Although the survey approach described above would lead to "invalid" results, I would never argue that survey studies should be thrown out and that we should only be conducting RCTs. My point is that there are strengths and limitations of various study designs, and causal inference is enhanced by having research evidence from multiple study designs. On balance, the strengths and weaknesses balance each other out, leading to a much greater ability to draw valid conclusions.

Harris et al., in perhaps the most comprehensive and definitive on this issue, found a clear increase in cancer risk associated with non-filtered cigarettes. As they reported: "Compared with smokers of medium tar (15-21 mg) filter cigarettes, risk was higher among men and women who smoked high tar (> or = 22 mg) non-filter brands (hazard ratio 1.44, 95% confidence interval 1.20 to 1.73, and 1.64, 1.26 to 2.15, respectively)."

The Rest of the Story

Thus, the authors readily acknowledge that it is impossible from this study to determine whether or not e-cigarettes lead to smoking or whether smoking leads to e-cigarette experimentation.

LAST：Hot sale e cigarette TGVenus electronic cigarettes China manufacturer |NEXT：2014 high quality 100% innokin itaste dual coil gladius atomizer rda atomizer dual coil clearomizer with adjustable airflow