Smoking and vaping

Smoking is behavioural science's clearest success and its largest remaining question. Adult prevalence in the United Kingdom, the United States and most of Western Europe is at a record low, around one in ten, having halved since the millennium; the deaths follow with a lag of two to three decades, which is why tobacco still accounts for roughly one death in eight to ten in high-income countries and why lung cancer mortality is still falling in men and only now turning in women. The remaining smokers are concentrated: by deprivation, by mental illness, and in the countries and regions where prevalence fell least, so that smoking is now among the largest single causes of the gap in life expectancy between rich and poor.

Vaping is the thing that could move the trajectory either way. As a route out of smoking it has the best evidence of any cessation aid in the randomised trials, and England has made it a plank of policy; as a route into nicotine it has spread fast among people who never smoked, with regular use among adolescents reaching around one in ten in several countries before the first restrictions on flavours and disposables. Its long-term harm is unknown because it is too new to be known, and the trials that would settle whether adolescent vaping leads to smoking are impossible to run. Heated tobacco has taken a large share of the Japanese market, and nicotine pouches are rising in Sweden and the United States; each shifts harm without removing it, by amounts the evidence does not yet give.

The policy frontier is the generational sales ban — no tobacco sale to anyone born after a chosen year — legislated in the United Kingdom, repealed before it took effect in New Zealand, and proposed elsewhere. It has no track record at all. The measures that do have one — price, plain packaging, smoke-free places, advertising bans — have effects that are well measured and mostly banked; their evaluations file under What changes behaviour at scale.

Status

established

Direction

mixed

Would change our view

Health outcomes in the first cohorts covered by a generational ban; long-term outcome data on vaping, especially among never-smokers; or evidence that the fall in prevalence has stalled in the most deprived groups.

Trackers

Gaps

Measured rather than self-reported prevalence; consistent vaping measures across countries and years; anything on heated tobacco beyond Japan; the mental-health population, where prevalence is highest and least studied.