When you question him about statistical links between cigarettes and lung cancer, for example, he immediately becomes a walking compendium of other peoples sayings: Cancer is a biological, not a statistical, problem. Smoking is one of the leading causes of statistics. Statistics are like a bikini bathing suit: what they reveal is interesting, what they conceal is vital. Repeatedly, Dwyer used his audience (me) as the example to back up his point: If you can decide not to smoke on your own, why not let others do the same? When asked about how independent a decision about smoking could be, given its addictive properties and the industrys massive advertising budget, he replied that there is no addictive effect, and that ads influence only the brand choice of those who already smoke, just as soap ads only talk to consumers about buying Tide instead of Fab, not about whether to wash. This skilled persuasion on the public-relations level is backed up by widespread campaign contributions to Congressional candidates

and her sister, Princess Margaret, also faced the consequences of smoking, undergoing a lung operation in 1985 before dying of strokes in 2002
It is relatively refreshing because it has little bitterness or miscellaneous taste.
Health Educ J 56:199209 Welshman J (1996) Images of youth: the issue of juvenile smoking, 18801914
These are abnormal tissue growths that form in the intestines