Causal Inference? It’s Only E-cigarettes, So Let’s do it Anyway
“[Causal inference] is the process of drawing a conclusion about a causal connection based on the conditions of the occurrence of an effect” with the word causal meaning “the agency or efficacy that connects one process (the cause) with another (the effect).”
So using causal inference, you attach one process to another.
In this case we are talking about two e-cigarette studies, which looked at advertising and teen uptake of vaping.
Both stated in their write up that causal inference could not be used.
Study #1: the analyses were cross-sectional which prohibits causal inference.
Study #2: data were cross-sectional, and thus, causal relationships between e-cigarette advertising and use cannot be made.
But they went ahead and did it anyway.
Writing about this Professor Siegel highlights the very obvious bias that has crept (or stormed?) into these studies. A University in Texas produced one of the studies; the other was by the CDC. Both were published in highly respected journals, and both studies used causal inference.
The press then stated that advertising increased teen vaping.
Study #1: This study highlights widespread environmental influences promoting e-cigarette use through a variety of platforms, and that these influences increase the odds that a young person might also be using e-cigarettes.
Study #2: These findings suggest that comprehensive tobacco prevention and control strategies, including efforts to reduce youth exposure to advertising, are critical to prevent all forms of tobacco use among US youth, including e-cigarettes.
The quotes are taken from The Rest of the Story, a tobacco news website, and show quite clearly the causal inference that “couldn’t be used”.
In the same article that the quotes are taken from, Professor Siegel then asks this very important question,
“The question that naturally arises is what is the point of this research? If the researchers were going to conclude that e-cigarette marketing influences e-cigarette use anyway, then why do the research? The apparent drawing of a priori conclusions essentially negates the purpose of conducting research in the first place. Both studies mention the need to conduct longitudinal research to determine the causal direction. But why is that research needed if the causal conclusions have already been drawn?”
We, our company and our suppliers own the copyright, trademarks and all other intellectual property rights in all material and content on this website, which you may use, download and copy only for your own personal, non-commercial use. Any other use or reproduction of the material or content is strictly prohibited. You may not create any link to this website without our prior written consent, nor may you restrict or inhibit the use or enjoyment of it by anyone else.
WARNING: The Electronic Cigarette Company makes no claims that the electronic cigarette will cure a smokers addiction to nicotine; the electronic cigarettes it sells serves the same purpose as a tobacco cigarette- it delivers its user nicotine. If you do suffer from the disease of Tobacco/Nicotine Dependence Syndrome and want to take steps to give up smoking or cut down the quantity of cigarettes you currently smoke. We recommend you visit your health care provider to discuss NRT (Nicotine Replacement Therapy) or a tobacco harm reduction programme.
Please Note: We are not a pharmaceutical company and we do not produce medical products.