It sounds like the cancer rates really are dramatically lower, and there really is very little spit involved. Even the oral cancer rates drop down to near baserate levels (lazily averaging studies here), and the pancreatic cancer rates, as noted by someone above, are over-stated*. I do note, however, that the Swedes really are making a concerted effort to minimize carcinogens. Somehow, the record of American tobacco companies isn't too reassuring in this regard, however, which limits the degree to which we can generalize the results of the Swedish study to the US.

So I'm thinking this would be a tremendous mistake, unless there is accompanying regulation to emulate the carcinogen-reduction strategies of the Swedes, because it might lull people into an unwarranted sense of security regarding the safety of this product.

Researchers used to speculate more often about the relative elevation of cancer risk due to smoking in the US vs. smoking in other countries (the effect sizes were much smaller - don't know if they still are). Production was cited frequently as a potential culprit, but the population differences (baseline health, mortality, nutrition etc.) makes the comparison difficult. Even if the Swedish study turns out to be a good indicator of potential risk, I wouldn't be at all surprised to see higher mortality rates in the US, purely as a function of differences in medical primary care between good old Socialist Sweden and the US, where early diagnosis is somewhat less likely to happen [hurray for the free market].

My 2-cents.

*Using a description like "double the risk" can be terribly misleading, of course, if the probability within a population is already very low. The description works as well for a change from once incident to two in a sample of 10,000 as it does for a change of 100 to 200. It sounds like the risk of pancreatic cancer probably has been over-hyped. I'd be interested in someone contrasting this increase with increases due to other known risk factors we don't do much about (moderately elevated Radon levels in homes, for example, or the prevalence of lung cancer among auto technicians routinely exposed to brake-lining dust).