tried cigarettes (Table 2).  Four percent of exclusive smokers responded that they had never used any tobacco product (Table 1).

The “Y” in NYTS stands for youth, and responses from these surveys are known to be inconsistent.  In working with the data, if one eliminates participants who gave inconsistent responses, there would be considerably fewer valid participants.  I discussed this problem in 2015 (here), when I called on the CDC to “issue a comprehensive report on the internal consistency and relative validity of the NYTS data.” The agency appears to have ignored that issue.


Wednesday, January 4, 2017

Confirmed: Snus Use Protective for Parkinson’s Disease



Investigators in Sweden, Italy and the United States report that “non-smoking men who used snus had a substantially reduced risk of Parkinson’s disease…”

The research, published in the International Journal of Epidemiology (abstract here), combined data from seven Swedish cohort studies involving nearly 350,000 men.  Subjects were classified according to tobacco use and diagnosis of Parkinson’s disease (an illness of the nervous system affecting movement) over an average 16 years of follow-up.

The principal results are impressive:

“Among never-tobacco smokers, Parkinson’s disease risk in ever-snus users was lower than in never-users (pooled [hazard ratio, similar to relative risk] HR = 0.41, 95% [confidence interval] CI 0.28-0.61, for the fully-adjusted model).  Current-snus use was associated with a lower Parkinson’s disease risk than former use.  In addition, there was evidence of dose-response relationships such that moderate-heavy amount (pooled HR 0.41, 95% CI 0.19-0.90) and long-term current-snus users (pooled HR 0.44, 95% CI 0.24-0.83) had the lowest Parkinson’s disease risks.”

The bottom line: Current snus use, not former use, was strongly protective against Parkinson’s disease, with more protection from heavier and long-term use.

This is not the first such finding.  In 2009, I discussed (here) research from the American Cancer Society showing a similar strong protective effect (Relative risk, RR = 0.22, CI = 0.07 – 0.67) (abstract here).  Further, Parkinson’s may not be the only nerve illness for which smokeless tobacco and/or nicotine use is protective.  Snus users have a significantly lower risk for multiple sclerosis than nonusers of tobacco (here).  Nicotine has been found to improve performance in people with mild cognitive impairment, and it may also benefit those with Alzheimer’s disease (discussed here).

The current study represents a new era in Swedish snus research.  It was conducted by the Swedish Collaboration on Health Effects of Snus Use, “which brought together Swedish prospective cohort studies with detailed information on tobacco smoking and snus use.”

In the past, the snus research field was dominated by investigators at the Karolinska Institute; they published a series of studies that featured obvious technical problems and contradictions, and routinely found significant, small risks.  I documented these flawed studies in professional journals and in my blog (here, here, here, and here).

It is my hope that the Swedish Collaboration, with investigators from multiple universities in Sweden and beyond, will produce valuable, unbiased research on the health impact of snus use.


Wednesday, December 28, 2016

Sacramento Bee Errs Twice with Fake News on E-Cigarettes



The Sacramento Bee on January 21 published a column written by a hearing aids company executive.  Titled “E-cigarettes may also cause hearing loss” (here), the piece asserted that “damage to inner ear of teen [sic] is an overlooked potential health risk to vaping” and that “nicotine – regardless of whether it is inhaled in smoke or in vapor – presents a significant risk to hearing.”

These claims are fallacious.  On January 27, the newspaper published my correction online:

“There is virtually no scientific evidence to support Dave Fabry's claim.  I conducted a search of Medline, which contains journal citations and abstracts for biomedical literature from around the world for the period 1946-2016.  Nicotine is identified as a topic in 22,218 medical publications, and hearing loss is identified in 11,984 articles.  There are only two articles matching both terms: a 1956 article on vitamin therapy of chronic deafness published in Italian, and a 1964 article entitled "Are You Smoking More But Hearing Less?"  It is almost impossible for Dave Fabry's claim to be valid if these two articles are the only relevant scientific publications in the world's biomedical literature for the past 70 years.”

Days later, the Bee deleted the correction but left other comments.  Reader Jim McDonald observed:

“Why did you delete Dr. Rodu's comments? He did a search for studies on this topic, going back to 1946 and found nothing to support Mr. Fabry's claim. Dr. Rodu is a professor at the School of Medicine at the University of Louisville. That seems relevant.

“You also deleted mine from earlier today. I was not disrespectful.

“If you print opinions and offer a place for comment, you should expect opposing points of view.”

Nicotine has nothing to do with hearing loss, but smoking might worsen age-related impairment (here, here and here) via damage to small blood vessels in the ear.

Kudos to Mr. McDonald and another reader who brought the deletion to my attention.  The newspaper erred in publishing fake news, then compounded its mistake by suppressing truthful corrective responses.