Vaping and Heart Attacks

Posted 17th April 2019 by Dave Cross
The American Journal of Preventative Medicine has published a letter by Konstantinos Farsalinos and Raymond Niaura. “E-cigarette Use and Myocardial Infarction: Association Versus Causal Inference” counters that lies put out in a recent Wichita study.

Konstantinos Farsalinos said: “I am sure you remember the story that e-cigarettes increase the risk of heart attacks. I characterised such statements as epidemiological and scientific malpractice, and I strive to prove that [in this letter].”

The authors write: “We read with particular interest the study by Alzahrani et al., who examined the association between e-cigarette use and myocardial infarction (MI). With this letter, we would like to express our concern about the study conclusions.”

The first error they noted was with the selection of the data sets for analysis: “Although annual data sets of the National Health Interview Survey (NHIS) have been released consistently, the authors chose to pool non-consecutive years (2014 and 2016) and did not mention any particular reason for this. According to the National Centre for Health Statistics (NCHS), a new sampling design was implemented in 2016.2 Therefore, 2014 and 2016 fall into different sampling design periods.”

The selective procedure allowed the Wichita researchers to conclude: “E-cig users have higher odds of myocardial infarction, stroke, depression, anxiety, emotional problems, circulatory problems and lower risk of hypertension and diabetes compared to non-E-cigarette users.”

Stanton Glantz embraced the findings and proclaimed: “e-cigarettes are a lot more dangerous than people used to think…e-cigarettes make it harder, not easier, to quit smoking.”

The study and Glantz’ comments were widely lambasted at the time. Michael Siegel commented: “This is yet another example of the junk science that is rapidly being spewed out by anti-tobacco researchers who are apparently more interested in demonizing vaping than in using rigorous scientific reasoning.”

Farsalinos and Niaura point out that it is impossible for the study’s authors to arrive at their stated conclusion: “[It] is a misinterpretation and misrepresentation of the study findings. The ‘increased risk’ claim clearly implies causality and a specific temporal definition of events (i.e., that e-cigarette use precedes MI and e-cigarette use caused the MI). This disagrees with what Alzahrani and colleagues mention in the Limitations section of their article: the study cannot permit identifying causal relationships in part because it is not known when the MIs occurred relative to e-cigarette use. It also violates a basic principle of epidemiology that no causal inference can be derived from any cross-sectional study, such as the NHIS.”

They concluded with a demonstration of how flawed the conclusion is by performing an identical operation to demonstrate that the use of cholesterol medications can also be associated with heart attacks – clearly a nonsense.

 

Resources:

 

 Dave Cross
Article by Dave Cross
Freelance writer, physicist, karateka, dog walker