There was recently a study conducted among high school students in Hawaii, and it seemed to conclude that vaping leads to smoking. But are the results reliable?
Let's take a look!
Wills et al, 2015
This study was a longitudinal study, which is where they take data from a set group of subjects at set time intervals.
In this case, they surveyed 2,338 students aged between 14-15 at one point in time then followed them up one year later. They examined whether they vaped and/or smoked cigarettes, along with whether their parents influenced their behaviour, how rebellious they were and how disposed towards cheap thrills they were (psychosocial covariates for short).
The researchers busted out a questionnaire designed to answer the following questions:
1. What their family support structure was like (single parent, step-parent, biological parents, extended family).
2. How educated their parents were
3. Have they ever heard of e-cigarettes
4. Do they think e-cigs are healthier than tobacco
5. How often they vape
6. How often they smoke
They also hit them with the old "how true is this statement on a scale" type of questions, with answers ranging from "not at all true for me" to "very true for me" on either a five- or seven-point scale. The statements included:
1. "When I feel bad about something, my parent will listen"
2. "My parent knows where I am after school"
3. "I like to do dangerous things for fun"
4. "I like to break the rules"
The surveys were completed once, then again one year later.
Both cigarette and e-cigarette use increased over the space of the year.
The authors of the paper determined that (to paraphrase) for whatever reason, e-cigarette use seemed to act as a gateway to cigarette use.
Are these results reliable?
It may come as no surprise to you that I have a few bones to pick with this study.
The survey says...?
The study did not publish the full results of the survey, which is fair enough. You can't cram all of that paperwork into a six-page PDF.
But they also didn't publish all of the questions they asked. The questions I wrote above were a sample of the kinds of questions they asked. So we can't know if all the questions were appropriate for collecting the data they required. We also can't know if any of the questions were liable to skew the results at all.
OK, fair enough, I am nitpicking here, but still...
The gateway hypothesis...?
Is the gateway hypothesis based on fact? Is it such an established theory that it can be taken as a given that it exists?
No, don't panic, I'm not muddying my boots in these particular waters. What I am concerned about, though, is that this paper seems to use the gateway hypothesis to make the results seem self-evident.
"The reasons for the effect of e-cigarette use on transition to smoking remain to be clarified, but plausible hypotheses have been suggested" (Wills et al, 2015).
So... your data shows that both e-cigarette use and tobacco use increased, but doesn't explain how or why? OK, so your data basically shows... nothing.
We all know that teenagers are rebellious and thrill-seeking. We all know that a number of them experiment with tobacco and other drugs at young ages. There's a lot of noise in there; did throwing e-cigs into the mix really have an effect?
Correlation equals causation...?
I must admit I had a little chuckle when I was reading the results. To paraphrase (heavily), it went a little something like this:
"OK, so, kids who used e-cigs also started smoking. We don't know why, but here are some really good ideas..."
Essentially, all this paper did was show a correlation between kids who started using e-cigs and kids who started smoking. And for those playing at home...
Correlation does NOT imply causation.
To make this easier to understand, here are some shocking correlations:
Thanks to tylervigen.com for these. You can also make your own spurious correlations here.
You get the idea.
This is all very well and good, Lloyd, but what's the takeaway here? How is the knowledge of this one particular study supposed to help me?
I'm glad you asked!
First and most importantly, you may have recently heard that there is movement in the SA legislature to regulate e-cigs and the like under the same laws as tobacco. One of the common cries of adherents to this movement is that e-cigarette use is linked to increased tobacco use amongst minors. You may even hear this study cited at you because of its recency. Well, now you can say "actually, NO, that study did not show a causative link, and failed to take into account the tendency of teenagers to try whatever is going around at the time, and wasn't transparent in publishing its conclusions, so NER!"
Secondly, being the skeptic that I am, I am always keen to encourage anyone who will listen to me to take a skeptical look at things. And with such a politically charged issue such as the usage of e-cigarettes, you need to be on your guard and take every bit of evidence, both for and against your position, with a grain of salt. This study, as is the case with a lot that I've seen, seemed biased from the outset with statements such as:
"...it is important to have evidence from different settings on the relation between ecigarette [sic] use and smoking" (Wills et al, 2015).
This is a case of begging the question - "we know that e-cig use leads to smoking, we just need some more evidence that shows it". Sorry, but this study fails to represent evidence. You may as well just submit it to tylervigen.com for a laugh.
There is a great article on Forbes that basically says the same thing as I just have, but in far fewer and more professional-sounding words. Check it out here.
Hastily mashed out by Sam Lloyd.