Correspondence (letter to the editor): No Solid Evidence Base

The authors of the cohort study (1) asked 10^th grade students how regularly they smoked conventional cigarettes or e-cigarettes. However, Morgenstern et al. did not publish the relevant data. Instead they reported whether the youngsters had “ever smoked,” that is, if once in their lives they had tried a few puffs of a cigarette or e-cigarette. The authors themselves imply that ever smoking does not constitute a clinically or health relevant parameter. To obviate the conclusion that this might also mean that their study could be irrelevant, Morgenstern et al. switch from statistics to semantics. Phrases such as “initiation of daily tobacco use” or “onset of smoking” imply more frequent follow-up consumption than is supported by the published data.

Although the authors concede on several occasions that they were unable to draw causal conclusions, elsewhere they claim that the adolescents had been “animated” by the use of e-cigarettes to use conventional cigarettes. Assumptions of causality would require to also investigate the opposite question—namely, how many adolescents had been animated after using e-cigarettes not to smoke conventional cigarettes. The authors could have calculated this on the basis of their data, but they did not do so (or have not published their findings).

Similar reservations (2– 4) have been expressed about cohort studies in the past, which Morgenstern et al. cite in their article, and whose study design informed their own study. The blatant weakness of these studies lay in deducing from adolescents’ willingness to experiment that they entered into long-term cigarette consumption, and to not even consider possible exit effects. The succession of a long list of one-sided, speculative publications obviously extends the publication list of the participating authors, but it does not amount to a solid evidence base on which to balance the benefits and risks of e-cigarettes.

References