Summary: A systematic review and meta-analysis finds that controlled breathing interventions significantly reduce substance cravings across a range of drug classes, including alcohol, nicotine, opioids, stimulants and polysubstance use.
This synthesis pooled data from 20 studies involving 1,249 participants to evaluate the effectiveness of various breathwork approaches. Overall, breathing-based interventions produced modest but statistically significant reductions in craving. Single-session, non–heart rate variability biofeedback (non‑HRVB) breathing techniques tended to deliver immediate craving relief, whereas HRVB approaches—using heart-rate feedback—showed stronger effects when delivered across multiple sessions for longer-term craving reduction.
Key Facts
- Craving reduction: Meta-analysis of 20 trials (N = 1,249) identified a small but consistent effect on lowering substance craving.
- Protocol differences: Non‑HRVB techniques (for example, slow-paced or resonant breathing) were most effective in providing rapid, single-session relief. HRVB, which requires heart-rate sensors, demonstrated greater benefit across repeated sessions targeting sustained craving control.
- Across substances: Positive effects were consistent across nicotine, alcohol, opioids, stimulants and polysubstance use.
- Dependence and withdrawal: Evidence for withdrawal and dependence outcomes is limited by few studies. Effects on dependence were large in the small number of trials available, indicating the need for further investigation.
- Scalability: Because breathing interventions are low cost and require minimal resources, smartphone-based delivery has high potential to scale globally as an adjunct to conventional addiction treatment.
Source: Society for the Study of Addiction
Controlled breathing interventions can reduce cravings in people who use substances, including alcohol, tobacco and other drugs, according to a new review published in the journal Addiction.
The review combined evidence from 20 studies that measured outcomes such as craving, withdrawal symptoms and measures of dependence in people using alcohol, nicotine and other drugs. The interventions varied in format and duration but shared a primary focus on paced or structured breathing.
Breathing approaches evaluated included heart-rate variability biofeedback (HRVB), resonant breathing, yogic breathing sequences, rhemercise (slow-paced breathing with behavioural elements), smoking-mimic breathing patterns and three-part breathing techniques. Across interventions, the strongest and most consistent evidence supported craving reduction, with pattern differences by protocol type and delivery format.
Non‑HRVB methods—such as fixed-rate resonant breathing (typically around six breaths per minute)—showed reliable immediate effects useful for single-session or momentary craving management. HRVB protocols, which require heart-rate sensors to provide real-time feedback and guide breathing to align with the individual’s cardiac rhythm, showed greater promise in multi-session programs targeting sustained reductions in craving.
Results were consistent across substance types. For withdrawal symptoms and dependence, the evidence base was smaller: four studies evaluated withdrawal and two examined dependence. Findings were mixed for withdrawal, while the observed effect size for dependence was relatively large in the limited data available—underscoring the need for larger, pre-registered clinical trials with standardized outcomes and longer follow-up.
Lead author Dr Ariel Dart Roxburgh (Monash University) notes that controlled breathing is a low-cost, scalable option that could augment existing addiction treatments. Given global smartphone ownership, digital delivery of guided breathing protocols could extend access to evidence-based coping tools. Future research priorities include larger trials, standardized outcome measures and mechanistic studies to clarify how breathwork produces clinical benefit.
- Heart‑rate variability biofeedback (HRVB): Uses a sensor to track heart rate and provides real‑time feedback so users can pace breathing to align with their cardiac rhythm.
- Resonant breathing: A paced breathing method approximating the HRVB target rhythm (commonly around six breaths per minute) that can be practiced without sensors.
- Yogic breathing: Structured breathing patterns that vary in depth, speed and retention (for example, practices that cycle through different rhythms).
- Rhemercise: Slow-paced breathing combined with behavioural elements like smiling, yawning and positive verbalization coordinated with breaths.
- Mimicking smoking: Deep, timed inhalations and holds designed to replicate the physical sensations of smoking without nicotine.
- Three-part breathing: A sequence of breaths that emphasize expansion of different parts of the torso (abdomen, lower chest, upper chest) to enhance bodily awareness.
Primary funding: The review itself was not directly funded. Two authors are supported by NHMRC Fellowships (GNT1196892 and GNT2008193).
Declaration of interests: The authors reported no competing financial or non-financial interests related to this project.
Key Questions Answered
A: HRVB uses heart-rate sensors to synchronize breathing with cardiac rhythm and is best delivered across multiple sessions for lasting change. Non‑HRVB approaches use fixed pacing (for example, six breaths per minute) without sensors and are effective for immediate, single-session craving relief.
A: Yes. The review found consistent craving-reduction effects across nicotine, alcohol, opioids, stimulants and polysubstance use.
A: Breathwork requires no medication and minimal resources, so delivering evidence-based protocols via smartphone apps can expand access and provide low-cost, scalable support alongside conventional treatment.
Editorial Notes
- This article was edited by a Neuroscience News editor.
- The full journal paper was reviewed.
- Additional contextual information was added by staff editors.
About this addiction research news
Author: Jean O’Reilly
Source: Society for the Study of Addiction
Contact: Jean O’Reilly – Society for the Study of Addiction
Image credit: Neuroscience News
Original research (open access): “Controlled breathing and heart rate variability biofeedback for substance use-related outcomes: A systematic review and meta-analysis” by Ariel Dart Roxburgh et al., Addiction. DOI: 10.1111/add.70542
Abstract
Controlled breathing and heart rate variability biofeedback for substance use-related outcomes: A systematic review and meta-analysis
Background and aim
Substance use disorders impose substantial social and economic burdens worldwide. Breathing-based interventions are low-cost, scalable options that may complement or augment traditional treatments. This review aimed to evaluate the effectiveness of controlled breathing and HRVB interventions for substance use-related outcomes, including craving, withdrawal and dependence.
Methods
Multiple databases were searched through 14 November 2025. Eligible studies were quantitative, included repeated measures and reported at least one relevant substance use outcome. Interventions were required to center on controlled breathing and were classified by whether HRVB was used. The review adhered to PRISMA guidelines and used random-effects meta-analysis to estimate pooled effects.
Results
Of 3,502 records screened, 20 studies met inclusion criteria and 15 were included in meta-analyses. Breathing interventions reduced craving compared with control conditions with a small but statistically significant pooled effect (Hedges’ g = 0.30, 95% CI = 0.12–0.48; I2 = 23.5%; 12 studies; 696 participants). Effects on withdrawal were uncertain (g = 0.46, 95% CI = −0.28 to 1.19; I2 = 89.7%; 4 studies; 289 participants). For dependence, two studies yielded a larger pooled effect (g = 0.85, 95% CI = 0.21–1.48; I2 = 70.9%; 144 participants), though data were limited. No overall difference in craving outcomes was observed between HRVB and non‑HRVB approaches.
Conclusions
Breathing-based interventions appear to produce small but reliable reductions in craving among people who use substances, supporting their potential as a low-cost, scalable strategy to support recovery. Further large, pre-registered trials with standardized measures and longer follow-up are needed to clarify effects on withdrawal, dependence and the mechanisms driving benefit.