Monday, November 21, 2016

New Study on E-Cigarettes and Bronchitis: An Example of Scientific Dishonesty and Deception

A new study published online ahead of print in the American Journal of Respiratory and Critical Care Medicine along with a press release issued by the American Thoracic Society (publisher of the journal) conclude that youth use of electronic cigarettes contributes to "the burden of chronic respiratory symptoms in youth." Specifically, the research concludes that there appears to be a causal relationship between e-cigarette use and "chronic bronchitic symptoms (chronic cough, phlegm or bronchitis)" among youth.

According to the press release: "E-cigarette use among teenagers is growing dramatically, and public health experts are concerned that these devices may be a gateway to smoking. Now, new research indicates that even if these young e-cigarette users do not become tobacco smokers, e-cigarettes may harm their health."

One of the study authors was quoted in the press release as stating: "The Food and Drug Administration recently banned the sale of e-cigarettes to children under 18 years of age, and California just prohibited sale to young adults under 21. Our results suggest that these regulations and an environment that discourages the initiation of any tobacco product may reduce the burden of chronic respiratory symptoms in youth."

While the paper considers the possibility that the relationship between e-cigarette use and chronic bronchitic symptoms may not be causal, it essentially concludes that there is most likely a causal relationship: "It is, therefore, more likely that the observed (2014) relationship of bronchitic symptoms with e-cigarette use was causal, because the e-cigarette effect estimates were unchanged after adjustment for 2010 symptoms and after restriction to adolescents with no bronchitic symptoms in 2010."

The study involved a cross-sectional examination of the relationship between e-cigarette use and a past-year history of bronchitis symptoms and asthma symptoms (wheezing). The sample consisted of 2,086 southern California 11th and 12th graders, interviewed in 2014. Students were asked to report whether they had never used e-cigarettes, used e-cigarettes but not in the past month, or used e-cigarettes in the past month. Based on this question, they were categorized, respectively, as never users, past users, or current users of e-cigarettes. Current users were further classified into infrequent or frequent users based on whether they reported using e-cigarettes more than twice in the past month.

In the analysis, the investigators examined the relationship between the presence in the past 12 months of bronchitis or asthma symptoms and e-cigarette use. These relationships were then examined after controlling for two critical, potential confounding variables: (1) individual cigarette use; and (2) secondhand smoke exposure in the home.

The paper's abstract (a summary of the most important findings) reports a total of five findings:

1. There was a significant positive relationship between past e-cigarette use and bronchitis symptoms (not controlling for smoking or secondhand smoke exposure).

2. There was a significant positive relationship between current e-cigarette use and bronchitis symptoms (not controlling for smoking or secondhand smoke exposure).

3. The relationship between current e-cigarette use and bronchitis symptoms was significant for both infrequent and frequent users and was stronger for frequent users than infrequent users (not controlling for smoking or secondhand smoke exposure).

4. After controlling for smoking and secondhand smoke exposure or restricting the analysis to never smokers, there was still a significant positive relationship between past e-cigarette use and bronchitis symptoms.

5. After controlling for smoking and secondhand smoke exposure, there was no significant relationship between e-cigarette use (past or current) and asthma symptoms.

The abstract also notes that the relationship between current e-cigarette use and bronchitis symptoms was "attenuated" after controlling for smoking and secondhand smoke exposure.

The Rest of the Story

What the abstract does not tell you is that there were several other important findings of the paper:

1. After controlling for individual smoking and secondhand smoke exposure, there was no significant relationship between current e-cigarette use and bronchitis symptoms.

2. Among never smokers, there was no significant relationship between current e-cigarette use and bronchitis symptoms.

3. After controlling for individual smoking and secondhand smoke exposure, there was no significant relationship between either infrequent or frequent current e-cigarette use and bronchitis symptoms, and there was no significant trend.

Why weren't these critical results reported in the abstract? It certainly creates the appearance that the paper is trying to hide these important findings.

Moreover, it is deceptive to state that the relationship between current e-cigarette use and bronchitis symptoms was "attenuated" after controlling for individual smoking and secondhand smoke exposure. Actually, the relationship was no longer statistically significant. In fact, in none of the many analyses conducted was there a significant relationship between current e-cigarette use and any adverse respiratory symptoms - bronchitic or asthmatic - after controlling for individual smoking and secondhand smoke exposure.

In lieu of reporting these important "negative" findings, the paper instead uses the abstract to articulate the unadjusted study findings (results that do not control for smoking or secondhand smoke exposure), which are essentially meaningless because without controlling for these variables, one would actually expect to see an increase in respiratory symptoms in e-cigarette users (since we know e-cigarette use is highly correlated with smoking). So we can basically throw out the first 3 of the 5 reported results in the abstract.

What we are left with then, is that there was no association detected between e-cigarette use and asthma symptoms and that there was a positive association between past e-cigarette use and bronchitis symptoms, but not between current e-cigarette use and bronchitis symptoms. Taken as a whole, these findings do not support the conclusion that e-cigarette use has any adverse respiratory effects.

If there is no significant effect of current e-cigarette use on respiratory symptoms, then how can the study conclude that there is such an effect? The paper seems to gloss over the fact that the paper fails to find a dose-response relationship and in fact finds the opposite. Youth who use e-cigarettes frequently (at least once a month) are no more likely than never users to experience respiratory symptoms. The increase in respiratory symptoms was found only in youth who almost never use e-cigarettes!

While this study does not prove that e-cigarettes have no respiratory effects among youth, neither does it provide evidence that there is such an effect.

In fact, the headline of the paper - and its primary conclusion - should have been that there was no observed adverse effect of current e-cigarette use (including frequent e-cigarette use) on either bronchitic or asthmatic symptoms in youth vapers. Thus, there is no basis for the conclusion that electronic cigarette use contributes to "the burden of chronic respiratory symptoms in youth."

What concerns me even more than the fact that the study conclusion appears to be severely biased is the fact that the paper and the press release are both very deceptive. The paper abstract is hiding important findings. The same thing is true of the press release. In fact, nowhere in the press release does it even mention that there was no significant relationship found between current vaping and respiratory symptoms of any kind.

The press release also emphasizes the unadjusted findings, which is going to greatly confuse reporters and lead to inaccurate reporting of the findings of the study.

The rest of the story is that essentially what we have here is an example of scientific dishonesty and apparently intentional deception of the journal readers and the public. For a movement that has devoted so much attention to attacking the tobacco industry for its deception and scientific dishonesty, I believe that we need to adhere to the highest standards of honesty and transparency in our scientific reporting. This is not happening in our reporting of the health effects of vaping, and it is certainly not happening in this study and the dissemination of its results.

No comments: