REF-1463 · Tattan-Birch et al., 2022 · Cochrane Database of Systematic Reviews 2022, CD013790 · Paper · PubMed · Free full text

Does heated tobacco spare a cigarette's CO?

COHb was lower after switching in 9 trials, all tobacco-funded: about half lower overall, about a fifth with 4+ weeks' follow-up (our back-calculation). A marker; no exercise tested.

Main caveat: All 11 RCTs were funded by tobacco companies; no independent replication; publication bias not ruled out by the authors.
fairly strong evidenceMeasured on non-athletes and recreational people

Tattan-Birch and colleagues reviewed heated tobacco products for cessation, safety and smoking prevalence, for the Cochrane Tobacco Addiction Group. Thirteen completed studies were included: 11 RCTs in 2,666 people, median follow-up 13 weeks, and two time-series studies. All RCTs were funded by tobacco companies; eight were judged at unclear and three at high risk of bias.

No study reported smoking cessation. Against continued smoking, COHb was lower with heated tobacco (log mean difference −0.74; 9 studies, 1,807 people; I² 96%; moderate certainty, limited by risk of bias) and every study pointed the same way, though the sensitivity table shows −0.24 in the six trials with at least 4 weeks' follow-up.

Exhaled CO was 9.13 ppm lower (3 studies, 1,322; I² 4%). Against attempted abstinence, COHb was higher in intention-to-treat analyses (+0.69; 3 RCTs, 212) and not different per protocol (−0.32; 2 RCTs, 170), very low certainty. Adverse event comparisons were inconclusive.

Authors' declared interests: no tobacco or e-cigarette funding; two authors declare funding from makers of smoking cessation medication (Pfizer, Johnson & Johnson). Our reading: for the session question, "less carbon monoxide than cigarettes" stands, from industry-funded trials; "as low as not using" is not settled; and the review says nothing about exercise.

Protocol

Search: Cochrane Tobacco Addiction Group register, CENTRAL, MEDLINE and six other databases to January 2021, reference checks, contact with authors.
Eligible trials: smokers randomised to exclusive heated tobacco use or a control (continued smoking, attempted abstinence, snus); time-series studies on prevalence or sales.
Outcomes: abstinence, adverse and serious adverse events, sales or prevalence; biomarkers of exposure including NNAL and COHb.
Analysis: random-effects models; mean differences on the log scale (LMD); GRADE.
Funding and interests (as printed): review funded through university posts and public or charity studentships (University College London, King's College London, Oxford; Public Health England, Cancer Research UK, NIHR Maudsley BRC, UK Prevention Research Partnership); "None of the authors have received cash or kind, hospitality, or any subsidy from manufactures of tobacco products or electronic cigarettes." Also declared: LS has received honoraria, an unrestricted research grant and travel expenses from Pfizer and an advisory-panel honorarium from Johnson & Johnson; JB has received unrestricted research funding from Pfizer and Johnson & Johnson, makers of smoking cessation medication.The included RCTs are all tobacco-company funded.

Key Results

COHb, heated tobacco vs continued smokingn (studies)LMD (95% CI)I²
All data1807 (9)−0.74 (−0.97 to −0.52; abstract and summary table print −0.92)96%
No high risk of bias1611 (7)−0.76 (−1.07 to −0.44)97%
Only electronic devices1659 (7)−0.84 (−1.07 to −0.60)96%
At least 4 weeks' follow-up1511 (6)−0.24 (−0.36 to −0.12)95%
What the log difference means: −0.74 is about 52% lower and −0.24 about 21% lower than the smoking arm, by back-transformation (our arithmetic). The text calls the sensitivity results "similar"; the direction is, the size is not.
Exhaled CO vs continued smoking: −9.13 ppm (−10.49 to −7.78), 1322 people, 3 studies, I² 4%.
COHb vs attempted abstinence (very low certainty, reported narratively): intention to treat +0.69 (0.07 to 1.31), 212 people, 3 RCTs; per protocol −0.32 (−1.04 to 0.39), 170 people, 2 RCTs; all data 0.30 (−0.40 to 1.00), 382 people, 5 studies.
Adverse events vs continued smoking: RR 1.03 (0.92 to 1.15), 6 studies, 1713 people; inconclusive by the authors' judgement.
Numbers that do not add up: the abstract and summary-of-findings table give the COHb interval as −0.92 to −0.52; the Results, the sensitivity table and Analysis 1.7 give −0.97 to −0.52.

Mechanisms

Measured: lower blood and breath CO markers after switching, in the trial setting.
Authors' interpretation: moderate-certainty lower exposure to selected toxicants than smoking, very low- to moderate-certainty higher exposure than abstinence; independent replication is needed; real-world exposure is unclear because six of 11 RCTs held people in a clinic with limited access to cigarettes.
Inference: the smaller difference with longer follow-up could reflect less confinement and more cigarette use in the switch arm, or the different trials. The review does not test this. Heating without combustion would be expected to make less CO, but the marker shows only what was measured here.

Practical Application

Does heated tobacco spare a cigarette's CO? COHb was lower after switching in 9 trials, all tobacco-funded: about half lower overall, about a fifth with 4+ weeks' follow-up (our back-calculation). A marker; no exercise tested.
Before a session: supports "less CO than a cigarette", and does not support "no cost": nicotine and airway effects are separate (REF-1464, REF-1471, REF-1473). Not from this paper; the base's own practice: any hour window or race-morning advice is practice.
When an athlete asks if it is clean: say what was measured (biomarkers, people switching from cigarettes, industry-funded trials) and what was not (session, race, fitness).
What it does not show: performance, heart rate at work or fitness; whether heated tobacco helps people stop smoking; long-term effects. The reference list includes a registered trial on IQOS and exercise capacity (NCT03887117); its results were not in this review.

Limitations

All 11 RCTs were funded by tobacco companies; no independent replication; publication bias not ruled out by the authors.
Biomarkers, not performance or health; short follow-up (median 13 weeks).
Very high heterogeneity on COHb (I² 96%), and the size depends on follow-up length.
Optimised settings: six of 11 RCTs had a clinic stay, so real-world exposure is unclear.
Mixed devices, including carbon-tip products, and the market has moved on.
Search to January 2021; Crossref finds no later version (no .pub3, checked 2026-10-30), so newer trials are not in it.
Who was measured: adult smokers, not athletes.
How this entry was checked and changed
Why this level

a systematic review of RCTs (rubric A) held one step down: I² 96% on COHb, eight RCTs at unclear and three at high risk of bias, all RCTs industry-funded, and the outcome is a biomarker, not performance or health

Full text (checked 2026-10-30)
We included 13 completed studies, of which 11 were RCTs assessing safety (2666 participants) and two were time-series studies; eight RCTs 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 (full text)
No studies reported smoking cessation outcomes (full text)
COHb: LMD −0.74, 95% CI −0.97 to −0.52; 1807 participants; I2 = 96% (Results); LMD −0.74, 95% CI −0.92 to −0.52 (abstract) (full text)
Sensitivity table, COHb: No high risk of bias 1611 (7), −0.76 (−1.07 to −0.44); Only electronic devices 1659 (7), −0.84 (−1.07 to −0.60); ≥ 4 weeks' follow-up 1511 (6), −0.24 (−0.36 to −0.12) (full text)
Exhaled CO: MD −9.13ppm, 95% CI −10.49 to −7.78; 1322 participants; I2 = 4% (full text)
COHb vs abstinence: 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 low (full text)
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) (full text)
None of the authors have received cash or kind, hospitality, or any subsidy from manufactures of tobacco products or electronic cigarettes (full text)