In short

  • A randomized trial in regular drinkers with AFib found fewer recurrences and lower burden after a large reduction in alcohol.1
  • In people with paroxysmal AFib, recent drinking was associated with higher odds of a recorded episode. This does not give an individual probability or a safe waiting time.2
  • Long-term studies disagree at low intake. They do not establish that drinking protects against AFib or identify a safe personal dose.4,5,6

Reading the evidence

One topic. Four different questions.

Recurrence

Does AFib return?

A randomized reduction trial in regular drinkers with AFib.

Burden

How much time in AFib?

A separate outcome in the same trial—not another replication.

Acute association

Does drinking precede an episode?

Short-term, within-person comparisons in paroxysmal AFib.

Incident AF

Who first develops AFib?

Long-term population studies; low-intake results differ.

Schematic distinctions, not measurements. Evidence about a first diagnosis cannot automatically answer questions about episodes after diagnosis.

Does cutting back change AFib?

The strongest direct test enrolled 140 adults with paroxysmal or persistent AFib who were in sinus rhythm at entry and drank at least ten study drinks a week. They were randomly assigned to abstinence or usual drinking for six months.1

Randomized trial · six months

AFib returned in 53% versus 73%

140 regular drinkers with established AFib; 70 in each group. Recurrence was counted after the first two weeks.

20 percentage pointsFewer people with a recurrence in this trial
What changed—and a different outcome
GroupMean drinks/week
before → during		Median burden
Assigned to abstinence16.8 → 2.10.5%IQR 0–3.0%
Usual drinking16.4 → 13.21.2%IQR 0–10.3%

Burden means estimated percentage of time in AFib. These are follow-up medians in two groups, not a before-and-after change. IQR describes the middle half of participants, not uncertainty around the treatment effect.

Voskoboinik and colleagues, 2020. The abstinence goal led to substantial reduction, not universal abstinence. Time-to-recurrence HR 0.55; 95% CI 0.36–0.84. Mixed continuous and intermittent ECG monitoring limits completeness.1

The intervention was a large reduction, with complete abstinence as its goal. It did not compare several low doses to find a threshold. Its time-to-recurrence hazard ratio was 0.55 (95% confidence interval 0.36–0.84); this is different from the absolute difference shown above.1

The trial was small, open label and mostly male. Participants were willing to attempt abstinence; alcohol dependence was excluded. Rhythm monitoring combined implanted devices with intermittent ECG recordings. The result supports a benefit in the group studied, without guaranteeing that an individual's AFib will stop.

Can a drink precede an episode?

A prospective case-crossover study followed 100 people with paroxysmal AFib for four weeks. Continuous ECG monitoring captured rhythm, participants marked drinks, and a skin sensor measured alcohol exposure. Each person served as their own comparison.2

Within-person association · not a randomized trial

Higher odds after recent drinks

100 people with paroxysmal AFib, followed for four weeks. Drinks in the preceding four hours were compared with no recent drinks.

Odds ratio · ECG-confirmed AFib episode

1 = no association. Dots are estimates; lines show 95% confidence intervals. The ratio axis uses equal spacing for doubling.

Marcus and colleagues, 2021. “Twice the odds” does not mean a 50% chance or twice the absolute probability. The wider interval for two or more drinks shows greater uncertainty. These results do not identify an individual's risk or a safe time after drinking.2

The sensor analysis also linked exposure over the preceding twelve hours to episodes. These are study windows, not a countdown until alcohol becomes safe. Drinking was not randomized: sleep, meals and other circumstances can change alongside it. The results cannot tell how likely a particular drink is to precede an episode in one person.2

What trigger testing adds

I-STOP-AFib used randomized weeks of exposure or avoidance for participants' chosen triggers. Forty-three people initially chose alcohol. Across all testing periods, the alcohol analysis based on assigned weeks was inconclusive: odds ratio 1.17, with a 95% Bayesian credible interval of 0.81–1.72.3

When researchers instead analyzed reported exposure, alcohol was associated with more reported AFib: odds ratio 1.77 (1.20–2.69). That analysis gives up some protection from randomization. Unblinded reporting, adherence and attrition matter. The trial's main quality-of-life outcome did not improve.3 This adds qualified support; it is not a prescription to test alcohol as a trigger.

Why long-term findings differ

Population studies ask who develops AFib over years—called incident AF. That differs from asking whether drinking precedes an episode or whether reducing it changes recurrence after diagnosis.

In pooled European cohorts of 100,092 people initially without AFib, 5,854 developed it over a median 13.9 years. Each additional 12 g of usual daily intake was associated with a hazard ratio of 1.16 (95% confidence interval 1.11–1.22). That is a relative association, not sixteen extra cases per hundred people.4

A 2022 review of thirteen prospective studies estimated a 6% higher relative risk per additional daily drink, with differences by sex and uncertainty about the low-intake curve.5 A 2026 review of 26 observational studies instead found lower incident AF at low intake: under 12 g/day versus none had a pooled risk ratio of 0.72 (0.63–0.83).6

The newer review is relevant, but it does not prove protection. Its low-intake results varied greatly between studies; drinking was self-reported and groups could differ in health, social circumstances and AFib detection. Its comparisons between drinking levels relied on indirect evidence. Reviews also reuse many of the same cohorts.5,6

These findings cannot establish a reason to start drinking, a protective beverage, or a safe personal limit. They also do not overturn a randomized result about recurrence in people who already have AFib.

After ablation and possible mechanisms

A Japanese observational study analyzed 1,720 drinkers undergoing ablation. People who reduced their drinking had less recurrent AFib or atrial tachycardia. They were not randomly assigned to reduce it; other differences and changes after treatment could help explain the association. It does not establish an exact reduction that makes ablation successful.7

A separate placebo-controlled experiment in 100 people undergoing ablation found that intravenous alcohol shortened an electrical recovery measurement in the pulmonary veins. Overall AFib inducibility did not clearly differ. This offers a possible mechanism, rather than a prediction of what happens after an ordinary drink.8

What counts as a drink?

Comparing quantities

A drink is not the same unit everywhere

12 gPer drink in the reduction trial1
13.45 gPer Canadian standard drink13
10 gPer drink in the ESC recommendation12

Grams refer to pure alcohol. In Canada, 341 mL beer at 5%, 142 mL wine at 12%, and 43 mL spirits at 40% each contain one standard drink. Larger or stronger servings contain more.13

A weekly average also does not describe how drinking is distributed across days. Neither a serving size nor a guideline quantity is a boundary below which an AFib episode cannot occur.

What guidelines say

These summarize professional guidance for clinical care, rather than an individual drinking plan.

  • Canada, CCS/CHRS 2020: includes alcohol misuse within systematic assessment and management of modifiable conditions. Its broad recommendation for recurrence and symptom outcomes is strong, with low-quality evidence; it is not a specific alcohol-dose trial.9 A replacement AFib guideline is listed in development for 2026–2027.10
  • United States, ACC/AHA/ACCP/HRS 2023: recommends minimizing or eliminating alcohol for people with AFib pursuing rhythm control, to reduce recurrence and burden (Class 1, B-R).11
  • Europe, ESC 2024: recommends reducing intake to no more than 30 g/week, described as three 10 g drinks, within comprehensive care to reduce recurrence (Class I, B). This is about 2.2 Canadian standard drinks, not three Canadian servings.12,13

Guideline targets describe care strategies; they do not promise zero risk below a number. They also do not establish that fewer episodes remove a need for stroke-prevention treatment. Decisions about treatment require an individual clinical assessment.

Understanding your own record

Illustrative record · invented example

Recorded close together does not establish a cause

  1. 7:30 PM
    Drink noted

    Reported time and serving; quantity may be approximate.

  2. Between entries
    Rhythm unknown

    No continuous ECG. A gap does not establish normal rhythm.

  3. 10:40 PM
    Symptoms noticed

    Reported onset of sensations—not a measured rhythm start.

  4. 11:05 PM
    ECG saved

    A rhythm observation at that moment; earlier onset remains uncertain.

This is a schematic example, not an Evenbeat screenshot or study result. Entries are arranged in order, not to a proportional time scale. They do not show a causal link or the true start of an episode.

A useful record distinguishes a reported drink and its quantity, noticed symptoms, and the time an ECG was actually saved. Similar evenings without recorded AFib matter too, but an empty interval does not establish normal rhythm. A spot ECG cannot reveal the exact start of an earlier episode.

Evenbeat's role is to help make the record understandable alongside its gaps. A note, a watch observation and a Kardia recording can add different kinds of context. Seeing them together cannot establish that alcohol caused an episode, calculate a safe dose or predict the next one.

What this research cannot show

  • A personal safe amount, a safe waiting time, or whether one particular drink caused an episode.
  • Whether small reductions have the same effect as the large reduction in the six-month trial, or whether results generalize to groups it poorly represented.
  • Whether reducing alcohol prevents stroke, changes survival, or replaces medication or ablation.
  • A causal protective effect of low intake, wine or another beverage from observational comparisons.

Questions about drinking strategies remain under study. An ongoing UCSF trial compares avoidance with permissive or regulated drinking; a registration describes a study, not a result.14

Questions for your care team

  • How closely do the studied patients and drinking patterns match my circumstances?
  • Which times in my record are reported symptoms, ECG observations or uncertain episode starts?
  • What does the alcohol evidence mean for my rhythm-control plan, separately from stroke prevention?
  • If drinking is difficult to change or dependence is a concern, what support and clinical plan fit my situation?

Updates and corrections

  • · First edition: separates recurrence, burden, acute associations and incident AF; includes absolute trial results, low-intake uncertainty and Canadian, US and European context.

Sources

Evidence is labelled by study design and publication format. Its relevance depends on the question, population and measurements. Manufacturer testing is identified separately. Source checks record which text was available; they are not clinical review or a guarantee that a publication has no corrections.

  1. Randomized trial, 140 regular drinkers with AFib; six months

    Voskoboinik A, Kalman JM, De Silva A, et al. Alcohol Abstinence in Drinkers with Atrial Fibrillation. New England Journal of Medicine. 2020;382:20–28. doi:10.1056/NEJMoa1817591

    Assignment to abstinence produced a large reduction in drinking and fewer recurrences; the trial did not establish a safe dose.

    Limit: Small, motivated, mostly male sample; dependence excluded. Drinking self-reported. Mixed continuous and intermittent rhythm monitoring; six months, not stroke or survival evidence.

    Population, result and source check
    Applies to
    Paroxysmal or persistent AF, sinus rhythm at entry, ≥10 study drinks/week (about 12 g each); 85% men, mean age 62; six Australian hospitals.
    Comparison
    70 assigned to abstinence versus 70 continuing usual consumption; open label.
    Outcome and timing
    recurrence · Six months; recurrence after a two-week blanking period.
    Result
    Recurrence 37/70 (53%) versus 51/70 (73%); observed difference 20 percentage points; HR 0.55 (95% CI 0.36–0.84). Median burden 0.5% (IQR 0–3.0) versus 1.2% (0–10.3). Mean drinks/week 16.8→2.1 versus 16.4→13.2.
    Guidelines
    Central randomized evidence for US and European recommendations concerning recurrence and burden after AF diagnosis.
    Evidence checked
    2026-10-22 · abstract · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  2. Case-crossover study, 100 people with paroxysmal AFib; four weeks

    Marcus GM, Vittinghoff E, Whitman IR, et al. Acute Consumption of Alcohol and Discrete Atrial Fibrillation Events. Annals of Internal Medicine. 2021;174(11):1503–1509. doi:10.7326/M21-0228

    Recent drinks were associated with higher odds of an ECG-confirmed episode within the same person.

    Limit: Not randomized; factors changing alongside drinking can confound results. Odds ratios are not absolute probabilities or an individual prediction.

    Population, result and source check
    Applies to
    100 adults with paroxysmal AF; mean age 64, 79% men, 85% White; 56 had at least one recorded episode.
    Comparison
    Recent drinking versus periods without recent drinks within participants; continuous ECG, participant-marked drinks and transdermal alcohol sensing.
    Outcome and timing
    acute association · Four weeks; four-hour drink-count window and preceding 12-hour sensor exposure.
    Result
    One drink in prior four hours: OR 2.02 (95% CI 1.38–3.17); ≥2 drinks: OR 3.58 (1.63–7.89). Sensor exposure was also associated with episodes.
    Guidelines
    Supports short-term trigger plausibility in existing paroxysmal AF, not primary-prevention thresholds.
    Evidence checked
    2026-10-22 · abstract · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  3. Randomized trial, 43 selected alcohol for the first testing period

    Marcus GM, Modrow MF, Schmid CH, et al. Individualized Studies of Triggers of Paroxysmal Atrial Fibrillation: The I-STOP-AFib Randomized Clinical Trial. JAMA Cardiology. 2022;7(2):167–174; online November 2021. doi:10.1001/jamacardio.2021.5010

    The randomized alcohol comparison was inconclusive; analysis of reported exposure supported an association with reported AFib.

    Limit: Unblinded, self-reported exposure/events, incomplete adherence and attrition. Per-protocol analysis loses the full protection of randomization. Denominator inconsistencies are noted in the dossier.

    Population, result and source check
    Applies to
    Smartphone users with symptomatic paroxysmal AF interested in testing triggers; 82 alcohol tests across all periods, not 82 independent people.
    Comparison
    Randomized weekly exposure/avoidance assignments; intention-to-treat versus per-protocol analysis of reported exposure.
    Outcome and timing
    acute association · Six-week n-of-1 periods plus four-week follow-up; repeated testing available.
    Result
    All-period alcohol OR: intention to treat 1.17 (95% Bayesian credible interval 0.81–1.72); per protocol 1.77 (1.20–2.69). Primary quality-of-life outcome did not improve.
    Guidelines
    Supporting evidence with qualifications; does not validate an app that identifies personal causes or recommends alcohol challenges.
    Evidence checked
    2026-10-22 · full text · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  4. Observational cohort, 100,092 participants without AFib at baseline

    Csengeri D, Sprünker NA, Di Castelnuovo A, et al. Alcohol consumption, cardiac biomarkers, and risk of atrial fibrillation and adverse outcomes. European Heart Journal. 2021;42(12):1170–1177. doi:10.1093/eurheartj/ehaa953

    Higher usual alcohol intake was associated with more newly diagnosed AFib over years.

    Limit: Observational, residual confounding and exposure misclassification; does not estimate the effect of changing intake or recurrence after diagnosis.

    Population, result and source check
    Applies to
    Pooled European community cohorts; 100,092 participants free of AF at baseline.
    Comparison
    Self-reported usual intake, modeled per additional 12 g/day.
    Outcome and timing
    incident AF · Median 13.9 years.
    Result
    5,854 incident cases; HR per 12 g/day 1.16 (95% CI 1.11–1.22). The overall case proportion is not an absolute difference between intake groups.
    Guidelines
    Long-term prevention evidence addresses a different question from rhythm control in existing AF.
    Evidence checked
    2026-10-22 · abstract · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  5. Systematic review and meta-analysis, 13 prospective studies

    Jiang H, Mei X, Jiang Y, et al. Alcohol consumption and atrial fibrillation risk: An updated dose-response meta-analysis of over 10 million participants. Frontiers in Cardiovascular Medicine. 2022;9:979982. doi:10.3389/fcvm.2022.979982

    The pooled estimate rose with intake, with differences by sex and uncertainty at low intake.

    Limit: Observational exposure, between-study heterogeneity and overlapping cohorts with other reviews. Not an independent trial of drinking.

    Population, result and source check
    Applies to
    Prospective cohorts, over 10 million participants; incident AF.
    Comparison
    Dose-response modeling per additional daily drink, standardized to 12 g alcohol.
    Outcome and timing
    incident AF · Varied between included cohorts.
    Result
    Pooled RR 1.06 (95% CI 1.03–1.08) per additional drink/day; linear in men, potentially J-shaped in women.
    Guidelines
    Does not supply a personal low-risk drinking threshold or a recurrence estimate.
    Evidence checked
    2026-10-22 · full text · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  6. Systematic review and meta-analysis, 26 observational studies; nearly 15 million participants

    Asad ZUA, Jafry AH, Akbar U, et al. Alcohol consumption and risk of atrial fibrillation: a pairwise and network meta-analysis. EP Europace. 2026;28(6):euag111. doi:10.1093/europace/euag111

    Lower incident AF was associated with low intake in this review; that does not establish protection from drinking.

    Limit: All observational, self-reported intake, high heterogeneity, residual confounding; no direct comparisons between non-zero categories in the network. Internal reporting qualifications are documented in the dossier.

    Population, result and source check
    Applies to
    Studies of incident AF in people without prior AF; nearly 15 million participants.
    Comparison
    Harmonized intake categories versus no intake; indirect network comparisons between drinking levels.
    Outcome and timing
    incident AF · Weighted mean about 6.4 years.
    Result
    <12 g/day versus none RR 0.72 (95% CI 0.63–0.83), I² 90%; >60 g/day RR 1.75 (1.25–2.44).
    Guidelines
    Does not overturn recurrence trials; no burden assessment, personalized safe threshold or causal benefit established.
    Evidence checked
    2026-10-22 · full text · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  7. Observational cohort, 1,720 drinkers undergoing ablation; 88% men

    Takahashi Y, Nitta J, Kobori A, et al. Alcohol Consumption Reduction and Clinical Outcomes of Catheter Ablation for Atrial Fibrillation. Circulation: Arrhythmia and Electrophysiology. 2021;14(6):e009770. doi:10.1161/CIRCEP.121.009770

    Reduction in drinking was associated with less recurrent atrial arrhythmia after ablation; intake was not randomized.

    Limit: Not randomized; selection, confounding and intake assessed after treatment. A data-derived 1% split is not evidence that a tiny reduction causes benefit.

    Population, result and source check
    Applies to
    Japanese multicentre prospective ablation cohort; analyzed 1,720 baseline drinkers, mean age 64, 88% men.
    Comparison
    ≥1% versus <1% reduction in reported intake at one year; data-derived cutoff.
    Outcome and timing
    recurrence · Median 420 days.
    Result
    Adjusted HR for AF/atrial tachycardia recurrence 0.630 (95% CI 0.518–0.768). Median reductions in the compared groups were 75% versus 0%.
    Guidelines
    Supports broader risk-factor management around ablation, not a causal dose-response target.
    Evidence checked
    2026-10-22 · abstract · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  8. Randomized trial, 100 people undergoing AFib ablation

    Marcus GM, Dukes JW, Vittinghoff E, et al. A Randomized, Double-Blind, Placebo-Controlled Trial of Intravenous Alcohol to Assess Changes in Atrial Electrophysiology. JACC: Clinical Electrophysiology. 2021;7(5):662–670. doi:10.1016/j.jacep.2020.11.026

    Alcohol changed a pulmonary-vein electrical measurement; immediate AFib inducibility did not clearly differ.

    Limit: A mechanistic endpoint under procedural conditions is not a trial of real-world drinks or future clinical episodes.

    Population, result and source check
    Applies to
    Selected patients undergoing pulmonary-vein isolation, under general anesthesia.
    Comparison
    Intravenous alcohol titrated to about 0.08% blood alcohol concentration versus placebo.
    Outcome and timing
    acute association · Immediate laboratory measurements.
    Result
    Pulmonary-vein refractory periods shortened in the alcohol arm; no clear between-group difference in overall AF inducibility.
    Guidelines
    Biological plausibility, not an independent estimate of recurrence prevention.
    Evidence checked
    2026-10-22 · abstract · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  9. Clinical guideline

    Andrade JG, Aguilar M, Atzema C, et al. The 2020 CCS/CHRS Comprehensive Guidelines for the Management of Atrial Fibrillation: modifiable risk factors. Canadian Cardiovascular Society. 2020.

    Canadian guidance includes alcohol misuse within comprehensive risk-factor assessment and management.

    Limit: Broader risk-factor recommendation, not an isolated alcohol dose trial; predates 2021 acute and trigger-testing studies.

    Population, result and source check
    Applies to
    People with established AF or at high risk.
    Comparison
    Professional guidance; not an alcohol intervention trial.
    Outcome and timing
    guidance · 2020 guideline.
    Result
    Recommendation 11: systematic identification and management of modifiable conditions; strong recommendation, low-quality evidence for recurrence/symptom outcomes.
    Guidelines
    Canadian context; distinguish from the more specific US and European alcohol recommendations.
    Evidence checked
    2026-10-22 · official documentation · official URL
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  10. Official documentation

    Canadian Cardiovascular Society Guideline Development. Canadian Cardiovascular Society. 2026.

    A replacement Canadian AFib guideline is listed in development for 2026–2027.

    Limit: An expected date is not a published replacement or a guarantee.

    Population, result and source check
    Applies to
    Canadian AF guideline programme.
    Comparison
    Development listing, not completed guidance.
    Outcome and timing
    guidance · Listing checked October 22, 2026.
    Result
    Expected AF guideline publication: 2026–2027.
    Guidelines
    Explains the age of the Canadian guideline used on this page.
    Evidence checked
    2026-10-22 · official documentation · official URL
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  11. Clinical guideline

    Joglar JA, Chung MK, Armbruster AL, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation. Journal of the American College of Cardiology. 2023. doi:10.1016/j.jacc.2023.08.017

    For people pursuing rhythm control, US guidance recommends minimizing or eliminating alcohol to reduce recurrence and burden.

    Limit: Applies to rhythm control; not an individualized safe dose or proof of stroke prevention.

    Population, result and source check
    Applies to
    Patients with AF pursuing rhythm control.
    Comparison
    Professional guidance.
    Outcome and timing
    guidance · 2023 guideline.
    Result
    Section 5.2.4: minimize or eliminate alcohol; Class 1, B-R.
    Guidelines
    Specific AF recommendation rather than general-population drinking guidance.
    Evidence checked
    2026-10-22 · full text · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  12. Clinical guideline

    Van Gelder IC, Rienstra M, Bunting KV, et al. 2024 ESC Guidelines for the management of atrial fibrillation. European Heart Journal. 2024;45:3314–3414. doi:10.1093/eurheartj/ehae176

    European guidance specifies ≤30 g alcohol/week as part of comprehensive care to reduce recurrence.

    Limit: Here one drink means 10 g; a guideline target does not establish zero risk below it.

    Population, result and source check
    Applies to
    People with AF, within comprehensive risk-factor management.
    Comparison
    Professional guidance.
    Outcome and timing
    guidance · 2024 guideline.
    Result
    Recommendation Table 5: reduce to ≤3 standard drinks (≤30 g alcohol)/week to reduce AF recurrence; Class I, B.
    Guidelines
    Report grams alongside drinks; do not silently substitute Canadian servings.
    Evidence checked
    2026-10-22 · full text · DOI matched
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  13. Official documentation

    Health Canada Low-risk alcohol drinking guidelines: Standard drink. Government of Canada. 2025.

    A Canadian standard drink contains 13.45 g alcohol. Only the unit definition is used here.

    Limit: Serving sizes and strengths vary. This page also displays older drinking limits, which are not adopted in this article.

    Population, result and source check
    Applies to
    Canadian beverage unit definition.
    Comparison
    Pure alcohol by beverage volume and strength.
    Outcome and timing
    guidance · Official page checked October 22, 2026.
    Result
    13.45 g: 341 mL beer at 5%, 142 mL wine at 12%, or 43 mL spirits at 40%.
    Guidelines
    Unit conversion only; neither general drinking guidance nor an AF-specific recommendation.
    Evidence checked
    2026-10-22 · official documentation · official URL
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑

  14. Official documentation

    University of California, San Francisco Full-Avoidance vs. Permissive/Regulated Drinking & Outcomes On Fibrillation. UCSF Clinical Trials. 2026;NCT06754176.

    An ongoing trial is investigating different drinking strategies; its registration is not evidence of a result.

    Limit: Ongoing study; dates may change. Not counted as completed evidence.

    Population, result and source check
    Applies to
    Adults with AF enrolled in a new randomized alcohol-strategy study.
    Comparison
    Avoidance versus permissive/regulated drinking.
    Outcome and timing
    guidance · Sponsor listing checked October 22, 2026.
    Result
    Listing describes a trial that began in April 2025; no results reported on the consulted page.
    Guidelines
    Marks an unresolved research question, not a new recommendation.
    Evidence checked
    2026-10-22 · official documentation · official URL
    Publication status check
    No correction or retraction notice found in the consulted record. Scoped AI-assisted check, not exhaustive publication-status clearance or clinical review. Access limits are recorded in the research notes.
    Return to article summary ↑