name gateway theory hypothesis
class assumption
category conflation
motivation pretext
science aversion 🟥🟧🟧
used by TI, Bloomberg/WHO, CTFK et al.

The gateway hypothesis is easily debunked. By you know, just casually glancing at any usage statistic. Smoking goes down, whenever vaping goes up. Direct substitutes.

Claims to the contrary rest on misquoting studies and conflating individual effects for a generalizable thing. On the population level (and in particular on teenagers), vaping is observably a gateway out of smoking. (Which doesn't mean it should be encouraged though). The general gateway theorizing largely stem from prior claims to such effect (also long debunked), or the visual misdiagnosis.

individual gateway: smoking "upgrade"

Adolescents trying adult things is kinda their thing. And attention seekers can be predisposed to try vaping. On an individual level they might also start smoking (later or beforehand). Not because of the claimed heroin-like effects, but because nicotine e-cigarettes comparatively boring. Any migration to smoking is more plausibly explained by the lackluster intoxication and fairly delayed effects (in comparison to tobacco).

Albeit the flavour degration for any substance use "upgrade" from e-cigs to tobacco poses a marginal deterrent still. (Even from mint vapes to menthol cigarettes). -- Btw, the simpleminded conjecture flavours cause vaping, and then vaping leads to smoking is just that: simpleminded.

population gateway: out of smoking

Individual-level gateway effects are rare, else we would see an according effect at the population level. But smoking rates are declining even for teenagers. More even, an accelerated decline in teen smoking was particular during the "teen vaping epidemic" in the US. (Albeit even the frequent-vaping uptake can't explain all of it. Can't draw many conclusions from it either way.)

population prevalence → renormalization effect

"Renormalization" is a common pessimism to insinuate rising smoking rates. Which of course, they aren't. It's largely based on the visual misdiagnosis that "vaping looks like smoking". Which is essentially a technophobic mix-up - that teenagers aren't as prone to as policy makers.

Counter point:

As described in the paper, while there has been a considerable uptake of vaping by youth in the United States, smoking prevalence has decreased, suggesting no overall gateway effect. However, scientists warned that the recent phenomenon of the use of e-cigarettes by youth cannot dismiss the chance that if youth prevalence of e-cigarette use increases and is sustained over time, the result might be an increase in the prevalence of combustible tobacco use.

Albeit there hasn't yet been the proclaimed epidemic of teen or general vaping. But if there was, the argument might hold some clout. It's not likely that 20% of the population (= all current smokers) would switch at once to vaping. But hypothetically, if, then it would drag along social acceptance. If not smoking per se. (But I guess that explains the thought process behind "renormalization".)