Cold turkey or taper off nicotine: which works better?

Both work, and the gap between them is smaller than either camp claims. The best-known head-to-head trial, in 697 smokers in England, found that stopping abruptly beat cutting down first: 49.0 per cent against 39.2 per cent abstinent at four weeks, and 22.0 against 15.5 per cent at six months. But a Cochrane review pooling 22 trials and 9,219 people found no difference in long-term quit rates between reducing first and stopping abruptly, with moderate certainty. A fair reading is that a taper with a schedule, a quit date and support is a legitimate method, and a taper with no end date tends to become indefinite cutting down. The limit of the evidence matters. Nearly all of it is in cigarette smokers, the key trial gave nicotine replacement to both groups, and the Cochrane review of vaping cessation contains no trial comparing a taper with an abrupt stop.
What the head-to-head trial found
The study most often quoted for cold turkey is a randomised trial published in Annals of Internal Medicine in 2016. Researchers at Oxford recruited 697 adult smokers through primary care clinics in England, all of whom wanted to stop soon. Half set a quit day and smoked normally until it arrived. The other half cut their smoking by 75 per cent over the two weeks before the same quit day. Both groups had support from nurses and used nicotine replacement before and after quitting.
Four weeks after quit day, 49.0 per cent of the abrupt group and 39.2 per cent of the gradual group were abstinent, confirmed by carbon monoxide breath tests. At six months the figures were 22.0 and 15.5 per cent. The trial was designed to show that gradual was no worse than abrupt, and it failed to show that.
Two details matter more than the headline. First, according to the lead author, the gap opened on quit day itself: fewer people in the gradual group made a quit attempt at all, and among those in either group who did, success was equal. Cutting down was an extra task, and some never got past it. Second, people who said beforehand that they preferred to quit gradually did worse than those who preferred to stop abruptly — 38.3 against 52.2 per cent — whichever group they were put in.
The authors’ own caveat is the fairest summary. For people ready to stop and receiving support, the best advice is to pick a day and stop. For people who cannot imagine stopping completely, cutting down is much better than doing nothing.
What the pooled evidence says
One trial is not the evidence base. A Cochrane review published in 2019 gathered 51 trials with 22,509 participants, typically smoking about 23 cigarettes a day. Twenty-two of those trials, with 9,219 people, compared reducing-to-quit directly with abrupt quitting. The pooled risk ratio was 1.01, with a 95 per cent confidence interval from 0.87 to 1.17. In plain terms: no detectable difference in long-term quit rates, with moderate certainty, and an interval that leaves room for a modest advantage either way.
The same review found that reduction worked better when it was backed by medication. Reducing with pharmacotherapy beat reducing alone, and the benefit was clearest for varenicline and for fast-acting nicotine replacement such as gum or lozenges. Whether reducing-to-quit beats making no attempt was, perhaps surprisingly, left inconclusive, because the trials were few and inconsistent.
The Oxford trial and the review are not in conflict. The trial shows that in one well-run programme, for people ready to stop now, a two-week cut-down phase cost some of them their quit attempt. The review shows that across many programmes and populations, structured reduction ends in about the same place. Anyone citing only one of the two is arguing a side.
What “cold turkey” means in these studies
It is worth being exact, because the phrase is used loosely. In the trials, “abrupt” means stopping on a set day rather than cutting down first. It does not mean unaided. The abrupt group in the Oxford trial wore nicotine patches and saw a nurse. So the trial evidence is about the timing of stopping, not about whether to use support or medication.
On that separate question the evidence is not close. Across 133 trials and 64,640 people, nicotine replacement increased the rate of quitting by about 50 to 60 per cent compared with placebo or no treatment. Among smokers who try with no help at all, a review of the available studies put prolonged abstinence at six to twelve months at roughly 3 to 5 per cent per attempt, with most relapse in the first eight days.
It is also true that most former smokers stopped without assistance — two-thirds to three-quarters, in the research summarised by Chapman and MacKenzie. Both facts hold together. Unaided quitting is the commonest route to success because it is by far the commonest kind of attempt, not because any single unaided attempt has better odds.
Why tapering a vape is harder than tapering cigarettes
Here the evidence thins out sharply. Cochrane’s living review of interventions for quitting vaping, updated in November 2025, covers 15 studies and about 5,800 people, and rates the evidence on reducing nicotine or vaping as inconclusive. In its first edition that evidence was a single small trial, analysed in 17 participants, which could not rule out either benefit or harm. No included trial compares a taper with an abrupt stop. Everything above is borrowed from smokers.
The practical difficulty is that a cigarette is a unit and a puff is not. A smoker cutting from twenty to ten to five can count. A vaper has two dials — the nicotine strength of the liquid and how much of it is used — and only the first is easy to see. This is reasoning rather than trial evidence, but there is one relevant experiment.
Twenty experienced vapers used 6 and 18 milligram-per-millilitre liquid for a week each. On the weaker liquid they took more puffs a day (338 against 279 with the device at fixed power), drew for longer and used about a third more liquid. Their saliva cotinine was still lower — about 250 against 400 nanograms per millilitre — so stepping down did reduce nicotine intake, though by less than the label implies, and urges to vape and withdrawal symptoms were higher. The study was small, short and unblinded.
If you are going to taper, what the evidence suggests
The features shared by the reduction programmes that performed as well as abrupt quitting are not mysterious: a defined schedule, a quit date fixed in advance, and some form of support. The Oxford trial’s taper lasted two weeks. An open-ended intention to cut down is a different thing, and the trials do not speak for it.
For a vape, the experiment above suggests reducing both dials rather than one: step the strength down and also limit when and where the device is used, so that weaker liquid is not simply consumed in larger volume. Keeping the device’s power setting fixed is sensible, since vapers in that study tended to turn the power up on weaker liquid. This is inference from one small study, not a tested protocol.
Reduction is where the Cochrane review found that medication may help, most convincingly varenicline and fast-acting nicotine replacement, so a conversation with a doctor or pharmacist is worth having before starting. And one honest test applies to any taper: if the date for zero keeps moving, it has become maintenance.
What happens after the last dose, either way
Both routes end in the same place: a last dose, and then withdrawal. Nicotine’s half-life is about two hours, so blood levels fall steeply over the first day. The recognised symptoms — irritability, anxiety, low mood, poor concentration, restlessness and insomnia — peak within the first week and last two to four weeks.
It is commonly assumed that tapering makes this stage milder. That is plausible, since a lower starting dose means a smaller drop, but the trials reviewed here were not designed to answer it. What the Cochrane review does report, from sparse data, is that reduction did not produce an excess of withdrawal symptoms during the cut-down phase. A taper changes when withdrawal is felt; whether it changes how much remains uncertain.
When to see a doctor
- Chest pain, tightness, or pain spreading to the arm or jaw — call emergency services
- Low mood that keeps deepening after stopping or cutting down, or any thoughts of harming yourself — seek urgent help
- Nausea, vomiting, a pounding heartbeat or dizziness when combining nicotine replacement with vaping or smoking — the dose is too high; seek advice
- Pregnancy — talk to a midwife or doctor before choosing a method
- An existing mental health condition or regular prescription medicines — tell your doctor you are stopping nicotine
- Withdrawal symptoms that are severe or not easing after about four weeks
What 336 does not do
336 is cold turkey only: no step-down schedule, and no nicotine replacement sold or recommended inside it. That is a design choice, not a verdict on the evidence above, which shows that a structured taper with support is a legitimate way to stop. The app is built for the other route — a quit timestamp, then a 14-day hour-by-hour timeline — free for the first 24 hours and then a one-time purchase. If a taper is working for you, nothing on this page says to abandon it.
On the quit timeline
Withdrawal symptoms, hour by hour
자주 묻는 질문
Is cold turkey the best way to quit smoking?
Stopping on a set day did better than cutting down first in the best-known trial: 22.0 against 15.5 per cent abstinent at six months. A Cochrane review of 22 trials found no overall difference. Note that “abrupt” in these trials did not mean unaided; participants used nicotine replacement and had support, which separately improves the odds.
How do you taper off nicotine?
The reduction programmes with evidence behind them used a fixed schedule, a quit date set in advance and support. One major trial cut smoking by 75 per cent over two weeks before quit day. Cochrane found reduction worked better with varenicline or fast-acting nicotine replacement, which means speaking to a doctor or pharmacist first.
Is it better to quit vaping gradually or cold turkey?
Nobody knows from direct evidence. Cochrane’s review of ways to quit vaping includes no trial comparing a taper with an abrupt stop, and rates the evidence on reducing nicotine or vaping as inconclusive. In smokers the two approaches give similar long-term results. Either is reasonable; a taper needs a fixed end date to avoid becoming indefinite.
Does lowering the nicotine strength of a vape work?
Partly. In a study of 20 experienced vapers, switching from 18 to 6 mg/mL liquid led to more and longer puffs and about a third more liquid used, but cotinine levels were still lower. So intake falls, by less than the label suggests, with stronger urges to vape. Limiting when you vape as well as the strength is the logical response.
Does tapering make nicotine withdrawal easier?
It is plausible but not established. The trials comparing gradual and abrupt quitting were designed to measure quit rates, not symptom severity. Cochrane reports, from limited data, that cutting down did not cause excess withdrawal symptoms before quit day. After the last dose, symptoms peak within the first week and mostly ease over two to four weeks.
Do most people who quit do it cold turkey?
Most former smokers stopped without assistance — two-thirds to three-quarters in the research summarised by Chapman and MacKenzie. That reflects how many attempts are unaided rather than a higher success rate per attempt: among untreated smokers, roughly 3 to 5 per cent achieve prolonged abstinence at six to twelve months, and medication or support improves on that.
Sources
- Gradual versus abrupt smoking cessation: a randomized, controlled noninferiority trial. Annals of Internal Medicine (Lindson-Hawley et al.), 2016.
- If you want to quit smoking, do it now. Nuffield Department of Primary Care Health Sciences, University of Oxford, 2016.
- Smoking reduction interventions for smoking cessation. Cochrane Database of Systematic Reviews, 2019.
- Nicotine replacement therapy versus control for smoking cessation. Cochrane Database of Systematic Reviews, 2018.
- Shape of the relapse curve and long-term abstinence among untreated smokers. Addiction (Hughes, Keely, Naud), 2004.
- The global research neglect of unassisted smoking cessation: causes and consequences. PLoS Medicine (Chapman, MacKenzie), 2010.
- Interventions for quitting vaping. Cochrane Database of Systematic Reviews (living review), 2025.
- ‘Real-world’ compensatory behaviour with low nicotine concentration e-liquid: subjective effects and nicotine, acrolein and formaldehyde exposure. Addiction (Dawkins et al.), 2018.
- Effects of abstinence from tobacco: valid symptoms and time course. Nicotine & Tobacco Research (Hughes JR), 2007.
- Nicotine chemistry, metabolism, kinetics and biomarkers. Handbook of Experimental Pharmacology (Benowitz, Hukkanen, Jacob), 2009.
