Heated tobacco products for smoking cessation and reducing smoking prevalence ( Review )

Background Heated tobacco products (HTPs) are designed to heat tobacco to a high enough temperature to release aerosol, without burning it or producing smoke. They diFer from e-cigarettes because they heat tobacco leaf/sheet rather than a liquid. Companies who make HTPs claim they produce fewer harmful chemicals than conventional cigarettes. Some people report stopping smoking cigarettes entirely by switching to using HTPs, so clinicians need to know whether they are eFective for this purpose and relatively safe. Also, to regulate HTPs appropriately, policymakers should understand their impact on health and on cigarette smoking prevalence. Objectives To evaluate the eFectiveness and safety of HTPs for smoking cessation and the impact of HTPs on smoking prevalence. Search methods We searched the Cochrane Tobacco Addiction Group's Specialised Register, CENTRAL, MEDLINE, and six other databases for relevant records to January 2021, together with reference-checking and contact with study authors and relevant groups. Selection criteria We included randomised controlled trials (RCTs) in which people who smoked cigarettes were randomised to switch to exclusive HTP use or a control condition. Eligible outcomes were smoking cessation, adverse events, and selected biomarkers.  RCTs conducted in clinic or in an ambulatory setting were deemed eligible when assessing safety, including those randomising participants to exclusively use HTPs, smoke cigarettes, or attempt abstinence from all tobacco. Time-series studies were also eligible for inclusion if they examined the populationlevel impact of heated tobacco on smoking prevalence or cigarette sales as an indirect measure. Data collection and analysis We followed standard Cochrane methods for screening and data extraction. Our primary outcome measures were abstinence from smoking at the longest follow-up point available, adverse events, serious adverse events, and changes in smoking prevalence or cigarette sales. Other outcomes included biomarkers of harm and exposure to toxicants/carcinogens (e.g. NNAL and carboxyhaemoglobin (COHb)). We used a random-eFects Mantel-Haenszel model to calculate risk ratios (RR) with 95% confidence intervals (CIs) for dichotomous outcomes. Heated tobacco products for smoking cessation and reducing smoking prevalence (Review) Copyright © 2022 The Authors. Cochrane Database of Systematic Reviews published by John Wiley & Sons, Ltd. on behalf of The Cochrane Collaboration. 1 Cochrane Library Trusted evidence. Informed decisions. Better health. Cochrane Database of Systematic Reviews For continuous outcomes, we calculated mean diFerences on the log-transformed scale (LMD) with 95% CIs. We pooled data across studies using meta-analysis where possible. Main results We included 13 completed studies, of which 11 were RCTs assessing safety (2666 participants) and two were time-series studies. We judged eight RCTs to be at unclear risk of bias and three at high risk. All RCTs were funded by tobacco companies. Median length of follow-up was 13 weeks. No studies reported smoking cessation outcomes. There was insuFicient evidence for a diFerence in risk of adverse events between smokers randomised to switch to heated tobacco or continue smoking cigarettes, limited by imprecision and risk of bias (RR 1.03, 95% CI 0.92 to 1.15; I2 = 0%; 6 studies, 1713 participants). There was insuFicient evidence to determine whether risk of serious adverse events diFered between groups due to very serious imprecision and risk of bias (RR 0.79, 95% CI 0.33 to 1.94; I2 = 0%; 4 studies, 1472 participants). There was moderate-certainty evidence for lower NNAL and COHb at follow-up in heated tobacco than cigarette smoking groups, limited by risk of bias (NNAL: LMD −0.81, 95% CI −1.07 to −0.55; I2 = 92%; 10 studies, 1959 participants; COHb: LMD −0.74, 95% CI −0.92 to −0.52; I2 = 96%; 9 studies, 1807 participants). Evidence for additional biomarkers of exposure are reported in the main body of the review. There was insuFicient evidence for a diFerence in risk of adverse events in smokers randomised to switch to heated tobacco or attempt abstinence from all tobacco, limited by risk of bias and imprecision (RR 1.12, 95% CI 0.86 to 1.46; I2 = 0%; 2 studies, 237 participants). Five studies reported that no serious adverse events occurred in either group (533 participants). There was moderate-certainty evidence, limited by risk of bias, that urine concentrations of NNAL at follow-up were higher in the heated tobacco use compared with abstinence group (LMD 0.50, 95% CI 0.34 to 0.66; I2 = 0%; 5 studies, 382 participants). In addition, there was very low-certainty evidence, limited by risk of bias, inconsistency, and imprecision, for higher COHb in the heated tobacco use compared with abstinence group for intention-to-treat analyses (LMD 0.69, 95% CI 0.07 to 1.31; 3 studies, 212 participants), but lower COHb in per-protocol analyses (LMD −0.32, 95% CI −1.04 to 0.39; 2 studies, 170 participants). Evidence concerning additional biomarkers is reported in the main body of the review. Data from two time-series studies showed that the rate of decline in cigarette sales accelerated following the introduction of heated tobacco to market in Japan. This evidence was of very low-certainty as there was risk of bias, including possible confounding, and cigarette sales are an indirect measure of smoking prevalence. Authors' conclusions No studies reported on cigarette smoking cessation, so the eFectiveness of heated tobacco for this purpose remains uncertain. There was insuFicient evidence for diFerences in risk of adverse or serious adverse events between people randomised to switch to heated tobacco, smoke cigarettes, or attempt tobacco abstinence in the short-term. There was moderate-certainty evidence that heated tobacco users have lower exposure to toxicants/carcinogens than cigarette smokers and very lowto moderate-certainty evidence of higher exposure than those attempting abstinence from  all tobacco. Independently funded research on the eFectiveness and safety of HTPs is needed. The rate of decline in cigarette sales accelerated aOer the introduction of heated tobacco to market in Japan but, as data were observational, it is possible other factors caused these changes. Moreover, falls in cigarette sales may not translate to declining smoking prevalence, and changes in Japan may not generalise elsewhere. To clarify the impact of rising heated tobacco use on smoking prevalence, there is a need for time-series studies that examine this association. P L A I N   L A N G U A G E   S U M M A R Y Do heated tobacco products help people to quit smoking, are they safe for this purpose, and have they led to falls in smoking rates? Key messages Heated tobacco probably exposes people to fewer toxins than cigarettes, but possibly more than not using any tobacco. Falls in cigarette sales appeared to speed up following the launch of heated tobacco in Japan, but we are uncertain whether this is caused by people switching from cigarettes to heated tobacco. We need more independently funded research into whether heated tobacco helps people stop smoking, whether it results in unwanted eFects, and the impact of rising heated tobacco use on smoking rates. What are heated tobacco products? Heated tobacco products are designed to heat tobacco to a high enough temperature to release vapour, without burning it or producing smoke. They diFer from e-cigarettes because they heat tobacco leaf/sheet rather than a liquid. Many of the harmful chemicals in cigarette smoke are created by burning tobacco. So heating not burning tobacco could reduce the amount of chemicals a user ingests. Some people report stopping smoking cigarettes entirely by switching to using heated tobacco. Heated tobacco products for smoking cessation and reducing smoking prevalence (Review) Copyright © 2022 The Authors. Cochrane Database of Systematic Reviews published by John Wiley & Sons, Ltd. on behalf of The Cochrane Collaboration. 2 Cochrane Library Trusted evidence. Informed decisions. Better health. Cochrane Database of Systematic Reviews Why we did this Cochrane Review Because cigarette smoking is addictive, many people find it diFicult to stop despite the harm it causes. We aimed to find out whether trying to switch to heated tobacco helps people stop smoking cigarettes, and whether it results in unwanted eFects. We also wanted to find out whether rising heated tobacco use has aFected smoking rates or cigarette sales. What did we do? We looked for studies that reported on the use of heated tobacco for stopping smoking, and on unwanted eFects and toxin exposure in people asked to use heated tobacco. Here we only included randomised controlled trials, where treatments were given to people at random. This type of study is considered the most reliable way of determining if a treatment works. Finally, we searched for studies looking at changes in smoking rates and cigarette sales following the launch of heated tobacco to market. We included studies published up to January 2021. What we found Our search found 13 relevant studies. No studies reported whether heated tobacco helps people stop smoking cigarettes. Eleven trials, all funded by tobacco companies and with 2666 adult smokers, compared unwanted eFects and toxin levels in people randomly assigned to use heated tobacco or to continue smoking cigarettes or abstain from tobacco use. Two studies looked at how trends in cigarette sales changed following the launch of heated tobacco in Japan. What are the results of our review? We do not know whether using heated tobacco helps people to stop smoking cigarettes (no studies measured this). We are uncertain whether the chances of getting unwanted symptoms from being asked to use heated tobacco are diFerent compared with cigarettes (6 studies, 1713 participants) or no tobacco (2 studies, 237 participants). Serious unwanted symptoms in the short time period studied (average 13 weeks) were rare in all groups, which means we are uncertain about any diFerences. Toxin levels were probably lower in people using heated tobacco than those smoking cigarettes (10 studies, 1959 participants), but may be higher than in people not using any tobacco products (5 studies, 382 participants). The launch of heated tobacco products in Japan may have caused the decline in cigarette sales to speed up over time (two studies), but it is unclear whether the fall in the percentage of people who smoke also sped up because no studies looked at this. How reliable are these results? Results are based on data from a small number of studies, most of which were funded by tobacco companies. Results on unwanted eFects are likely to change as more evidence becomes available. However, we are moderately confident that levels of measured toxins are lower in people using heated tobacco than smoking cigarettes, but less confident that levels were higher than in people not using any tobacco. We are also less confident that the launch of heated tobacco caused the fall in cigarette sales to speed up, as results came from a single country. Heated tobacco products for smoking cessation and reducing smoking prevalence (Review) Copyright © 2022 The Authors. Cochrane Database of Systematic Reviews published by John Wiley & Sons, Ltd. on behalf of The Cochrane Collaboration. 3 H eted toacco prducts or sm oing ceation nd reucing sm oing prealence (Rview ) Coyright © 2022 he Auhors. Cohrane Daabase of Sstem tic Rview s pulished by John W ley & Sns, Ltd. on bhalf of he Cohrane Colloration. 4 S U M M A R Y   O F   F I N D I N G S Summary of findings 1.   Heated tobacco use compared with cigarette smoking Heated tobacco use compared with cigarette smoking Patient or population: people who smoke Setting: USA, Japan, UK, South Korea, Poland Intervention: heated tobacco use Comparison: cigarette smoking Anticipated absolute effects* (95% CI) Outcomes Risk with cigarette smoking Risk with heated tobacco use Relative effect (95% CI) No of participants (studies) Certainty of the evidence (GRADE) Comments Study population Adverse events – measured by self-report 235 per 1000 242 per 1000 (216 to 270) RR 1.03 (0.92 to 1.15) 1713 (6 RCTs) ⊕⊕⊝⊝ Lowa,b — Study population Serious adverse events – measured by self-report and medical records 13 per 1000 10 per 1000 (4 to 24) RR 0.79 (0.33 to 1.94) 2009 (9 RCTs) ⊕⊝⊝⊝ Very lowa,c — NNAL at follow-up – measured in urine — — LMD 0.81 lower (1.07 lower to 0.55 lower) 1959 (10 RCTs) ⊕⊕⊕⊝ Moderatea LMD has no units as it is calculated from the logarithm of biomarker measurements. COHb at follow-up – measured in blood — — LMD 0.74 lower (0.92 lower to 0.52 lower) 1807 (9 RCTs) ⊕⊕⊕⊝ Moderatea LMD has no units as it is calculated from the logarithm of biomarker measurements. *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; COHb: carboxyhaemoglobin; LMD: difference in means of log-transformed measurements; NNAL: 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol; RCT: randomised controlled trial; RR: risk ratio. GRADE Working Group grades of evidence Cohrane Lrary Tsted evence. Iform ed desions. Btter hlth. Cohrane Daabase of Sstem tic Rview s H eted toacco prducts or sm oing ceation nd reucing sm oing prealence (Rview ) Coyright © 2022 he Auhors. Cohrane Daabase of Sstem tic Rview s pulished by John W ley & Sns, Ltd. on bhalf of he Cohrane Colloration. 5 High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. aDowngraded one level for risk of bias: all studies were at either unclear or high risk of bias. bDowngraded one level for imprecision: confidence intervals contain clinically meaningful benefit and clinically meaningful harm. cDowngraded two levels for imprecision: confidence intervals contain large clinically meaningful benefit and clinically meaningful harm. Summary of findings 2.   Heated tobacco use compared with abstinence from tobacco Heated tobacco use compared with abstinence from tobacco Patient or population: people who smoke Setting: USA, Japan, UK, South Korea Intervention: heated tobacco use Comparison: abstinence from tobacco Anticipated absolute effects* (95% CI) Outcomes Risk with abstinence from tobacco Risk with heated tobacco use Relative effect (95% CI) No of participants (studies) Certainty of the evidence (GRADE) Comments Smoking cessation – not measured — — — — — — Study population Adverse events – measured by self-report 468 per 1000 525 per 1000 (403 to 684) RR 1.12 (0.86 to 1.46) 237 (2 RCTs) ⊕⊝⊝⊝ Very lowa,b — Study population Serious adverse events – measured by self-report and medical records See comment See comment — 533 (5 RCTs) ⊕⊝⊝⊝ Very lowc,d No serious adverse events were reported. NNAL at follow-up – measured in urine — — LMD 0.50 higher (0.34 higher to 0.66 higher) 382 (5 RCTs) ⊕⊕⊕⊝ Moderated LMD has no units as it is calculated from the logarithm of biomarker measurements. Cohrane Lrary Tsted evence. Iform ed desions. Btter hlth. Cohrane Daabase of Sstem tic Rview s H eted toacco prducts or sm oing ceation nd reucing sm oing prealence (Rview ) Coyright © 2022 he Auhors. Cohrane Daabase of Sstem tic Rview s pulished by John W ley & Sns, Ltd. on bhalf of he Cohrane Colloration. 6 COHb at follow-up – measured in blood LMD 0.69 higher (0.07 higher to 1.31 higher) for analyses using intention-to-treat, but LMD 0.32 lower (1.04 lower to 0.39 higher) for per-protocol analyses. 212 (3 RCTs) ⊕⊝⊝⊝ Very lowa,d,e Reported narratively due to inconsistency of results across subgroups. *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; COHb: carboxyhaemoglobin; LMD: difference in means of log-transformed measurements; NNAL: 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol; RCT: randomised controlled trial; RR: risk ratio. GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. aDowngraded one level for imprecision: confidence intervals contained clinically meaningful benefit and clinically meaningful harm. bDowngraded two levels for risk of bias: all studies were considered at high risk of bias. cDowngraded two levels for imprecision: no serious adverse events occurred so confidence intervals could not be calculated. dDowngraded one level for risk of bias: two of the five studies were considered high risk of bias, while three had uncertain risk of bias. eDowngraded two levels for inconsistency: there was unexplained heterogeneity and results were inconsistent across subgroups and sensitivity analyses. Summary of findings 3.   Heated tobacco use compared with snus use Heated tobacco use compared with snus use Patient or population: people who smoke Setting: USA Intervention: heated tobacco use Comparison: snus use Anticipated absolute effects* (95% CI) Outcomes Risk with snus use Risk with heated tobacco use Relative effect (95% CI) No of participants (studies) Certainty of the evidence (GRADE) Comments Smoking cessation – not measured — — — — — — Study population Adverse events – measured by selfreport 558 per 1000 726 per 1000 RR 1.30 (0.94 to 1.80) 87 (1 RCT) ⊕⊝⊝⊝ Very lowa,b — Cohrane Lrary Tsted evence. Iform ed desions. Btter hlth. Cohrane Daabase of Sstem tic Rview s H eted toacco prducts or sm oing ceation nd reucing sm oing prealence (Rview ) Coyright © 2022 he Auhors. Cohrane Daabase of Sstem tic Rview s pulished by John W ley & Sns, Ltd. on bhalf of he Cohrane Colloration.


A B S T R A C T Background
Heated tobacco products (HTPs) are designed to heat tobacco to a high enough temperature to release aerosol, without burning it or producing smoke. They di er from e-cigarettes because they heat tobacco leaf/sheet rather than a liquid. Companies who make HTPs claim they produce fewer harmful chemicals than conventional cigarettes. Some people report stopping smoking cigarettes entirely by switching to using HTPs, so clinicians need to know whether they are e ective for this purpose and relatively safe. Also, to regulate HTPs appropriately, policymakers should understand their impact on health and on cigarette smoking prevalence.

Objectives
To evaluate the e ectiveness and safety of HTPs for smoking cessation and the impact of HTPs on smoking prevalence.

Search methods
We searched the Cochrane Tobacco Addiction Group's Specialised Register, CENTRAL, MEDLINE, and six other databases for relevant records to January 2021, together with reference-checking and contact with study authors and relevant groups.

Selection criteria
We included randomised controlled trials (RCTs) in which people who smoked cigarettes were randomised to switch to exclusive HTP use or a control condition. Eligible outcomes were smoking cessation, adverse events, and selected biomarkers. RCTs conducted in clinic or in an ambulatory setting were deemed eligible when assessing safety, including those randomising participants to exclusively use HTPs, smoke cigarettes, or attempt abstinence from all tobacco. Time-series studies were also eligible for inclusion if they examined the populationlevel impact of heated tobacco on smoking prevalence or cigarette sales as an indirect measure.

Library
Trusted evidence. Informed decisions. Better health.

Cochrane Database of Systematic Reviews
For continuous outcomes, we calculated mean di erences on the log-transformed scale (LMD) with 95% CIs. We pooled data across studies using meta-analysis where possible.

Main results
We included 13 completed studies, of which 11 were RCTs assessing safety (2666 participants) and two were time-series studies. We judged eight RCTs to be at unclear risk of bias and three at high risk. All RCTs were funded by tobacco companies. Median length of follow-up was 13 weeks.
No studies reported smoking cessation outcomes.
There was insu icient evidence for a di erence in risk of adverse events between smokers randomised to switch to heated tobacco or continue smoking cigarettes, limited by imprecision and risk of bias (RR 1.03, 95% CI 0.92 to 1.15; I 2 = 0%; 6 studies, 1713 participants). There was insu icient evidence to determine whether risk of serious adverse events di ered between groups due to very serious imprecision and risk of bias (RR 0.79, 95% CI 0.33 to 1.94; I 2 = 0%; 4 studies, 1472 participants). There was moderate-certainty evidence for lower NNAL and COHb at follow-up in heated tobacco than cigarette smoking groups, limited by risk of bias (NNAL: LMD −0.81, 95% CI −1.07 to −0.55; I 2 = 92%; 10 studies, 1959 participants; COHb: LMD −0.74, 95% CI −0.92 to −0.52; I 2 = 96%; 9 studies, 1807 participants). Evidence for additional biomarkers of exposure are reported in the main body of the review.
There was insu icient evidence for a di erence in risk of adverse events in smokers randomised to switch to heated tobacco or attempt abstinence from all tobacco, limited by risk of bias and imprecision (RR 1.12, 95% CI 0.86 to 1.46; I 2 = 0%; 2 studies, 237 participants). Five studies reported that no serious adverse events occurred in either group (533 participants). There was moderate-certainty evidence, limited by risk of bias, that urine concentrations of NNAL at follow-up were higher in the heated tobacco use compared with abstinence group (LMD 0.50, 95% CI 0.34 to 0.66; I 2 = 0%; 5 studies, 382 participants). In addition, there was very low-certainty evidence, limited by risk of bias, inconsistency, and imprecision, for higher COHb in the heated tobacco use compared with abstinence group for intention-to-treat analyses (LMD 0.69, 95% CI 0.07 to 1.31; 3 studies, 212 participants), but lower COHb in per-protocol analyses (LMD −0.32, 95% CI −1.04 to 0.39; 2 studies, 170 participants). Evidence concerning additional biomarkers is reported in the main body of the review. Data from two time-series studies showed that the rate of decline in cigarette sales accelerated following the introduction of heated tobacco to market in Japan. This evidence was of very low-certainty as there was risk of bias, including possible confounding, and cigarette sales are an indirect measure of smoking prevalence.

Authors' conclusions
No studies reported on cigarette smoking cessation, so the e ectiveness of heated tobacco for this purpose remains uncertain. There was insu icient evidence for di erences in risk of adverse or serious adverse events between people randomised to switch to heated tobacco, smoke cigarettes, or attempt tobacco abstinence in the short-term. There was moderate-certainty evidence that heated tobacco users have lower exposure to toxicants/carcinogens than cigarette smokers and very low-to moderate-certainty evidence of higher exposure than those attempting abstinence from all tobacco. Independently funded research on the e ectiveness and safety of HTPs is needed.
The rate of decline in cigarette sales accelerated a er the introduction of heated tobacco to market in Japan but, as data were observational, it is possible other factors caused these changes. Moreover, falls in cigarette sales may not translate to declining smoking prevalence, and changes in Japan may not generalise elsewhere. To clarify the impact of rising heated tobacco use on smoking prevalence, there is a need for time-series studies that examine this association.

What we found
Our search found 13 relevant studies. No studies reported whether heated tobacco helps people stop smoking cigarettes. Eleven trials, all funded by tobacco companies and with 2666 adult smokers, compared unwanted e ects and toxin levels in people randomly assigned to use heated tobacco or to continue smoking cigarettes or abstain from tobacco use.
Two studies looked at how trends in cigarette sales changed following the launch of heated tobacco in Japan.

What are the results of our review?
We do not know whether using heated tobacco helps people to stop smoking cigarettes (no studies measured this).
We are uncertain whether the chances of getting unwanted symptoms from being asked to use heated tobacco are di erent compared with cigarettes (6 studies, 1713 participants) or no tobacco (2 studies, 237 participants). Serious unwanted symptoms in the short time period studied (average 13 weeks) were rare in all groups, which means we are uncertain about any di erences. Toxin levels were probably lower in people using heated tobacco than those smoking cigarettes (10 studies, 1959 participants), but may be higher than in people not using any tobacco products (5 studies, 382 participants).
The launch of heated tobacco products in Japan may have caused the decline in cigarette sales to speed up over time (two studies), but it is unclear whether the fall in the percentage of people who smoke also sped up because no studies looked at this.

How reliable are these results?
Results are based on data from a small number of studies, most of which were funded by tobacco companies.
Results on unwanted e ects are likely to change as more evidence becomes available. However, we are moderately confident that levels of measured toxins are lower in people using heated tobacco than smoking cigarettes, but less confident that levels were higher than in people not using any tobacco. We are also less confident that the launch of heated tobacco caused the fall in cigarette sales to speed up, as results came from a single country. LMD has no units as it is calculated from the logarithm of biomarker measurements.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

Risk with abstinence from tobacco
Risk with heated tobacco use -COHb at follow-up -measured in blood 6.0% saturation 3.75% saturation (2.5% higher to 5.0% higher) MD 2.24% saturation higher (0.69 higher to 3.79 higher)

52
(1 RCT) ⊕⊕⊝⊝ Low a -*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

GRADE Working Group grades of evidence
High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
a Downgraded two levels for indirectness: participants in the included study were given carbon-tip heated tobacco products, which are unlike heated tobacco products currently on the market. b Downgraded one levels for imprecision: confidence intervals incorporate no clinically meaningful di erence. c Downgraded two levels for imprecision: no serious adverse events occurred so confidence intervals could not be calculated.
1 study found that the yearly percentage decline in cigarette sales accelerated after the introduction of heated tobacco in Japan, increasing from a mean decline of −3.10% across 2011-2015 to −16.38% across 2016-2019. A second study found similar results using a different method; it found that per capita cigarette sales were increasing at a rate of 0.10 to 0.14 (depending on statistical approach) per month before the introduction of heated tobacco in Japan. After the introduction, they declined at a rate of 0.63 to 0.66 cigarettes per month.

GRADE Working Group grades of evidence
High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
a Downgraded one level for indirectness: cigarette sales do not necessarily translate to reductions in smoking prevalence, as smokers may reduce the amount they smoke rather than stop smoking entirely. b Downgraded one level for risk of bias: one study was considered to be at serious risk of bias, while the other was deemed at moderate risk. Cochrane Database of Systematic Reviews

Description of the condition
Tobacco use kills eight million people each year, making it one of the leading preventable causes of death worldwide (GBD 2021). Approximately 90% of these deaths result from the most harmful form of tobacco consumption: smoking (Drope 2018). Therefore, reducing smoking prevalence is one of the most e ective ways of improving population health (Holford 2014).
Although most smokers want to quit, smoking is highly addictive. Most people who make a serious attempt to quit fail, with only 3% to 10% still abstinent a er one year (Hughes 2004;Jackson 2019a). Available treatments such as behavioural support, varenicline, and nicotine replacement therapy (NRT) improve the chance that these attempts will succeed (Cahill 2016; Hartmann-Boyce 2018; Hartmann-Boyce 2019; Hartmann-Boyce 2021a). However, even with these treatments, success rates are typically under 25%, and many who try to quit do not use any support (Borland 2012; Jackson 2019b). There remains an urgent need to identify new, e ective, and safer alternatives to cigarettes to reduce smoking prevalence.

Description of the intervention
Heated (or heat-not-burn) tobacco products (HTPs) are designed to heat tobacco leaf/sheet to a high enough temperature to release nicotine-infused aerosol, without burning it or producing smoke. Many of the toxic and carcinogenic products of cigarette smoking are formed during combustion. HTPs are marketed as less harmful and as alternatives to conventional cigarettes because they are engineered to avoid combustion (Mathers 2017). The extent to which they help people quit smoking is largely unknown, and their impact on youth uptake to smoking is contentious (Czoli 2020). Therefore, it is unclear what impact HTPs will have on smoking prevalence across the population.
'Premier' was the first HTP made available for consumers. It resembled a cigarette, but the tobacco was not directly burned, instead it was heated by lighting a carbon-tip (i.e. not electronic). Premier was introduced to test markets throughout the US by RJ Reynolds in 1988, but it was not widely used and was discontinued in 1989 (Stapleton 1998). In the early 2000s, RJ Reyolds introduced another carbon-tip HTP, 'Eclipse', and they funded research to support marketing claims that it reduced health risks relative to cigarettes. A court case in the US succeeded in challenging these reduced risk claims, but trial evidence did suggest users of Eclipse had lower exposure to toxicants than people smoking cigarettes (Anderson 2008;Rennard 2002). The first electronic HTPs were produced by Philip Morris International (PMI). They introduced 'Accord' into the US in 1997 and a similar product, ' Heatbar', in Germany in 2007(Elias 2018. While these products have both since been discontinued, they acted as predecessors to 'IQOS'. The current HTP market is dominated by electronic rather than carbon-tip devices. Current brands include IQOS by PMI, 'glo' by British American Tobacco, and 'Ploom Tech' by Japan Tobacco International (WHO 2018). IQOS and glo produce aerosol by directly heating tobacco sticks which resemble small cigarettes. Conversely, Ploom Tech produces aerosol by heating a similar liquid to that found in e-cigarettes. This aerosol is then drawn through a bulb of tobacco to infuse it with flavour. Of these products, IQOS was the first to launch in 2014 in Japan and Italy, and it has since entered markets across Asia, Europe, and the Americas. Most recently, in 2019, the US Food and Drug Administration (FDA) permitted the sale of IQOS (FDA 2019) and in 2020 authorised their marketing as a modified-exposure tobacco product (FDA 2020). At the time of writing, HTPs were most popular in Japan and the Republic of Korea; tobacco sticks for HTPs constituted 15.8% and 8.0% respectively of each country's tobacco market in 2018 (WHO 2018). Market research by Euromonitor estimates that HTPs had an increased share of the retail value of all nicotine or tobacco products between 2017 and 2018, which was similar to e-cigarettes globally (Euromonitor 2020). However, HTP use remains rare in North America and much of Europe (Gallus 2021;Laverty 2021;Miller 2020;Tattan-Birch 2021).

How the intervention might work
Nicotine is the primary addictive compound in cigarettes. Neuroadaptation to repeated nicotine delivery from smoking causes people who quit to experience withdrawal and cravings (Benowitz 2010;West 2017). Like cigarettes, HTPs contain nicotine. They may aid smoking cessation in a similar way to NRT and e-cigarettes: people can use them to relieve nicotine cravings without smoking cigarettes (Wadgave 2016). HTPs may also provide certain advantages over NRT. One limitation of NRT is that it poorly addresses the behavioural and sensory cues associated with cigarette smoking, such as repeated hand-to-mouth actions and the scratch at the back of the throat when inhaling smoke. Evidence shows that denicotinised cigarettes reduce cravings and withdrawal symptoms among abstinent smokers, despite containing negligible levels of nicotine (Rose 2006). This suggests that these cues contribute to cigarette dependence. HTPs may more closely replicate these cues than NRT. Because HTP aerosol is delivered to the throat and lungs, nicotine absorption likely occurs more rapidly than from patches, gum, or lozenges, which are absorbed through the skin or buccal mucosa (Simonavicius 2018). The speed with which nicotine is absorbed may be one of the key determinants of dependence (Benowitz 2009), so HTPs may provide a better replacement for cigarette smoking than NRT. E-cigarettes also deliver nicotine rapidly to the throat and possibly lungs (Hajek 2020; Wagener 2017) and, like HTPs, they mimic the hand-to-mouth actions of cigarette smoking. But only HTPs contain tobacco leaf/sheet, so their flavour may more closely resemble cigarette smoke (Poynton 2017), which may make them more attractive to smokers (Tompkins 2021). Moreover, in some countries, the sale of nicotine e-cigarettes is banned or heavily restricted (Dyer 2019). In such environments, HTPs may be the only consumer product available that delivers nicotine rapidly through a potentially less harmful medium than tobacco smoke.
We refer to the complete replacement of cigarettes with HTPs as 'switching'. A substantial proportion of people who use HTPs for smoking cessation may continue using these products for some time a er they stop smoking cigarettes, as is the case with ecigarettes (Hajek 2019; Simonavicius 2020). Encouraging people to switch from smoking cigarettes to using HTPs would only be beneficial if HTPs are less harmful to health or if HTPs eventually help people taper o nicotine entirely. The safety of HTPs to users depends on both the acute harm, measured by adverse and serious adverse events, and the long-term harm of repeated inhalation of damaging compounds in HTP aerosols.
Biomarkers can be used to measure exposure to these harmful toxicants and carcinogens. Important exposure biomarkers include: 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL), a marker of tobacco-specific N-nitrosamine exposure that is linked to numerous cancers (IARC 2012); 1-hydroxypyrene (1-OHP) and 1-and 2-naphthol, indicators of exposure to polycyclic aromatic hydrocarbons that are associated with cancers and kidney and liver damage; 3-hydroxypropylmercapturic acid (3-HPMA), a marker of exposure to acrolein that is linked to respiratory disease (Yeager 2016); and carboxyhaemoglobin (COHb), a measure of recent carbon monoxide (CO) intake. Details about these and other exposure biomarkers are available in Appendix 1.
Manufacturers of HTPs claim that the aerosol they produce contains significantly lower levels of toxicants than cigarette smoke and, as a result, that they have reduced risk potential or are less harmful (BAT 2020;PMI 2018). Two systematic reviews supported claims about lower toxicant levels, but found that most research into HTPs was funded through sources a iliated with the tobacco industry (Jankowski 2019; Simonavicius 2018). In addition, reduced exposure does not necessarily indicate reduced harm. The US FDA judged that there was su icient evidence that IQOS reduced exposure to harmful chemicals (FDA 2020), but insu icient evidence on whether switching from smoking to HTPs reduces harm, such as pulmonary function or biomarkers linked to smoking-related harm (Glantz 2018;Moazed 2018). It is also the case that safety, especially of longer-term use, cannot be addressed with confidence until long-term cohort studies have collected su icient data.

Why it is important to do this review
There is substantial variation between countries in their regulatory approaches to HTPs, and within countries across di erent nicotine products. In order for policymakers to regulate HTPs e ectively and proportionately, there is a need for evidence to inform a judgement on their likely public health impact. The net impact of HTPs on public health will depend on a variety of factors. Three influential elements that could result in HTPs benefiting public health are if they increase smoking cessation, decrease smoking prevalence, and are less harmful than cigarette smoking. Conversely, even if these products are shown to be much less harmful than cigarettes, HTPs could damage public health if they hinder smoking cessation or increase smoking prevalence.
The e ect of HTP use on smoking prevalence will depend on whether they influence rates of attempted quitting among cigarette smokers, the proportion of these attempts that are successful, cigarette uptake among non-smokers, and relapse among people who had previously quit smoking. Therefore, we are not only interested in studies that report individual-level e ects of HTPs on smoking cessation, but also those that estimate their populationlevel e ects on smoking prevalence. This review will investigate upto-date evidence for both, using appropriate study designs.
The growing popularity of HTPs means that people who smoke may be increasingly likely to seek advice from practitioners who need to know whether HTPs are e ective for smoking cessation and how their safety compares with cigarettes and other alternative nicotine products. If HTPs are found to be safe and e ective for smoking cessation, they would o er a novel treatment for cigarette addiction. Moreover, evidence on associations between HTP use and smoking prevalence will help to guide the regulation of HTPs.
Licensed smoking cessation medications tend to be used for a short time while quitting, whereas people may continue using HTPs for extended periods a er they quit. This means that it is especially important to evaluate indicators of the long-term safety of HTP use (such as exposure to toxicants and carcinogens) in addition to adverse events occurring in the short term.

O B J E C T I V E S
To evaluate the e ectiveness and safety of HTPs for smoking cessation and the impact of HTPs on smoking prevalence.

Types of studies
We divided the methods into the three subsections, representing the di erent objectives of the review: e ectiveness for smoking cessation, safety, and smoking prevalence.

E ectiveness for smoking cessation
Individual-level and cluster-randomised controlled trials (RCTs) to examine the e ectiveness (or e icacy) of HTPs for tobacco smoking cessation.

Safety
Individual-level, randomised cross-over and cluster-RCTs to explore adverse and serious adverse events and biomarkers of toxicant and carcinogen exposure. RCTs in optimised settings for smoking cessation, such as those where participants stayed in a clinic with restricted access to tobacco products, were eligible for inclusion, as were studies in naturalistic or ambulatory settings.

Smoking prevalence
Interrupted and multiple time-series studies were included to examine the population-level e ect of HTPs on cigarette smoking prevalence. Smoking cessation interventions may not be representative of the way most people use HTPs, which is without support from a researcher or trained specialist. Moreover, even if HTPs encourage smoking cessation among those trying to quit, their impact on smoking prevalence depends on how they a ect smoking initiation and the number of people who make a quit attempt and are successful in remaining abstinent. We used time-series studies to assess how changes in HTP prevalence are associated with changes in smoking prevalence (or cigarette sales), with the limitation that associations might not reflect causal e ects.
We included studies regardless of language or status of publication.

E ectiveness and safety
We included adults who were defined as current cigarette smokers by the study at the time of enrolment.

Library
Trusted evidence. Informed decisions. Better health.
Cochrane Database of Systematic Reviews who indicated their smoking status or consumption and HTP use or consumption, measured by survey or by record of sales.

Types of interventions
HTPs, defined as hand-held devices that aim to heat tobacco to a temperature high enough to produce a nicotine-infused aerosol but too low to cause self-sustaining combustion. HTPs di er from ecigarettes in that they heat compressed tobacco leaf rather than a liquid that is infused with nicotine.

E ectiveness and safety
We were interested in studies that compared HTPs, or the addition of HTPs, to no treatment (i.e. continued tobacco smoking), placebo or any other smoking cessation treatment, including NRT, e-cigarettes, snus, varenicline, bupropion, and behavioural support. HTPs could be provided in addition to any other smoking cessation treatment, providing there was equivalent provision of the additional treatment for the control group. We only included studies where participants in the HTP arm were instructed to stop smoking combustible cigarettes for at least seven days.

Smoking prevalence
For interrupted time-series studies, the interventions of interest were the introduction of HTPs to market or the time point where HTPs began gaining popularity. For multiple time-series studies, we were interested in the extent to which changes in the prevalence of HTP use were associated with changes in the prevalence of cigarette smoking (or cigarette sales as a proxy), a er adjusting for other influences that could a ect changes in the prevalence of smoking at the population level.

E ectiveness
• Tobacco smoking cessation at the longest follow-up point available, using intention-to-treat and biochemically verified abstinence where possible. While HTPs contain tobacco, they are designed to avoid or minimise combustion and smoke. Therefore, HTP use was not classified as tobacco smoking. If review updates find studies reporting smoking cessation, we will only include those which report abstinence at fourweek follow-up or longer. We will use the strictest definition of abstinence recorded, that is, prolonged or continuous abstinence over point prevalence, and biochemically verified over self-reported abstinence. Typically, Cochrane Tobacco Addiction Group reviews only include data on smoking cessation at six months or longer. We will include short-term outcomes in the next update of this review because we anticipate a paucity of longer-term data. In subsequent updates, as and when more data become available, we may change the inclusion criteria accordingly.

Safety
• Number of people reporting adverse events and serious adverse events. We defined serious adverse events as medical incidents that are potentially life-threatening, require hospitalisation, result in disability or death, or a combination of these. Adverse events were medical problems -including cough, headache, and dry mouth -that did not fulfil the above criteria to be considered serious.

Smoking prevalence
• Change in the prevalence of cigarette smoking, measured as the proportion of people in a given locality that regularly smoke cigarettes or other combustible tobacco products, over a defined time period. We included cigarette sales as a proxy for prevalence, measured as the number of cigarettes sold in a given locality over a given time period. This was used as a proxy because, in a population where mean cigarette consumption among smokers remains stable, declines in cigarette sales imply falls in smoking prevalence. However, it should be considered an indirect measure of prevalence because smokers can reduce their cigarette consumption without quitting.

Secondary outcomes
All secondary outcomes are measures of safety. We only included studies that reported safety outcomes at one-week follow-up or longer.
• Biomarkers of toxicant and carcinogen exposure. We included measures of exposure to tobacco-specific N-nitrosamines, polycyclic aromatic hydrocarbons, volatile organic compounds, and CO (see Appendix 1 for details on associations with health outcomes). • Biomarkers of harm, also known as surrogate endpoints. We included measures of lung function (forced expiratory volume in one second (FEV 1 ), forced vital capacity (FVC), and FEV 1 / FVC), blood pressure, heart rate, heart rate variability, and blood oxygen saturation.

Electronic searches
We searched the following databases on 19 January 2021: • As we were only interested in studies that used humans, we excluded those with the terms animal$ OR mice OR rat$ OR in vitro OR in silico OR in vivo in their title.

Searching other resources
We searched the reference lists of eligible studies found in the literature search.
In order to identify government reports and in-press or unpublished studies, we contacted relevant charities and authors of published research or trial protocols. We used the searches of ClinicalTrials.gov and the ICTRP detailed above to identify trial registry records.

Selection of studies
Two review authors (of HTB, JB, LK, ES, and LB) independently prescreened titles and abstracts of articles identified in the search, using a screening checklist. We resolved disagreements through discussion or referral to a third review author. We conducted screening using Covidence so ware (Covidence).
Two review authors (of HTB, JB, LK, ES, and LB) independently screened the full text of articles that passed prescreening. We consulted a third review author to resolve any disagreements that were not resolved through discussion.

Data extraction and management
We produced two custom data extraction forms: one for e ectiveness and safety, and the other for smoking prevalence. Details of these forms are available in Appendix 2.
Two review authors (of HTB, JB, LK, ES, and LB) independently extracted data from included studies. When discrepancies could not be resolved through discussion, we referred to a third review author. We contacted authors of included studies if additional information was needed.

E ectiveness and safety
Two review authors (of HTB, JB, LK, ES, and LB) independently assessed risks of bias for all included RCTs using the Cochrane risk of bias tool version 1. We followed the guidance as set out in the Cochrane Handbook for Systematic Reviews of Interventions to evaluate the following domains: sequence generation; allocation concealment; blinding of outcome assessment; incomplete outcome data; selective reporting; and other sources of bias (Higgins 2011).

Smoking prevalence
Two review authors independently assessed risk of bias for included time-series studies using the ROBINS-I tool (Sterne 2016).

E ectiveness and safety
We calculated risk ratios (RRs) and 95% confidence intervals (CIs) for dichotomous outcomes.
For continuous safety data, we calculated mean di erences on the raw (MD) or log-transformed (LMD) scale and the corresponding 95% CIs between the heated tobacco and control groups at follow-up. When studies reported geometric means, we converted these onto the (natural) log scale, and when studies being pooled reported mixtures of geometric and arithmetic means, we converted them all onto the log scale, using Method 1 described in Higgins 2008 where appropriate.
We used the longest follow-up data reported, with treatment e ects calculated on an intention-to-treat basis where possible.

Smoking prevalence
For interrupted time-series studies, the treatment e ect could have been reflected by the step change and change in trends in smoking prevalence or cigarette sales following the introduction of HTPs to the market (or the time point where they started gaining popularity), a er adjusting for confounding variables.
For multiple time-series studies (in future review updates), the treatment e ect of interest will be the association between HTP prevalence and smoking prevalence or cigarette sales, a er adjusting for confounding variables. Where variables are logtransformed, the resulting coe icient describes the percentage change in cigarette smoking prevalence associated with a 1% change in HTP prevalence.

E ectiveness and safety
For RCTs with more than two intervention arms, we combined data from all relevant intervention conditions where HTPs were o ered. For RCTs with more than two control arms, we combined data from each of these arms, and we chose the most appropriate comparator. If it is not appropriate to pool the intervention arms (in future updates) then we will split the control arm to act as a comparator to each separate intervention arm. If future updates of this review identify cluster-RCTs, we will attempt to extract an estimate of the e ect that accounts for the cluster design of the study. Where this is not reported, we will attempt to perform the correct analysis if required data are available.

E ectiveness
If we assess smoking cessation in future updates of this review, we will assume that people with missing data at follow-up have not stopped smoking, as is common in the field.

Safety
When assessing adverse and serious adverse events, we calculated the proportion of those available at follow-up who experienced an event (when such data are available) rather than the proportion of people who were randomised, when follow-up information was reported. When assessing biomarkers, we removed participants with missing follow-up data from the analysis.

Library
Trusted evidence. Informed decisions. Better health.

Smoking prevalence
We did not expect issues with missing data in time-series studies.

Assessment of heterogeneity
To assess whether to conduct meta-analyses, we considered the characteristics of included studies to identify substantial clinical or methodological heterogeneity. If we deemed the studies to be su iciently homogeneous to be combined meaningfully, we assessed statistical heterogeneity using the I 2 statistic. If the I 2 statistic was greater than 50%, we reported substantial heterogeneity. If I 2 was greater than 75%, we considered the appropriateness of presenting pooled results, and based this decision on consistency in the direction of e ect across included studies.

Assessment of reporting biases
In future updates of this review, we will assess reporting bias using funnel plots if we deem it appropriate to pool 10 or more studies in any analysis. The greater the asymmetry in the plots, the higher the risk of reporting bias.

E ectiveness
The primary outcome of smoking cessation provides dichotomous data. Following the standard methods of the Cochrane Tobacco Addiction Group, we aimed to combine RRs and 95% CIs from individual studies using a Mantel-Haenszel random-e ects model, to calculate pooled overall RRs with 95% CIs.

Safety
For dichotomous safety outcomes (i.e. adverse and serious adverse events), we combined RRs and 95% CIs from individual studies using a Mantel-Haenszel random-e ects model to calculate pooled overall RRs with 95% CIs.
For continuous safety outcomes measuring biomarkers, we pooled the MDs or LMDs and measures of variance of individual studies using a generic inverse variance random-e ects model.

Smoking prevalence
We aimed to calculate pooled estimates and their standard errors using a random-e ects model for each of three coe icients, when reported: step change in smoking prevalence or cigarette sales following the introduction of HTPs; change in these trends a er the introduction; and changes associated with changes in prevalence or sale of HTPs. We did not pool time-series studies with notably di erent time periods (e.g. weekly versus annual).

Subgroup analysis and investigation of heterogeneity
For biomarker outcomes, we undertook subgroup analyses to investigate di erences by whether analyses were per-protocol or intention-to-treat. We define per-protocol analyses as those that only included participants who exclusively (or almost exclusively) used the product they were assigned, whereas intention-to-treat analyses include all participants regardless of actual product use.
If appropriate for future updates of this review, we will undertake subgroup analyses to investigate di erences by: • intensity of behavioural support provided; • characteristics of HTP device (e.g. model used).

Sensitivity analysis
We aimed to carry out sensitivity analyses removing studies: • judged at high risk of bias for at least one domain; • with a minimum length of follow-up of less than four weeks (safety outcomes only); • where participants were given carbon-tip, rather than electronic, HTPs.
If appropriate for future updates of this review, we will also carry out sensitivity analyses where we: • remove studies that are funded by (or authors have received funding from) the tobacco industry; • only classify participants as HTP users if they use their product daily (smoking prevalence only); • only include interrupted time-series studies in localities where HTPs achieved widespread use a er they were introduced to market.

Summary of findings and assessment of the certainty of the evidence
We created summary of findings tables using GRADEpro GDT for all primary outcomes and for two biomarkers of exposure (NNAL and COHb), following the guidelines in Cochrane Handbook of Systematic Reviews of Interventions (GRADEpro GDT; Higgins 2021; Schünemann 2020). We chose NNAL and COHb because they are well-established indicators of tobacco smoke exposure (Chang 2017;Hedblad 2005). We used the five GRADE considerations (risk of bias, inconsistency, imprecision, indirectness, and publication bias) to assess the certainty of the body of evidence for each of these outcomes.

Results of the search
Our bibliographic database searches identified 1504 non-duplicate records ( Figure 1). We found a further four records through screening references in the papers identified through electronic searches. We screened all records and retrieved the full-text of 121 potentially relevant articles. A er screening and checking the full texts, we included 23 records, representing 13 completed (Characteristics of included studies) and three ongoing studies (Characteristics of ongoing studies). We excluded 98 records during full text screening, and we give reasons for exclusion for 11 studies (Characteristics of excluded studies).

Library
Trusted evidence. Informed decisions. Better health.
Cochrane Database of Systematic Reviews

Included studies
A summary of the 13 included studies is given below. Further details of each study can be found in the Characteristics of included studies section.

Participants
Of the 13 included studies, 11 collected data from participants. Two studies used sales data and are thus excluded from subsequent discussion of participant characteristics. A total of 2666 participants were recruited across the 11 RCTs. Three studies were conducted in Japan, three in the USA, two in Poland, two in the UK, and one in South Korea. These studies were conducted in adults who smoked cigarettes. Seven studies exclusively recruited participants who were not motivated to quit smoking cigarettes. One study only recruited participants diagnosed with generalised chronic periodontitis (NCT03364751). Three studies only recruited people who were Japanese or of "Japanese ethnicity" (

Interventions and comparators
All 11 included RCTs gave HTPs to participants. Two studies provided participants with the carbon-tip products 'CHTP 1.2' and 'Eclipse' (Bosilkovska 2020; Ogden 2015). All others provided electronic heating devices alongside tobacco sticks, with PMI's IQOS-family products (or their predecessors) provided in eight studies and BAT's glo-family products in one study (Gale 2020).
All 11 RCTs compared participants randomised to receive a HTP or to continue smoking cigarettes. Five studies also had tobacco abstinence as an additional comparator and one study had snus use as an additional comparator (Ogden 2015). Summaries of study results by comparator are available in E ects of interventions. Further details on the intervention and comparator groups for each are available in the Characteristics of included studies section.
There were two interrupted time-series studies using cigarette sales data from Japan. The intervention in these studies was the introduction of heated tobacco to market, with the launch of IQOS in 2015 or 2016 (depending on region).

Outcomes
Of the 13 included studies: • none reported smoking cessation rates; • 10 reported data on adverse events (four of which did not provide data in each trial arm). Commonly reported adverse events included cough, headache, gastrointestinal issues (e.g. diarrhoea), dry mouth, hyperglycaemia, and decreased haemoglobin; • 10 reported data on serious adverse events. Most studies defined serious adverse events as medical incidents that were potentially life-threatening, require hospitalisation, resulted in disability or death, of a combination of these; • 11 reported data on at least one biomarker of toxicant and carcinogen exposure; • five reported data on at least one biomarker of harm; • none reported time-series data on smoking prevalence; • two reported time-series data on cigarette sales.

Study types and funding
Eleven studies were RCTs and two were observational time-series studies. All 11 RCTs were funded by the tobacco industry. One timeseries study was funded through government grants, while the other had no specific funding. Figure 1 shows the most common reasons for exclusion of studies during full-text screening, which were: duplicate reports; less than one week of follow-up; and wrong study design (e.g. testing acute rather than extended e ects of HTP use).

Excluded studies
In the Characteristics of excluded studies table, we list more detailed exclusion reasons for 11 of these studies. This list is not comprehensive, only containing studies that a reader might plausibly expect be included.

Risk of bias in included studies
Overall, we judged eight of the 11 included RCTs at unclear risk of bias and three at high risk of bias, assessed using the ROB v1 criteria (Higgins 2011). Figure 2 shows judgements across the risk of bias domains for each RCT. Detailed rationale for these judgements can be found in the Characteristics of included studies.

Library
Trusted evidence. Informed decisions. Better health.
Cochrane Database of Systematic Reviews Cochrane Database of Systematic Reviews Risk of bias for the two included time-series studies was assessed using the ROBINS-I tool (Sterne 2016). One time-series study was at moderate risk of bias, while the other was at serious risk. Detailed risk of bias assessments for these time-series studies can be found in the appendices (Appendix 3; Appendix 4).

Allocation
All included RCTs were at unclear risk of selection bias, as there was no or insu icient information about random sequence generation or allocation concealment, or both.

Blinding
We judged all studies at low risk of detection bias, as most reported outcomes were biochemical and hence judged at low risk of di erential misreport. We planned to assess performance bias for smoking cessation outcomes, with studies judged at low risk if intervention and control arms received similar levels of behavioural support. As no study reported on smoking cessation outcomes, performance bias was not assessed.

Incomplete outcome data
Seven studies were at low risk of attrition bias, due to high and similar rates of follow-up across treatment and comparator arms (Bosilkovska 2020; Gale 2020; Lüdicke 2018; Lüdicke 2019; Martin 2012; NCT03364751; Ogden 2015). Three studies were at unclear risk as they did not provide su icient details about attrition (Tricker 2012a; Tricker 2012b; Tricker 2012c). Haziza 2019 was at high risk of attrition bias due to substantial loss to follow-up that was greater in the heated tobacco arm.

Selective reporting
We judged five studies at low risk of reporting bias, as all prespecified outcomes were reported (Bosilkovska 2020; Gale 2020; Haziza 2019; Lüdicke 2019; NCT03364751). Five studies were at unclear risk as there was no preregistered study protocol (Martin 2012; Ogden 2015; Tricker 2012a; Tricker 2012b; Tricker 2012c). Lüdicke 2018 was at high risk of reporting bias, as one preregistered outcome of interest was not reported (FEV 1 /FVC).

Other potential sources of bias
One study was at high risk of other bias as it did not report results across randomised trial arms (NCT03364751). Instead, they only reported results based on actual product use.

E ects of interventions
See: Summary of findings 1 Heated tobacco use compared with cigarette smoking; Summary of findings 2 Heated tobacco use compared with abstinence from tobacco; Summary of findings 3 Heated tobacco use compared with snus use; Summary of findings 4 Population-level impact of heated tobacco on cigarette smoking prevalence See: Summary of findings 1: heated tobacco use compared with cigarette smoking; Summary of findings 2: heated tobacco use compared with abstinence from tobacco; Summary of findings 3: heated tobacco use compared with snus use; Summary of findings 4: population-level impact of heated tobacco on cigarette smoking prevalence.
Data on each outcome are summarised below, alongside links to forest plots. In these forest plots, benefit of HTPs is usually shown on the le , as lower toxicant levels or risk of adverse events indicates benefits of HTPs relative to the comparator.

Tobacco smoking cessation
No studies reported on the e ectiveness of heated tobacco for smoking cessation.

Adverse events
Pooled data from six studies showed insu icient evidence of a di erence in the number of participants reporting adverse events between those in the heated tobacco use and cigarette smoking groups, but the CI contained the possibility of small but clinically meaningful di erences in both directions (RR 1.03, 95% CI 0.92 to 1.15; I 2 = 0%; 1713 participants; Analysis 1.1; Summary of findings 1). Two studies were at high risk of bias, while the remaining four were at unclear risk. Removing studies judged at high risk of bias did not substantially change the interpretation of results (RR 0.98, 95% CI 0.87 to 1.11; I 2 = 0%; 1472 participants), neither did removing the two studies that used carbon-tip, rather than electronic, HTPs (RR 1.04, 95% CI 0.82 to 1.30; I 2 = 35%; 1510 participants). All six studies had a follow-up of at least four weeks.

Serious adverse events
Pooled data from four studies showed insu icient evidence of a di erence in serious adverse events reported in the heated tobacco use compared with cigarette smoking group, with a wide CI that contained no di erence as well as the possibility of more events in either group (RR 0.79, 95% CI 0.33 to 1.94; I 2 = 0%; 1472 participants; Analysis 1.2; Summary of findings 1). All pooled studies were at unclear risk of bias and had a follow-up of at least four weeks. Removing the two studies that used carbontip, rather than electronic, HTPs did not substantially change the interpretation of results (RR 0.93, 95% CI 0.34 to 2.58; I 2 = 0%; 1269 participants). In a further five studies, there were no serious adverse events reported, which meant their data could not be pooled (Haziza 2019; Lüdicke 2018; Tricker 2012a; Tricker 2012b; Tricker 2012c).

Toxicant and carcinogen exposure
Pooled data from 1960 participants across 10 studies showed: • lower 1-OHP at follow-up in heated tobacco use compared with cigarette smoking groups (LMD −0.42, 95% CI −0.67 to −0.17; Analysis 1.3). Heterogeneity was high at I 2 = 94%, but the direction of the di erence was consistent across all studies except Ogden 2015, where carbon-tip HTPs were provided. It was also consistent across sensitivity analyses removing two studies at high risk of bias, two studies using carbon-tip HTPs, and three studies with less than four weeks of follow-up (

Cochrane Database of Systematic Reviews
• lower 3-HPMA at follow-up in heated tobacco use compared with cigarette smoking groups (LMD −0.40, 95% CI −0.62 to −0.17; Analysis 1.8). Heterogeneity was high at I 2 = 95%, but the direction of the di erence was consistent across sensitivity analyses and all studies except Ogden 2015 (Table 1); • lower MHBMA at follow-up in heated tobacco use compared with cigarette smoking groups (LMD −1.15, 95% CI −1.52 to −0.78; Analysis 1.9). Heterogeneity was high at I 2 = 94%, but the direction of the di erence was consistent across studies and sensitivity analyses (Table 1); • lower NNAL at follow-up in heated tobacco use compared with cigarette smoking groups (LMD −0.81, 95% CI −1.07 to −0.55; Analysis 1.10; Summary of findings 1). Heterogeneity was high at I 2 = 92%, but the direction of the di erence was consistent across sensitivity analyses and all studies except Ogden 2015 (Table 1). Another study also reported NNAL; as data were analysed based on actual product use rather than randomised group, it was not pooled (NCT03364751). It found results that were compatible with those from pooled data (LMD −1.46, 95% CI −1.81 to −1.10; 151 participants).
Pooled data for nine studies showed lower levels of COHb at followup in heated tobacco use compared with cigarette smoking groups (LMD −0.74, 95% CI −0.97 to −0.52; 1807 participants; Analysis 1.7; Summary of findings 1). Heterogeneity was high at I 2 = 96%, but estimates from each study were consistently in favour of the heated tobacco group. Results were similar a er removing two studies at high risk of bias, two studies using carbon-tip HTPs, and three studies with less than four weeks of follow-up (Table 1).
In addition, pooled data from three studies showed lower levels of exhaled CO at follow-up in heated tobacco use compared with cigarette smoking groups (MD −9.13ppm, 95% CI −10.49 to −7.78; 1322 participants; Analysis 1.6). There was low heterogeneity at I 2 = 4% and e ects for each study were in the same direction. All three studies were at unclear risk of bias, used electronic HTPs, and had at least four weeks of follow-up.
Ogden 2015 reported data from 63 participants showing insu icient evidence of a di erence in 1-naphthol between the heated tobacco use and cigarette smoking groups, with the CI containing the possibility of clinically meaningful e ects in either direction (MD 2.60 μg/24 hours, 95% CI −16.11 to 21.31; Analysis 1.4). The study also found that 2-naphthol was lower in the heated tobacco use group compared with the cigarette smoking group; however, the CIs were wide (MD −4.00 μg/24 hours, 95% CI −7.89 to −0.11; Analysis 1.5). This study was at unclear risk of bias, used a carbon-tip HTP, and had a follow-up of greater than four weeks.
No studies reported on exposure to lead or cadmium.

Biomarkers of harm
Pooled data from five studies showed greater lung function, measured usingFEV 1 , at follow-up among participants in the heated tobacco use compared with cigarette smoking groups (LMD 0.02, 95% CI 0 to 0.03; I 2 = 0%; 1290 participants; Analysis 1.11).
Results were similar a er removing two studies at high risk of bias and one study using carbon-tip HTPs. All five studies had a followup of at least four weeks (Table 1).
Pooled data from 196 participants across two studies found no evidence of a di erence in FVC between those randomised to heated tobacco use versus cigarette smoking, but the CI contained the possibility of clinically meaningful di erences in both directions (MD −0.12 L, 95% CI −0.45 to 0.21; I 2 = 38%; Analysis 1.14). Both studies had at least four weeks of follow-up, were judged at high risk of bias, and provided electronic rather than carbon-tip devices.
Pooled data from 288 participants across three studies showed no evidence of a di erence in systolic blood pressure (LMD 0.00, 95% CI −0.02 to 0.02; I 2 = 0%; Analysis 1.12) or diastolic blood pressure (LMD 0.00, 95% CI −0.03 to 0.03; I 2 = 0%; Analysis 1.13) at follow-up between heated tobacco use and cigarette smoking groups. Results were similar a er removing two studies at high risk of bias and one study using carbon-tip HTPs. All three studies had a follow-up of at least four weeks (Table 1).
No studies reported on FEV 1 /FVC, heart rate, or blood oxygen saturation.

Adverse events
Pooled data from two studies showed insu icient evidence of a di erence in the number of participants reporting adverse events between the heated tobacco use and attempted tobacco abstinence groups, with the CI containing the possibility of clinically meaningful di erences in both directions (RR 1.12, 95% CI 0.86 to 1.46; I 2 = 0%; 237 participants; Analysis 2.1; Summary of findings 2). Both studies were at high risk of bias, used electronic HTPs, and had a follow-up of at least four weeks.

Serious adverse events
Five studies reported that no serious adverse events occurred across either the heated tobacco or tobacco abstinence groups (Haziza 2019; Lüdicke 2018; Tricker 2012a; Tricker 2012b; Tricker 2012c), which meant that data could not be pooled (533 participants; Analysis 2.2, Summary of findings 2). Two studies were at high risk of bias, while the remaining three were at unclear risk. All studies used electronic HTPs and two had at least four weeks of follow-up.

Toxicant and carcinogen exposure
All five studies reporting on biomarkersof toxicant and carcinogen exposure for this comparison used electronic rather than carbontip HTPs. Pooled data from 382 participants across these studies showed: • higher 1-OHP at follow-up in heated tobacco use groups compared with tobacco abstinence groups, but CIs were wide and contained no di erence (LMD 0.12, 95% CI −0.03 to 0.28; Analysis 2.3). Heterogeneity was moderate with an I 2 of 54%, which reduced to 12% in a sensitivity analysis where the two studies at high risk of bias were removed. The direction of the e ect was unchanged a er removing these studies and a er removing three studies with less than four weeks of follow-up ( Cochrane Database of Systematic Reviews groups for intention-to-treat analyses (LMD 0.69, 95% CI 0.07 to 1.31; I 2 = 96%; 3 studies, 212 participants; Analysis 2.4), but lower COHb, limited by imprecision, for per-protocol analyses (LMD −0.32, 95% CI −1.04 to 0.39; I 2 = 91%; 2 studies, 170 participants; Analysis 2.4). Because of these subgroup di erences and high overall heterogeneity (I 2 = 99%), we did not present pooled results (Summary of findings 2). Heterogeneity was 96% when we removed the two studies at high risk of bias and 91% when we removed the three studies with less than four weeks of follow-up. The direction of the di erence was reversed when studies with less than four weeks of follow-up were removed (Table 2); • higher 3-HPMA in heated tobacco use compared with tobacco abstinence groups (LMD 0.56, 95% CI 0.33 to 0.80; Analysis 2.5). Heterogeneity was high with an I 2 of 85%, which reduced to 0% when removing three studies with less than four weeks of followup. Di erences were smaller when we removed these studies (LMD 0.35, 95% CI 0.20 to 0.50; 170 participants), but larger when we removed two studies at high risk of bias (LMD 0.64, 95% CI 0.32 to 0.96; 212 participants) ( Table 2); • higher MHBMA in heated tobacco use compared with tobacco abstinence groups (LMD 0.67, 95% CI −0.12 to 1.45; Analysis 2.6), but CIs contained the potential for no di erence. Heterogeneity was high with an I 2 of 96%, which reduced to 0% when removing three studies with less than four weeks of follow-up. Di erences were smaller when we removed these studies (LMD 0.07, 95% CI −0.16 to 0.30; 170 participants), but larger when we removed two studies at high risk of bias (LMD 0.97, 95% CI 0.02 to 1.92; 212 participants); • higher NNAL in heated tobacco use compared with tobacco abstinence groups (LMD 0.50, 95% CI 0.34 to 0.66; I 2 = 0%; Analysis 2.7; Summary of findings 2). Results were similar in sensitivity analyses removing two studies at high risk of bias and three studies with less than four weeks of follow-up.

Biomarkers of harm
Both of the studies that reported on biomarkers of harm were at high risk of bias, used electronic rather than carbon-tip HTPs, and had at least four weeks of follow-up. Pooled data from 170 participants across these two studies showed: • insu icient evidence of a di erence in lung function, measured usingFEV 1 at follow-up, among participants in the heated tobacco use compared with tobacco abstinence groups, with the CI including the possibility of clinically meaningful di erences in both directions (LMD −0, 95% CI −0.06 to 0.06; I 2 = 38%; Analysis 2.8); • higher systolic blood pressure at follow-up in the heated tobacco use compared with tobacco abstinence groups, but the CI included no di erence (LMD 0.02, 95% CI −0.01 to 0.05; I 2 = 0%; Analysis 2.9); • insu icient evidence of a di erence in diastolic blood pressure at follow-up between heated tobacco use and tobacco abstinence groups, with the CIs including the possibility of clinically meaningful di erences in both directions (LMD 0, 95% CI −0.04 to 0.04; I 2 = 0%; Analysis 2.10).
Both studies also reported data from 172 participants on FVC, with insu icient evidence for a di erence between those randomised to use heated tobacco versus tobacco abstinence (MD −0.02 L, 95% CI −0.29 to 0.26; I 2 = 0%; Analysis 2.11). The CIs contained the possibility of clinically meaningful di erences in both directions.
No studies reported FEV 1 /FVC, heart rate, or blood oxygen saturation.

Adverse events
In Ogden 2015, a higher number of participants reported adverse events in the group assigned to use heated tobacco compared with snus, but the CI was wide and included no di erence (RR 1.30, 95% CI 0.94 to 1.80; 87 participants; Analysis 3.1; Summary of findings 3). The study had a follow-up of at least four weeks, was at unclear risk of bias, and used carbon-tip HTPs.

Serious adverse events
Ogden 2015 reported that no serious adverse events occurred across either the heated tobacco or snus use groups (87 participants; Analysis 3.2; Summary of findings 3).

Secondary outcomes Toxicant and carcinogen exposure
Data from 50 participants (52 participants for COHb) in Ogden 2015 showed: • higher 1-OHP at follow-up in the heated tobacco compared with snus group (MD 252 μg/24 hours, 95% CI 100 to 404; Analysis 3.3); • insu icient evidence of a di erence in 1-naphthol between the heated tobacco and snus groups, but the CI was wide and it contained the possibility of clinically meaningful e ects in either direction (MD −2.4 μg/24 hours, 95% CI −27.7 to 22.9; Analysis 3.4); • lower 2-naphthol at follow-up in the heated tobacco compared with snus group, but the CI was wide and contained no di erence as well as the possibility of clinically meaningful e ects in either direction (MD −3.4 μg/24 hours, 95% CI −10.4 to 3.6; Analysis 3.5); • higher COHb at follow-up in the heated tobacco compared with snus group (MD 2.24% saturation, 95% CI 0.69 to 3.79; Analysis 3.6; Summary of findings 3); • higher 3-HPMA at follow-up in the heated tobacco compared with snus group (MD 1.07 mg/24 hours, 95% CI 0.39 to 1.75; Analysis 3.7); • insu icient evidence of a di erence in MHBMA between the heated tobacco and snus groups, with the CI containing the possibility of clinically meaningful e ects in either direction (MD 0.33 μg/24 hours, 95% CI −1.36 to 2.02; Analysis 3.8); • lower NNAL at follow-up in the heated tobacco compared with snus group, but the CI was wide and contained no di erence (MD −160 ng/24 hours, 95% CI −339 to 19; Analysis 3.9, Summary of findings 3).
No studies reported on exposure to exhaled CO, lead, or cadmium. Cochrane Database of Systematic Reviews

Biomarkers of harm
No studies reported on FEV 1 , FVC,FEV 1 /FVC, systolic blood pressure, diastolic blood pressure, heart rate, or blood oxygen saturation.

Cigarette sales
Cummings 2020 found that the yearly percentage decline in cigarette sales accelerated a er the introduction of HTPs in Japan, increasing from a mean decline of −3.10% across 2011-2015 to −16.38% across 2016-2019 (Summary of findings 4). This study was considered at serious risk of bias due to the limited number of time points (five) used to calculate the pre-intervention trend. Stoklosa 2020 found similar results using a di erent method and monthly rather than annual data; it found that per capita cigarette sales were increasing at a rate of 0.10 to 0.14 (depending on statistical approach) per month before the introduction of heated tobacco in Japan. A er the introduction, per capita cigarette sales declined at a rate of 0.63 to 0.66 cigarettes per month. This study was at moderate risk of bias, due to possible confounding and lack of a preregistered protocol. However, risk of confounding was partially accounted for using regional controls, with the monthly data enabling a su icient number of time points used to determine pre-and postintervention trends across regions.

Summary of main results
Our searches found no studies that reported the e ectiveness of heated tobacco for smoking cessation, but they did find 11 RCTs assessing the safety of heated tobacco -all of which were funded by tobacco companies. Results on adverse and serious adverse events were inconclusive, with insu icient short-term evidence of di erences between smokers randomised to switch to heated tobacco use or to cigarette smoking, attempted tobacco abstinence, or snus use (Summary of findings 1; Summary of findings 2; Summary of findings 3). No studies detected serious harms considered to be related to heated tobacco use. Pooled data showed there was moderate-certainty evidence that exposure to some measured toxicants and carcinogens was lower in smokers randomised to switch to heated tobacco than continue smoking cigarettes (Summary of findings 1), but very low-to moderatecertainty evidence of higher exposures than in those attempting abstinence from all tobacco (Summary of findings 2).
No studies directly assessed how trends in smoking prevalence changed following the introduction of heated tobacco to market, but we found two time-series studies on cigarette sales. Results from both studies showed that the rate of decline in cigarette sales accelerated from before to a er the launch of IQOS in Japan (Summary of findings 4). However, declining cigarette sales might not translate to falling smoking prevalence, as smokers can reduce the number of cigarettes they smoke without quitting entirely. Moreover, because data were observational, it is possible that changes were caused by other factors (e.g. demographic shi s or delayed e ects of tobacco control policies).

Overall completeness and applicability of evidence
Although included studies had conditions in which they asked smokers to switch completely to HTP or attempt abstinence from all tobacco, none reported smoking cessation outcomes. This means that the e ectiveness of heated tobacco for smoking cessation remains uncertain. However, we found one ongoing study that will evaluate their e ectiveness relative to e-cigarettes (Caponnetto 2020).
Safety data came from a wide range of locations across Europe, Asia, and North America. Conversely, both time-series studies used data from a single country (Japan), which limits the generalisability of conclusions. For instance, Japan di ers from many countries because it is illegal to sell nicotine e-cigarettes unless they are registered as a pharmaceutical product. This may have le a gap in the market for heated tobacco.
The types of heated tobacco devices produced continues to change over time. While carbon-tip HTPs such as Eclipse were once the only type available, electronic devices such as IQOS and glo now dominate the market. These products could di er in their safety. It is possible that using newer electronic products, such as those that heat tobacco through induction, could lead to di erent exposures than those reported here. Therefore, it is important to continue tracking the research into new developments in heated tobacco technology.
All studies on safety that we included were funded by tobacco companies. These companies have a financial incentive to produce results that are favourable towards the products they sell. Data from independent sources are, therefore, needed to confirm the results reported in this review. We cannot rule out the possibility of publication bias.
Safety data came from studies that used optimised settings for switching to exclusive HTP use. Six of the 11 RCTs had an extended period where participants stayed in a clinic, preventing those in the HTP group from easily accessing cigarettes (and vice versa). This means that, while trial data consistently show reduced exposure in people completely substituting HTPs for cigarettes, it remains unclear how exposure changes in people using HTPs in real-world settings where they have greater access to cigarettes.
Serious adverse events were rare as safety data came from studies where participants used heated tobacco for one year at most (median of 13 weeks). Trials with larger samples and longer followup periods are likely needed to establish how switching from cigarettes to heated tobacco a ects rates of these events.
Biomarker studies assessing exposure to toxicants and carcinogens are only relevant if reducing exposure prevents disease and premature death. Animal studies have shown a dose-response relationship between some exposures, such as nitrosamines, and cancer development, suggesting reduced exposure may indeed reduce disease incidence (Frank 2007 Cochrane Database of Systematic Reviews we only reported on biomarkers for a sample of the toxicants and carcinogens present in cigarette smoke or heated tobacco aerosol. Previous reviews found similar reductions in exposure to a broader range of potentially harmful chemicals among those switching from cigarettes to heated tobacco (Simonavicius 2018; Znyk 2021).

Quality of the evidence
We considered the certainty of evidence for e ectiveness and safety of heated tobacco compared with cigarette smoking, tobacco abstinence, and snus use, along with population-level data on smoking prevalence and cigarette sales (Summary of findings 1; Summary of findings 2; Summary of findings 3; Summary of findings 4).
Summary of findings 1; Summary of findings 2; and Summary of findings 3 show evidence from RCTs. Reasons for downgrading certainty of evidence included: risk of bias, when most studies pooled were judged at unclear or high risk of bias; imprecision, when confidence intervals were wide and included no di erence; inconsistency, when heterogeneity was high and unexplained; and indirectness, when all the studies pooled used carbon-tip HTPs, which di er substantially from the electronic devices currently on the market.

E ectiveness
We remain uncertain about the e ectiveness of HTPs for smoking cessation, as no studies assessed this.

Safety
For all comparisons, e ect estimates for adverse events or serious adverse events were of low or very-low certainty, mainly due to imprecision. This means that we remain uncertain about the direction and size of e ects. None of the analyses found serious adverse events that were judged to be caused by HTPs or comparators. For the selected biomarker outcomes NNAL and COHb, evidence was moderate certainty when the comparison was with cigarette smoking; moderate or very-low certainty compared with tobacco abstinence, respectively; and low or verylow certainty compared with snus use. This means we are more confident about the e ects of heated tobacco on biomarkers relative to cigarettes than to tobacco abstinence or snus.

Smoking prevalence
Summary of findings 4 shows evidence from time-series studies investigating smoking prevalence or cigarette sales. We remain uncertain about the impact of rising heated tobacco use on smoking prevalence, as no studies directly assessed this. There was very low-certainty evidence for an impact on cigarette sales, meaning our confidence in results is limited. We downgraded certainty one level for risk of bias, as the studies were considered at moderate or serious risk of bias. We also downgraded certainty one level for the indirectness of cigarette sales as a proxy for smoking prevalence. This is because falls in cigarette sales do not necessarily translate to reductions in smoking prevalence; people can reduce the number of cigarettes they smoke rather than stopping smoking entirely.

Potential biases in the review process
We took several steps to ensure the review process was robust. We followed standard methods used by the Cochrane Tobacco Addiction Review Group. Our search strategy included a broad range of databases, including the Cochrane Tobacco Addiction Group Specialised Register. We also contacted researchers who have worked on relevant reports by charities or public health bodies to capture studies that we may have otherwise missed. We followed standard Cochrane practice of requiring two review authors to independently screen studies, extract data, and assess risk of bias. None of the authors of this review were also authors of included studies.

Agreements and disagreements with other studies or reviews
Our results were similar to those from an earlier systematic review by Simonavicius 2018, which concluded that HTPs expose "users and bystanders to toxicants, although at substantially lower levels than cigarettes" and noted the lack of studies without links to the tobacco industry. There were analogous results in the Public Health England report into HTPs (McNeill 2018). Our current review di ers from these reports because it only uses safety data from RCTs with at least one week of follow-up. In addition, it includes several studies published between 2018 and 2021 and adds analysis of time-series studies.
One systematic review by Jankowski 2019 examined data from a wide range of study types, including those using animals and cellular models and those examining the chemical composition of heated tobacco aerosol. Because of these less stringent inclusion criteria, their search identified a greater number of studies than our review (97 versus 16). Nonetheless, they found similar results: "in vitro and in vivo assessments of HTP aerosols revealed reduced toxicity, but these were mainly based on studies sponsored by the tobacco industry". They also reported that exposure to toxicants is likely higher in HTP users compared with those not using any tobacco product.
One more recent systematic review by Znyk 2021 found that, as we did, there was no evidence on the e ectiveness of HTPs for smoking cessation. Their results into the toxicology of HTPs also aligned with ours and with those from the aforementioned reviews.
Finally, prior to the US FDA allowing marketing of IQOS as a "reduced exposure" tobacco product in the US, it reviewed evidence into the safety of these products relative to cigarettes (FDA 2019). This review concluded that "switching completely from conventional cigarettes to the IQOS system significantly reduces your body's exposure to harmful or potentially harmful chemicals" (FDA 2020). It also emphasised that "the evidence is not su icient to demonstrate substantiation of either of the claims about reduced risk of tobacco-related disease or harm". These statements align with our conclusions about the overall completeness of results.

Implications for practice
No studies reported on the use of heated tobacco for cigarette smoking cessation, so their e ectiveness for this purpose remains uncertain. There was insu icient evidence for di erences in risk of adverse or serious adverse events between people randomised to use heated tobacco products (HTPs) or to smoke cigarettes, attempt abstinence, or use snus, but participants only used these Library Trusted evidence. Informed decisions. Better health.
Cochrane Database of Systematic Reviews for a very short time. However, there was moderate-certainty evidence that users of heated tobacco have lower exposure to selected toxicants and carcinogens than cigarette smokers, and very low-to moderate-certainty evidence of higher exposure than those attempting abstinence from all tobacco.
The rate of decline in cigarette sales accelerated a er the introduction of heated tobacco to market in Japan but, as data were observational, it is possible other factors caused these changes. Moreover, falls in cigarette sales may not translate to declines in smoking prevalence, and changes in Japan may not generalise elsewhere.

Implications for research
Studies from independent sources are needed that attempt to replicate the randomised controlled trials (RCTs) on safety included in this review -all of which were funded by tobacco companies.
Users are likely to continue using HTPs for a prolonged period, so studies should allow for this and build in long-term followup. Studies are also needed to determine how rates of adverse and serious adverse events di er between those randomised to use heated tobacco, continue smoking cigarettes, or use another treatment. Ideally, studies that measure serious adverse events should be powered on this outcome, which is relatively rare, but of key clinical and policy importance. Further studies should measure how biomarkers of exposure and harm di er across groups, especially for new devices. In the longer-term, large cohort studies and RCTs are needed to examine how long-term switching from smoking to heated tobacco a ects disease incidence and death. If HTPs are determined to be substantially less harmful than traditional cigarettes, RCTs will be needed into their use for cigarette smoking cessation, preferably following up participants for at least six months.
Our literature searches only found population-level studies examining cigarette sales rather than smoking prevalence. Future research is needed to determine whether the increased rate of decline in cigarette sales following the launch of heated tobacco in Japan translated to similar changes in smoking prevalence trends. Furthermore, to assess whether results generalise, studies need to be conducted in other countries that have also seen substantial growth in heated tobacco use.

Additional references Anderson 2008
Anderson SJ, Ling PM."And they told two friends…and so on": RJ Reynolds' viral marketing of Eclipse and its potential to mislead the public. Cochrane Database of Systematic Reviews

FDA 2020
US Food and Drug Administration.FDA authorizes marketing of IQOS tobacco heating system with 'reduced exposure' information. www.fda.gov/news-events/press-announcements/ fda-authorizes-marketing-iqos-tobacco-heating-systemreduced-exposure-information (accessed prior to 21 November 2021).  • Medical conditions that were a safety concern or would interfere with study • BMI < 18.5 or ≥ 32 kg/m 2 • Use of nicotine-containing products other than cigarettes in 6 weeks prior to admission • Use of drugs likely to affect CYP1A2 or CYP2A6 activity within 14 days or 5 half-lives of the drug (whichever was longer) before admission • Current or past alcohol problems • Positive urine drug test • Given or received blood in 3 months prior to admission • Current or past employee of the tobacco industry or their close relatives • Pregnant or breastfeeding • People of childbearing potential who do not agree to use contraception

Interventions
Randomised ( Tobacco industry funded: quote: "Philip Morris International is the sole source of funding and the sponsor of this study".

Author conflicts of interest
Quote: "All authors are employees of Philip Morris International (PMI) or work for PMI under contractual agreements".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Low risk Protocol mentions that: "At the end of the baseline period enrolled subjects will be randomized using an interactive web and voice response system (IxRS) on day −1 at any time during the day".
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological and assessors were blinded to randomised group allocation.
Incomplete outcome data (attrition bias) All outcomes Low risk Low attrition in both groups, 76/80 participants in heated tobacco arm and 39/40 in cigarette arm completed final follow-up.
Selective reporting (reporting bias) Low risk All preregistered outcomes reported.

Funding source
No funding from the tobacco industry: quote: "K.M.C. and G.J.N. receive funding support from grants from the US National Cancer Institute (P01 CA200512, P30 CA138313)".

Author conflicts of interest
Quote: "K.M.C. has been a consultant and received grant funding from Pfizer, Inc. in the past five years. K.M.C. has also been a paid expert witness in litigation against the cigarette industry. D.T.S. does not accept money from any entity with a financial interest in promoting any tobacco or nicotine product, nor from any organization that promotes an abstinence-only position on nicotine and tobacco products".

Cochrane Database of Systematic Reviews
• Agreed to refrain from alcohol for 24 hours before study visits • No intention to quit smoking Exclusion criteria • Medical conditions that were a safety concern or would interfere with study • BMI < 17.6 or ≥ 32.0 kg/m 2 • Bodyweight < 50 kg for men or < 40 kg for women • Use of nicotine-containing products other than cigarettes in 14 days prior to screening • Use of drugs likely to interfere with study • Current or past alcohol or drug problems • Self-report not inhaling smoke from cigarettes into lungs

Funding source
Tobacco industry funded: quote: "The study was supported by British American Tobacco (Investments) Limited".

Author conflicts of interest
Quote: "All authors are current employees of British American Tobacco (Investments) Limited".

Bias Authors' judgement Support for judgement
Gale 2020 (Continued) Heated tobacco products for smoking cessation and reducing smoking prevalence (

Cochrane Database of Systematic Reviews
Random sequence generation (selection bias) Low risk Randomised using (quote): "blocks of computer generated random number sequences in PROC PLAN".

Allocation concealment (selection bias)
Unclear risk No details on whether random sequence was concealed from investigators.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.
Incomplete outcome data (attrition bias) All outcomes Low risk Similar attrition in both groups, with 127/197 participants in heated tobacco arm and 59/79 in cigarette arm completing the final follow-up.
Selective reporting (reporting bias) Low risk All preregistered outcomes reported. Tobacco industry funded: quote: "Philip Morris International is the sole source of funding and sponsor of this project".

Author conflicts of interest
Quote: "All authors are employees of Philip Morris International".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Low risk Randomised using an interactive voice response system.
Blinding of outcome assessment (detection bias) Low risk All outcomes were biological. Cochrane Database of Systematic Reviews

All outcomes
Incomplete outcome data (attrition bias) All outcomes High risk Attrition differed substantially across groups, with 59% participants in heated tobacco arm, 78% in cigarette arm, and 23% in abstinent arm completing final 90-day follow-up -with others removed due to protocol violations, non-compliance, or discontinuation.
Selective reporting (reporting bias) Low risk All preregistered outcomes reported.

Other bias Unclear risk
Reported that there were more people in the 90-day follow-up sample than at baseline. Behavioural support: 5 days in a clinic setting, where product use was monitored. Record use of nicotine and tobacco products in electronic diary. Carbon monoxide breath tests were used to assess compliance

Study characteristics
Instructions for smoking cessation/switching: switch entirely to menthol THS 2.2 for study period

Cigarette smoking arm
Behavioural support: record use of nicotine and tobacco products in electronic diary Instructions for smoking cessation/switching: not advised to stop smoking

Abstinence arm
Behavioural support: record use of nicotine and tobacco products in electronic diary. Carbon monoxide breath tests used to assess compliance. Nicotine replacement therapy was allowed, but not provided

Author conflicts of interest
Quote: "All authors are employees of Philip Morris Products S.A.".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Unclear risk Randomised was performed using an interactive voice response system.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.
Incomplete outcome data (attrition bias) All outcomes Low risk Low attrition reported across all groups, with 97% of participants who were randomised the heated tobacco, 98% to cigarette smoking, and 95% to abstinence arms completing the final follow-up.

Cochrane Database of Systematic Reviews
Selective reporting (reporting bias) High risk Did not report FVC outcomes, which were preregistered. Tobacco industry funded: Quote: "PMI is the sole source of funding and sponsor of this project".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Low risk Randomised using an interactive voice response system.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.
Incomplete outcome data (attrition bias) All outcomes Low risk Low attrition in both groups (15.1% in heated tobacco arm, 10.5% in cigarette smoking arm).
Selective reporting (reporting bias) Low risk All preregistered primary outcomes reported. Cochrane Database of Systematic Reviews Quote: "All authors are or were Philip Morris International (PMI) R&D employees or worked for PMI R&D under contractual agreements".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias) Unclear risk No details on random sequence generation.

Allocation concealment (selection bias)
Low risk Randomised using an Interactive Voice Response System.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.
Incomplete outcome data (attrition bias) All outcomes Low risk Low attrition across both groups, with 99% of participants randomised to heated tobacco use and 95% to cigarette smoking remaining in the study until the final follow-up.

Unclear risk
No protocol or analysis plan was registered online.

Funding source
Tobacco industry funded: quote: "PMI is the sole source of funding and sponsor of this project".

Author conflicts of interest
Quote: "The work reported in this publication involved a candidate reduced risk product developed by PMI Research & Development. All authors are employees of PMI".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Low risk Randomised using an interactive web/voice response system.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.
Incomplete outcome data (attrition bias) All outcomes Low risk Attrition was low across all groups, with only 1 participant failing to complete follow-up in each study arm.

NCT03364751 (Continued)
Heated tobacco products for smoking cessation and reducing smoking prevalence (

Cochrane Database of Systematic Reviews
Selective reporting (reporting bias) Low risk Protocol was preregistered and all outcomes were reported.
Other bias High risk Results analysed based on actual product use, rather than product participants were randomly allocated to use.

Cochrane Database of Systematic Reviews
Other instructions and details: provided with heated tobacco sticks, with choice of menthol or nonmenthol products Behavioural support: provided information about HTP. Compensated for recording use of nicotine and tobacco products in interactive voice recording system diary Instructions for smoking cessation/switching: switch entirely to Eclipse for study period

Snus arm
Other instructions and details: provided snus with 3 flavour options (spice, original, and frost) Behavioural support: provided information about snus. Compensated for recording use of nicotine and tobacco products in interactive voice response system diary Instructions for smoking cessation/switching: switch to snus use for study period

Cigarette smoking arm
Other instructions and details: provided with ultra-low machine yield cigarettes, with choice of menthol or non-menthol cigarettes

Author conflicts of interest
Quote: "All authors are current employees of RAI Services Company or R.J. Reynolds Tobacco Company".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Low risk Randomised using an Interactive Voice Response System.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.

Ogden 2015 (Continued)
Heated tobacco products for smoking cessation and reducing smoking prevalence (

Funding source
No funding from the tobacco industry: quote: "This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors".

Author conflicts of interest
Quote: "None declared". Tobacco industry funded: quote: "The work reported in all eight parts of this supplement was funded by PMI R&D".

Author conflicts of interest
Quote: "All authors are or were Philip Morris International (PMI) R&D employees or worked for PMI R&D under contractual agreements".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Unclear risk No information.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.
Incomplete outcome data (attrition bias) All outcomes

Unclear risk No information on attrition.
Selective reporting (reporting bias)

Unclear risk
No trial registration or analysis plan. Tobacco industry funded: quote: "The work reported in all eight parts of this supplement was funded by PMI R&D".

Author conflicts of interest
Quote: "All authors are or were Philip Morris International (PMI) R&D employees or worked for PMI R&D under contractual agreements".

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Unclear risk
No details on random sequence generation.

Allocation concealment (selection bias)
Unclear risk No information.
Blinding of outcome assessment (detection bias) All outcomes Low risk All outcomes were biological.
Incomplete outcome data (attrition bias) All outcomes

Unclear risk
No information on attrition.
Selective reporting (reporting bias) Unclear risk No trial registration or analysis plan. Tobacco industry funded: quote: "The work reported in all eight parts of this supplement was funded by PMI R&D".

Author conflicts of interest
Quote: "All authors are or were Philip Morris International (PMI) R&D employees or worked for PMI R&D under contractual agreements".

Risk of bias
Bias Library Trusted evidence. Informed decisions. Better health.

Study Reason for exclusion
Franzen 2020 Wrong study design as it only assessed acute effects of heated tobacco.

Gale 2017
Wrong study design as participants used heated tobacco product for < 7 days.

Haziza 2016b
Wrong study design as participants used heated tobacco product for < 7 days.

Ioakeimidis 2021
Wrong study design as it only assessed acute effects of heated tobacco.

Lee 2020
Wrong study design and outcome as formal interrupted or multiple time-series not used to assess change in trends in smoking prevalence or cigarette sales.

Martin 2016
Wrong study design as participants used heated tobacco product for < 7 days.

Pataka 2019
Wrong study design, as it only measures acute effects of heated tobacco use.
Tran 2020 Wrong study design as participants used heated tobacco product for < 7 days.

Yuki 2018
Wrong study design as participants used heated tobacco product for < 7 days.

Characteristics of ongoing studies [ordered by study ID]
Study name Non-inferiority trial comparing cigarette consumption, adoption rates, acceptability, tolerability, and tobacco harm reduction potential in smokers switching to heated tobacco products or electronic cigarettes: study protocol for a randomized controlled trial The study protocol was written by EM who was also the principal investigator of the study. Philip Morris Products SA had no role in the design of the study protocol and will not have any role during its execution, analysis, data interpretation or writing of the manuscript".

Author conflicts of interest
Quote: "EM, DS and RP are full-time employee of the University of Catania, Italy. PC, MC and RE are fixed-term researcher at University of Catania, Italy. MM is fixed-term researcher at Centro per la Prevenzione e Cura del Tabagismo, University of Catania. BB is full-time employee of ARNAS Garibaldi, Catania, Italy. AP is full-time employee of Casa di Cura Musumeci-Gecas, Gravina di Catania, Italy. UP is full-time employee of Ospedale "San Vincenzo" -Taormina, Italy. In relation to his work in the area of tobacco control, RP has received lecture fees and research funding from Pfizer and GlaxoSmithKline, manufacturers of stop smoking medications. He has also received sup-Caponnetto 2020 (Continued) Heated tobacco products for smoking cessation and reducing smoking prevalence (

Funding source
Assumed to be tobacco industry funded, as PMI is the study sponsor.

Author conflicts of interest
No conflicts of interest statement available.

NCT03837704 (Continued)
Study name Effect of switching from cigarette smoking to IQOS on exercise capacity Cochrane Database of Systematic Reviews • No plan to quit smoking within next 6 months Exclusion criteria • Medical conditions that were a safety concern or would interfere with study • BMI < 18.5 or > 32.0 kg/m 2 • Use of nicotine-containing products other than menthol cigarettes in 4 weeks prior to screening • Performs < 45 minutes of vigorous exercise per week • Use drugs likely to interfere with study • Current or past alcohol problems • Positive urine drug test • Pregnant or breastfeeding • People of childbearing potential who do not agree to use contraception • Current or past employee of tobacco industry (or their close relatives)

Interventions
Randomised to use an electronic HTP (with or without an exercise program), continue smoking cigarettes, or become abstinent

Heated tobacco with exercise training arm
Device heating method: electronic Assumed to be tobacco industry funded, as PMI is the study sponsor.

Author conflicts of interest
No conflicts of interest statement available.

Library
Trusted evidence. Informed decisions. Better health.