A Decade of Data Reveals Heated Tobacco in Japan Was Never Just Cigarette Substitution

When heated tobacco products arrived in Japan in 2014, they triggered one of the fastest product transitions the tobacco market has ever recorded. Cigarette sales, already falling, plummeted further, and public health observers around the world began asking a deceptively simple question: were smokers actually switching, or was the tobacco industry simply widening its net? A new decade-long analysis published in The Lancet Regional Health – Western Pacific now offers the most detailed answer yet, and the findings complicate the tidy substitution narrative that has dominated the debate.

Researchers Satomi Odani and Takahiro Tabuchi drew on the Japan Society and New Tobacco Internet Survey, known as JASTIS, an internet panel study that has tracked tobacco behaviour annually since 2015. Their analysis combined two complementary approaches: repeated cross-sectional estimates of prevalence across 161,583 annual observations from 58,751 unique respondents, and longitudinal modelling of 93,540 consecutive-year transition pairs from 35,479 respondents. By classifying each participant into one of five mutually exclusive states — Never, Non-current, Cigarette-only, HTP-only, or Dual use — the team could trace not just how many people used heated tobacco, but how they arrived there and how long they stayed.

The headline trend is striking. Weighted prevalence of current heated tobacco use, defined as use on at least one day in the past 30 days, climbed from just 0.1 percent in 2015 to 14.9 percent in 2025. But the trajectory was anything but linear. Segmented statistical models identified 2018 as a decisive turning point: before that year, standardised prevalence grew at an annual rate of 283.4 percent, an absolute gain of 3.33 percentage points per year; after 2018, growth slowed to 5.7 percent per year, or 0.92 percentage points annually. The explosive early phase gave way to slower but stubbornly persistent expansion.

The pattern differed sharply across groups. Prevalence remained consistently higher among men than women, reaching 22.2 percent versus 7.7 percent in 2025, and it was highest among current cigarette smokers, among whom heated tobacco use hit 43.7 percent by 2025. Former smokers followed at 18.4 percent, while never smokers stayed far lower at just 1.4 percent. That gradient matters: it suggests heated tobacco has not swept through the entire population, but the fact that uptake continued among former and never smokers signals diffusion beyond the smoking population that substitution arguments would predict.

The longitudinal data reveal what was really driving growth. Among cigarette-only users, the probability of remaining in that state fell from 76.8 percent in 2016–2018 to 64.4 percent in 2019–2025, while transitions to Non-current use rose from 13.5 to 22.0 percent. Yet direct conversion to HTP-only use stayed low, at 2.1 percent before 2018 and 2.7 percent after, and movement to dual use rose only modestly from 7.6 to 11.0 percent. Crucially, these cigarette-to-HTP pathways stopped accelerating after 2018, undermining the idea that continued market growth reflected ever-faster switching.

What changed instead was persistence. Once people adopted heated tobacco, they increasingly stayed. The probability of remaining HTP-only from one year to the next jumped from 43.1 percent in the early period to 67.9 percent after 2018, while persistence in dual use rose from 44.9 to 57.0 percent. Corresponding exits into Non-current use fell dramatically, from 36.4 to 15.5 percent among HTP-only users and from 26.0 to 15.3 percent among dual users. In other words, the later market was sustained less by new converts from cigarettes and more by users who simply did not leave.

To quantify these dynamics formally, the researchers applied a Markov decomposition, reconstructing year-to-year changes in heated tobacco prevalence and splitting each annual change into four pathways: uptake from Never or Non-current states, switching from cigarettes to heated tobacco, switching back to cigarettes, and exit to Non-current use. The results are revealing. Before 2018, cigarette switching contributed more to growth than uptake from Never or Non-current states, with mean annual contributions of 1.43 versus 0.73 percentage points. After 2018, the picture inverted: the contribution from cigarette switching, though still averaging 1.56 percentage points, declined over time, while uptake from Non-current states surged from 0.34 to 1.40 percentage points annually. Even never smokers, whose individual-level transitions remained rare at 0.6 to 1.3 percent per year, made a small but persistent positive contribution.

Dual use emerged as a central finding with uncomfortable implications. Throughout the entire decade, dual use remained more prevalent than HTP-only use, reaching 8.4 percent versus 6.5 percent in 2025. This means a substantial share of heated tobacco consumers never abandoned cigarettes at all. The researchers are careful to note that this does not negate genuine declines in cigarette smoking or complete switching among some individuals — cigarette-only use did fall from 19.9 to 10.9 percent over the study period. But it does mean that population-level expansion of heated tobacco cannot be read as a proxy for harm reduction, because the product landscape expanded rather than simply shifted.

Context matters for interpreting these dynamics. Japan regulated heated tobacco differently from combustible cigarettes: they initially carried a lower tax burden, and the revised Health Promotion Act permitted their use in designated heated-tobacco-only rooms. The 2018 inflection point coincided with changes in taxation, smoke-free regulation, and market maturation, though the study design cannot attribute the shift to any single policy. The World Health Organization maintains that heated tobacco products are tobacco products and that the Framework Convention on Tobacco Control applies to them, while cautioning against unproven harm-reduction claims. Japan’s experience now stands as a cautionary case study for Western Pacific nations weighing their own regulatory approaches, from permissive South Korea to restrictive Australia.

The authors acknowledge limitations. JASTIS is an internet panel and may carry selection bias, so temporal patterns deserve more weight than exact prevalence figures. Longitudinal analyses were unweighted and conditional on consecutive-wave participation, and retained respondents skewed older and more male than those lost to follow-up. One-year transitions cannot capture multi-year trajectories, relapse, or whether dual use is temporary. Still, sensitivity analyses adjusting for time in the sample and restricting to adults aged 25 to 69 produced materially similar results. The core message survives scrutiny: heated tobacco growth in Japan was sustained by persistence, uptake from non-current users, and incomplete switching — not by cigarette substitution alone. For regulators everywhere, that distinction may prove decisive.

Subject of Research: Long-term trends and transition dynamics of heated tobacco product use in Japan from 2015 to 2025

Article Title: Trends and transition dynamics of heated tobacco product use in Japan, 2015–2025: a repeated cross-sectional and longitudinal study

Article References: Odani, S., & Tabuchi, T. (2026). Trends and transition dynamics of heated tobacco product use in Japan, 2015–2025: a repeated cross-sectional and longitudinal study. The Lancet Regional Health – Western Pacific, 75, Article 101990. https://doi.org/10.1016/j.lanwpc.2026.101990

Image Credits: AI Generated

DOI: 10.1016/j.lanwpc.2026.101990

Keywords: heated tobacco products, Japan, IQOS, tobacco control, dual use, smoking cessation, JASTIS survey, Markov decomposition, tobacco prevalence, public health, nicotine, Western Pacific