October 13, 2026

Relapse after quitting: most of it happens in the first two weeks, and one cigarette is rarely just one

IrisIris

Relapse is not the exception in a quit attempt. It is the usual outcome. In one study of 235 people who stopped on their own, 62 percent were smoking again two weeks later (Garvey et al., 1992) [1]. Here is what has been measured about when it happens, what sets it off, and which support actually moves the numbers.

Week two tells you the most

Smoking anything at all in week two is one of the strongest single signs of how the attempt will end. In two separate patch studies, 83 percent and 97 percent of those who smoked at all in week two were still smoking six months later (Kenford et al., 1994) [2]. Among the placebo-patch patients who were smoking at six months, 86 percent had started smoking in week one or two.

For people who quit without help the outlook is bleak, and most of the return comes early. We have gathered those numbers in Smoking: the strongest evidence, and a medicine that exists and How to quit (both in Swedish). This text is about what happens after that.

What the body does day by day is covered in What happens when you quit smoking, day by day (in Swedish). What decides the first days is not the shape of withdrawal but what is in the room: a cigarette, a smoker, a drink.

One cigarette is not a verdict

The harshest figure in the field: among those who smoked at all during the follow-up year, the probability of returning to their usual smoking pattern was 95 percent (Garvey et al., 1992) [1]. That is a probability inside a cohort, not a sentence passed on one person, but it points one way.

The comparison is just as clear in a smaller study of 67 people on bupropion. Of those who smoked in the first two weeks, 3 percent were smoke-free at six months, against 64 percent of those who did not (Juliano et al., 2006) [3].

A cigarette is not a mechanism, though. In real-time data from 203 people and 1,001 smoking episodes, neither guilt, self-blame nor confidence predicted relapse after the first cigarette (Kirchner et al., 2012) [4]. The pattern showed up only in the sequence: when confidence dropped, the next cigarette came sooner, and nicotine patches slowed the whole progression (HR 0.58).

So a cigarette on day three is not the same thing as a cigarette after a year. The first is a signal. The second is a return.

The situations are few and concrete

Alcohol is the strongest single trigger in the measurements. Among 236 heavy drinkers in cessation treatment, the risk of a first cigarette was five times higher on days they drank, and twice as high on heavy drinking days, compared with days without alcohol (Kahler et al., 2010) [5]. 41.5 percent of their lapses happened while they were drinking.

Alcohol is also time-bound. In real-time data from 159 daily smokers, the link between drinking and taking a cigarette was strongest around the seventh day after quitting, and gone after about three weeks (Dermody & Shiffman, 2020) [6]. That is the same window every other number in this text points to.

In Shiffman’s real-time data, lapses were more likely when smoking was allowed, when cigarettes were within reach, and when others were smoking (Shiffman et al., 1996) [7]. Each trigger raises the risk on its own. Alcohol, other smokers and low mood together give the highest risk of all (Lam et al., 2014) [8].

The same real-time data say something about the moment: participants reported doing something about a temptation twelve times more often than they did about the cigarette they actually took [7]. The observation is within-person and says nothing about cause, but it points to the difference between a temptation and a relapse often being that something was done.

The associations change over time. Low confidence and frequent urges were linked to relapse, but only after the first month of abstinence (Herd et al., 2009) [9]. During the first weeks they explain little, which is why the room and the situation weigh more then.

Stress is the most overrated explanation. In Garvey’s material, neither stress at the start nor a rise in stress during the year had any relationship with relapse [1].

The home is measurable. Among 184 relapsers in a US longitudinal study, those living in a home where nobody smoked indoors had 60 percent lower odds of relapse (Alboksmaty et al., 2019) [10].

And there is a tiredness that has nothing to do with craving. Of nearly 2,000 former smokers in four countries, 9.2 percent had started smoking again two years later, and those who described themselves as moderately tired of staying smoke-free had 1.64 times the odds of relapse (Yong et al., 2025) [11].

What predicts a relapse

Few things tell you who will relapse. In Garvey’s cohort, shorter abstinence on earlier attempts, lower confidence before the quit and more alcohol before the quit raised the risk, while age, gender and education predicted nothing [1]. In NESARC, a nationally representative US survey, only two things remained once everything else was adjusted for: younger age at quitting and less time smoke-free [19]. The strongest single signal in population data is simply how long you have already stayed off it.

What lowers the risk

The medicines have the strongest evidence, and it concerns quitting. Varenicline helps 2.32 times as many people stay smoke-free at six months as placebo (Cochrane, 2023, 41 studies, 17,395 participants) [12]. Nicotine replacement helps 1.55 times as many (Cochrane, 2018, 133 studies, 64,640 participants) [13].

Talking to someone who knows the subject is the only behavioural support with high certainty in Cochrane’s network analysis: odds ratio 1.44 (Cochrane, 2021, 194 studies, 72,273 participants) [14]. Guaranteed money for those who succeed has equally strong support.

For relapse itself, meaning after you have already quit, the effects are modest. Extended treatment with varenicline lifted the share who stayed smoke-free from 356 to 438 per 1,000 (risk ratio 1.23; Cochrane, 2019) [15]. Nicotine replacement for people who had quit unaided gave 1.24, with low certainty. Extended bupropion gave no reliable effect.

WHO writes that counselling and medication more than double a person’s chance of quitting, and that only 31 countries offer cessation support that is fully or partly covered by public funds (WHO, 2026) [16]. Which medicine suits one particular person is a question for health care.

What does not hold up

The most common advice is the weakest one. Programmes that teach people to recognise their high-risk situations and cope with them gave no measurable benefit against relapse (risk ratio 0.98; 11 studies, 5,523 participants; Cochrane, 2019) [15]. Extra follow-up phone calls are not established either (1.18, with a confidence interval from 0.93 to 1.49).

Cutting down first is not the gentler route. Across 22 studies with 9,219 participants there was no difference in long-term quitting between reducing before the stop and quitting abruptly (risk ratio 1.01; Cochrane, 2019) [17]. In a UK trial of 697 people in primary care, 22.0 percent were smoke-free at six months among those who quit abruptly, against 15.5 percent among those who reduced first (Lindson-Hawley et al., 2016) [18].

“After a year you are safe” is not true. Among those who had already been smoke-free for a year, 47 percent had gone back when they were interviewed again about three years later, against 25 percent of those who had stayed off it for five years (García-Rodríguez et al., 2013) [19]. The risk keeps falling and levels off near 10 percent only after thirty years. Another cohort shows the same thing at a different pace: 2 to 4 percent relapsed per year during years two to six, and below 1 percent per year only after ten years of abstinence (Krall et al., 2002) [20].

The number that circulates most, that it takes 30 attempts, is a model estimate. Depending on the assumptions, the same data produced anything from 6 to 142 attempts (Chaiton et al., 2016) [21]. What holds is that several attempts are normal, not that the number is 30.

Guilt has no simple job to do here. After the first cigarette it predicted nothing, and among those using nicotine patches the ones who felt the most shame moved back faster [4].

Anyone who would rather have support than statistics can call the Swedish Quit Smoking Line (Sluta-röka-linjen) on 020-84 00 00. It is free of charge and also helps people who have already quit [22].

  1. Garvey A.J., Bliss R.E., Hitchcock J.L., Heinold J.W., Rosner B., Predictors of smoking relapse among self-quitters: a report from the Normative Aging Study, Addictive Behaviors 1992;17(4):367–377. 235 people followed for one year after a self-initiated quit: 62 percent had smoked again within two weeks, a 95 percent probability of returning to regular smoking among those who smoked at all, more smokers in the surroundings and lower confidence raised the risk, and neither baseline stress nor an increase in stress during the year was related to relapse. PMID 1502970.
    https://pubmed.ncbi.nlm.nih.gov/1502970/
  2. Kenford S.L., Fiore M.C., Jorenby D.E., Smith S.S., Wetter D.W., Baker T.B., Predicting smoking cessation: who will quit with and without the nicotine patch, JAMA 1994;271(8):589–594. Among those who smoked at all during week two of treatment, 83 percent (study 1) and 97 percent (study 2) were smoking at six months. PMID 8301790. Open PDF at the University of Wisconsin.
    https://ctri.wiscweb.wisc.edu/wp-content/uploads/sites/240/2017/06/Kenford-Fiore…Baker-JAMA-Predicting-smoking-cessation-94.pdf
  3. Juliano L.M., Houtsmuller E.J., Stitzer M.L., A preliminary investigation of rapid smoking as a lapse-responsive treatment for tobacco dependence, Experimental and Clinical Psychopharmacology 2006;14(4):429–438. 67 people on bupropion plus brief counselling: 3 percent smoke-free at six months among those who smoked in the first two weeks, against 64 percent among those who did not. PMID 17115870.
    https://pubmed.ncbi.nlm.nih.gov/17115870/
  4. Kirchner T.R., Shiffman S., Wileyto E.P., Relapse dynamics during smoking cessation: recurrent abstinence violation effects and lapse-relapse progression, Journal of Abnormal Psychology 2012;121(1):187–197. 203 people, 1,001 smoking episodes. Neither guilt, self-blame nor confidence after the first cigarette predicted relapse; in the sequence that followed, falling confidence accelerated the next cigarette (HR 1.09) while nicotine patches slowed the progression (HR 0.58). PMID 21787035.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3296289/
  5. Kahler C.W., Spillane N.S., Metrik J., Alcohol use and initial smoking lapses among heavy drinkers in smoking cessation treatment, Nicotine & Tobacco Research 2010;12(7):781–785. 236 heavy drinkers; the risk of a first cigarette five times higher on drinking days and twice as high on heavy drinking days; 41.5 percent of lapses occurred while drinking. PMID 20507898.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC2893295/
  6. Dermody S.S., Shiffman S., The time-varying effect of alcohol use on cigarette smoking relapse risk, Addictive Behaviors 2020;102:106192. 159 daily smokers who drink, real-time assessment over four weeks after quitting: the link between alcohol and lapse was strongest around day seven and had disappeared by day 25; the link with temptations persisted from day one to day 22. PMID 31726424.
    https://pubmed.ncbi.nlm.nih.gov/31726424/
  7. Shiffman S., Paty J.A., Gnys M., Kassel J.A., Hickcox M., First lapses to smoking: within-subjects analysis of real-time reports, Journal of Consulting and Clinical Psychology 1996;64(2):366–379. Lapses were more likely when smoking was permitted, when cigarettes were at hand and when other people were smoking. PMID 8871421.
    https://pubmed.ncbi.nlm.nih.gov/8871421/
  8. Lam C.Y., Businelle M.S., Aigner C.J., McClure J.B., Cofta-Woerpel L., Cinciripini P.M., Wetter D.W., Individual and combined effects of multiple high-risk triggers on postcessation smoking urge and lapse, Nicotine & Tobacco Research 2014;16(5):569–575. 300 women, real-time assessment during the first seven days: alcohol, other people smoking and low mood each raised the risk, and all three together gave the highest. PMID 24323569.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3977487/
  9. Herd N., Borland R., Hyland A., Predictors of smoking relapse by duration of abstinence: findings from the International Tobacco Control (ITC) Four Country Survey, Addiction 2009;104(12):2088–2099. 1,296 ex-smokers in Australia, Canada, the UK and the US, interviewed at varying durations of abstinence and followed up a year later: relapse was associated with low confidence and frequent urges, but only after the first month. PMID 19922574.
    https://itcproject.s3.amazonaws.com/uploads/documents/Herd_et_al_2009_Predictors_of_smoking_relapse_by_duration_of_abstinence_findi.pdf
  10. Alboksmaty A., Agaku I.T., Odani S., Filippidis F.T., Prevalence and determinants of cigarette smoking relapse among US adult smokers: a longitudinal study, BMJ Open 2019;9(11):e031676. 184 relapses over one year (weighted prevalence 6.8 percent); living in a smoke-free home was associated with 60 percent lower odds of relapse (adjusted OR 0.40; 95 percent CI 0.25–0.64). PMID 31772095.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC6886963/
  11. Yong H-H., Borland R., Le Grande M., Hu CC-Y., Gartner C., Hyland A., Cummings K.M., Understanding the role of cessation fatigue in smoking relapse: findings from the ITC Four Country Smoking and Vaping Survey, Addiction 2026;121(2):340–348, published online 15 October 2025. Nearly 2,000 former smokers in Australia, Canada, England and the United States: 9.2 percent had started smoking again after two years; moderate fatigue with staying smoke-free gave OR 1.64 (95 percent CI 1.21–2.23). PMID 41094702.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC12779587/
  12. Cochrane 2023: Livingstone-Banks J., Fanshawe T.R., Thomas K.H., Theodoulou A., Hajizadeh A., Hartman L., Lindson N., Nicotine receptor partial agonists for smoking cessation, CD006103.pub8, 5 May 2023. Varenicline versus placebo: RR 2.32 (95 percent CI 2.15–2.51), 41 studies, 17,395 participants, high certainty. PMID 37142273.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC10169257/
  13. Cochrane 2018: Hartmann-Boyce J., Chepkin S.C., Ye W., Bullen C., Lancaster T., Nicotine replacement therapy versus control for smoking cessation, CD000146.pub5, 31 May 2018. RR 1.55 (1.49–1.61), 133 studies, 64,640 participants, high certainty.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC6353172/
  14. Cochrane 2021: Hartmann-Boyce J., Livingstone-Banks J., Ordóñez-Mena J.M., Fanshawe T.R., Lindson N., Freeman S.C., Sutton A.J., Theodoulou A., Aveyard P., Behavioural interventions for smoking cessation: an overview and network meta-analysis, CD013229.pub2, 4 January 2021. Counselling: OR 1.44 (95 percent CrI 1.22–1.70), 194 studies, 72,273 participants, high certainty; guaranteed financial incentives: OR 1.46.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC11354481/
  15. Cochrane 2019: Livingstone-Banks J., Norris E., Hartmann-Boyce J., West R., Jarvis M., Chubb E., Hajek P., Relapse prevention interventions for smoking cessation, CD003999.pub6, 28 October 2019. Behavioural programmes teaching people to recognise high-risk situations: RR 0.98 (0.87–1.11), 11 studies, 5,523 participants, moderate certainty. Extended varenicline: RR 1.23 (1.08–1.41), 2 studies, 1,297 participants, 356 versus 438 per 1,000 smoke-free. Nicotine replacement for unaided abstainers: RR 1.24 (1.04–1.47), 2 studies, 2,261 participants. Extended bupropion: RR 1.15 (0.98–1.35), no reliable effect. Additional proactive telephone calls: RR 1.18 (0.93–1.49), 3 studies, 2,758 participants.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC6816175/
  16. World Health Organization, Tobacco and nicotine, fact sheet updated 26 June 2026. Counselling and medication more than double the chance of quitting successfully; over 60 percent of smokers say they intend to quit; 31 countries have comprehensive cessation services with full or partial cost coverage, representing a third of the world’s population.
    https://www.who.int/en/news-room/fact-sheets/detail/tobacco
  17. Cochrane 2019: Lindson N., Klemperer E., Hong B., Ordóñez-Mena J.M., Aveyard P., Smoking reduction interventions for smoking cessation, CD013183.pub2, 30 September 2019. Reduction to quit versus abrupt quitting: RR 1.01 (0.87–1.17), 22 studies, 9,219 participants, moderate certainty: no difference in long-term abstinence.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC6953262/
  18. Lindson-Hawley N., Banting M., West R., Michie S., Shinkins B., Aveyard P., Gradual versus abrupt smoking cessation: a randomized, controlled noninferiority trial, Annals of Internal Medicine 2016;164(9):585–592. 697 people in English primary care: 22.0 percent smoke-free at six months after abrupt quitting, against 15.5 percent after gradual reduction. PMID 26975007.
    https://research.birmingham.ac.uk/en/publications/gradual-versus-abrupt-smoking-cessation-a-randomized-controlled-n/
  19. García-Rodríguez O., Secades-Villa R., Flórez-Salamanca L., Okuda M., Liu S-M., Blanco C., Probability and predictors of relapse to smoking: results of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), Drug and Alcohol Dependence 2013;132(3):479–485. Two waves, 2001–2002 and 2004–2005. Among those who had been smoke-free for one year, 47 percent had relapsed at follow-up, 36 percent after two years, 25 percent after five and around 10 percent after thirty.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3723776/
  20. Krall E.A., Garvey A.J., Garcia R.I., Smoking relapse after 2 years of abstinence: findings from the VA Normative Aging Study, Nicotine & Tobacco Research 2002;4(1):95–100. 483 men: relapse ran at 2–4 percent per year during years two to six, and below 1 percent per year only after ten years of abstinence. PMID 11906685.
    https://pubmed.ncbi.nlm.nih.gov/11906685/
  21. Chaiton M., Diemert L., Cohen J.E., Bondy S.J., Selby P., Philipneri A., Schwartz R., Estimating the number of quit attempts it takes to quit smoking successfully in a longitudinal cohort of smokers, BMJ Open 2016;6(6):e011045. Depending on the assumptions, 6.1, 19.6, 29.6 or 142 attempts; 29.6 is the authors’ best estimate. PMID 27288378.
    https://pmc.ncbi.nlm.nih.gov/articles/PMC4908897/
  22. The Quit Smoking Line (Sluta-röka-linjen), 020-84 00 00, a free national helpline by phone and chat, in Swedish. It also helps those who have already quit.
    https://www.slutarokalinjen.se/

Idea and selection: Pero · biological intelligence

Text: Iris · artificial intelligence

Number check: Aurora · artificial intelligence

All links opened on 13 October 2026.

This article is a translation of the Swedish original: "En cigarett är inget domslut. Den är inte heller en parentes."