OK, I read your commentary. I also read the presentation that Fould's gave at Penn State. I went into the PATH study wave 1 and looked at the questionnaire concerning e cigarettes. I did not look at the raw data, and I didn't count up all the question on e cigs from page 66 to 90 in the survey.
I've got a few problems with interpreting this stuff. First, retrospective surveys are limited in their utility to the questions asked. Now, they did ask all sorts of questions of the e cig users. There is only one question that asks the single most important issue :
"[Do | Did] you use e-cigarettes as an alternative to quitting tobacco altogether?
1 Yes
2 No
-8 DON’T KNOW
-7 REFUSED"
this is from page 90 of DS1001 Wave 1 Adult Data questionnaire. There are lots of other questions looking at those surveyed for what appeals to you, what doesn't, how much do you smoke, how often do you feel the need to smoke, and all sorts of questions. But what Fould never provides (and I'm not even sure if it can be extracted based on how this thing was set up) is: How many people who dual use (that big 87% group) tried or used or chose as a method, to stop smoking. If most of those users never planned to stop, and were only using it to fill in as a stopgap such as being somewhere cigarettes could not be used, purchased on a whim at the convenience store, or whatever OTHER than "I bought this specifically to stop smoking and I tried real hard to not smoke or cut down smoking while using an e cig" then that 87% figure is useless.
Dual use is not being documented solely among those who view e cigs ONLY as a smoking cessation aid/device. That dual use group identified provided no specific evidence that their e cig use was to stop smoking. Now, maybe the data is buried somewhere in there, but what I would want to know is this:
In the 13% only e cig users, what percentage used e cigs to achieve the goal of smoking cessation (I'd bet a pretty high percentage), and in the 87% dual users the same question (I'd bet lower than the e cig group). Without that all you have is selection bias that could have easily been avoided, and loaded the deck to miss that basic question. It is in fact the very first question required, as it would split the group and assure that things like dual use, e cig preferences, and satisfaction apply to the correct population, not just anyone deciding to mess around with an e cig without a strong motivation to quit smoking.