- Sunday, 20 November 2016 21:41
By Dr Farsalinos
A new study published in the American Journal of Respiratory and Critical Care has been widely covered by the media as showing that “e-cigarettes double the risk of bronchitis in teenagers”. The official press statement, which is released from the American Thoracic Society, is more moderate. It is titled: “E-cigarettes may harm teens' lung health”. In the text, it is mentioned that:
The study found that when compared to those who never tried e-cigarettes, the risk of the respiratory symptoms was
- approximately 85 percent higher among past users, and
- double among current users
But there is a problem with these statements. They are different from what the study found.
The study recruited adolescents (aged 16-18 years) but did not define those with chronic bronchitis based on physician diagnosis or medications intake. Instead, they ASKED them if they had “daily cough for 3 months in a row, congestion or phlegm other than when accompanied by a cold, or bronchitis in the previous 12 months”. Therefore, they based the “diagnosis” on a response to a question. However, all media articles discuss about lung disease and bronchitis. The latter is also wrong, because there is an acute (more common in adolescents) and a chronic (rare in adolescents) form of bronchitis. The question specifically addressed the definition of chronic bronchitis only. I will mention later in my comment why it is important to make the distinction.
The reported findings of 85% higher and double risk of respiratory symptoms among past and current e-cigarette users respectively is wrong for 2 reasons. First, they did not measure any risk at all. This was a cross sectional study in which they reported odds ratios, not relative risk. The word “risk” is inappropriately used, because there was no assessment of previous exposure (e-cigarette use) and subsequent outcome (symptoms). Additionally, the media report is inaccurate because such odds were found WITHOUT including smoking and second-hand exposure to smoking as covariates in the regression models. When these 2 factors were added to the model, ONLY past e-cigarette use was marginally associated with symptoms while current use was NOT associated with symptoms. Fortunately the authors did not suggest that it is better for adolescents to continue rather than stop using e-cigarettes. Additionally, in non-smoking adolescents who were using e-cigarettes, there was no increase in the incidence of symptoms. This clearly shows that any association between e-cigarette use and chronic bronchitis symptoms is coincidental.
Imagine doing a study to assess the association between varenicline use and chronic bronchitis. Since all varenicline users are current or former smokers, and smoking is the most important risk factor for chronic bronchitis, removing smoking as a covariate from the analysis would result in varenicline “increasing the risk” of chronic bronchitis. Obviously, this is wrong for varenicline.
But there is another more impressive problem with this study. When I first read the media stories I was puzzled because I thought chronic bronchitis is a rare condition at this young age. So, I looked for the CDC data on the incidence of chronic obstructive lung disease (COPD = chronic bronchitis and emphysema). For the 3-year period of 2007-2009, the prevalence of COPD among US adults was 5.1%. Prevalence increased with age, with the highest observed in age groups 75-84 year. For age group 18-24 years, the prevalence was 2.5%. The study in South California found that the age group 16-18 years had an amazing 19.2% prevalence of “bronchitis”. This is an unbelievable number. So, is there an epidemic of COPD among South Californian adolescents? Should the local health authorities and the WHO be alerted? I don’t think so. In reality, the self-reported prevalence of symptoms has been just mispresented as lung disease, but its accuracy is close to zero.
It is sad that the American Thoracic Society, a well-respected health organization, released a press statement with a headline that e-cigarettes may harm lung health when in reality lung health was not assessed and harm was never observed to anyone. Moreover, it is sad that the word “risk” is mentioned 3 times in the results section of the abstract although the study did not assess risk and did not assess any disease prevalence. But, as expected, such headlines “sell” to the media. Unfortunately, the issue of e-cigarettes and public health has been degraded to a media hunt for maximum publicity and a campaign for mis-informing the public and the regulators.