checkedHuberman Lab2026-11-19last verified 2026-11-20

do 75% of quitters relapse within a week, and do 95% of quit attempts fail?

verdict: overstated
directionally right; magnitude or certainty inflated
“75% of people relapse within one week, and overall failure rates are 95%”

a host said on Huberman Lab, 2026-11-19, at 33:25. hear it in context →

first, the part that holds

the core of it holds for cigarettes: a systematic review of people who quit smoking without treatment found that most relapse happens in the first 8 days, and only about 3-5% of those unaided attempts last 6-12 months. that is a 95-97% failure rate per unaided attempt.

what we found when we went and looked

we went and looked. the 95% figure comes from research on people quitting cigarettes without help. a 2004 systematic review in addiction puts long-term abstinence after an unaided attempt at 3-5%, and it says most relapse happens within the first 8 days. we could not find the exact "75% within one week" number in the record we read, though the direction is right. the claim also stretches past its evidence in two ways. "overall" folds in attempts made with treatment, which the data do not cover. and the numbers come from cigarettes, not vaping: in a large trial of young adults who wanted to quit vaping, 18.6% of the group who got no help reported no vaping in the past 30 days at 7 months, so roughly 81% had not quit, not 95%.

assertion by assertion

assertionverdictwhat the record says
75% of people who quit smoking relapse within one weekunverifiablethe 2004 systematic review of unaided quitters says most relapse occurs in the first 8 days, based on eight relapse curves that the authors call heterogeneous. the abstract gives no 75%-at-one-week figure, and the authors say the data were too few and too varied to pool. the direction (relapse is heavily front-loaded) is supported. the specific 75% is not, by anything we read. [1]
the overall quit-attempt failure rate is 95%overstatedfor a single unaided attempt, 3-5% stay abstinent for 6-12 months, so a 95-97% failure rate is accurate for that group. 'overall' goes further than the data. the 3-5% figure covers only self-quitters, not people using counseling or medication. cdc's 2022 data show 53.3% of adults who smoke tried to quit in the past year and 8.8% of them quit successfully for at least 6 months. those are per-person, per-year numbers, not per-attempt failure rates. [1][2]
the same relapse and failure figures apply to quitting vapingmisattributedboth numbers come from studies of cigarette smoking. in a randomized trial of 2,588 people aged 18-24 who vaped and wanted to quit within 30 days, the assessment-only control group had 18.6% self-reported 30-day abstinence at 7 months, and 24.1% with a text-message program. that is a failure rate of about 81% without help, not 95%. it uses a different measure (point prevalence, self-report) and a different population (motivated young adults).
most quitters do not succeed in a given year (population view)confirmedcdc, 2022: of 28.8 million us adults who smoked, about half (53.3%) tried to quit in the past year, but only 8.8% quit successfully, defined as staying quit for at least 6 months. the cdc report has no figures on vaping. [2]

both directions

for the speaker: the 2004 systematic review of untreated smokers puts long-term (6-12 month) abstinence after a single unaided attempt at 3-5%. that matches the 95% failure figure. the same review says most relapse occurs in the first 8 days and recommends that cessation programs focus on the first week. cdc's 2022 data back up the broader picture: about half of adults who smoke try to quit each year, and only 8.8% quit successfully.

against: the exact 75%-in-a-week figure does not appear in the abstract we read. the review says the curves were too heterogeneous to pool, which works against stating any single precise percentage. the 95% figure only covers people quitting without help, so calling it the "overall" rate goes beyond the evidence. the claim also applies smoking data to vaping. the one large vaping-cessation trial we read found about 19% of untreated young adults reporting abstinence at 7 months, so failure was closer to 81% than 95%.

what this cannot say

we read the 2004 review's abstract, not the full text, so a 75% figure could be in its tables, though the abstract does not report one. the review's long-term figure relies on earlier summaries, and its relapse curves come from only a handful of studies, mostly decades old. the cdc measure is per person per year and includes people who did not try to quit, so it cannot be turned directly into a per-attempt failure rate. the vaping trial used self-reported 30-day abstinence, not prolonged abstinence, and enrolled young adults who were motivated to quit, so it is not directly comparable to the smoking figures. we did not find a population-level relapse curve for vaping, and none may exist yet.

sources, each one fetched and read

  1. Hughes JR, Keely J, Naud S. Shape of the relapse curve and long-term abstinence among untreated smokers. Addiction 2004;99(1):29-38. doi:10.1111/j.1360-0443.2004.00540.x. PMID 14678060 What we read there: eight relapse curves from self-quitters and no-treatment control groups indicate most relapse occurs in the first 8 days. 3-5% of self-quitters achieve prolonged abstinence for 6-12 months after a quit attempt. the data were too few and heterogeneous for meta-analysis. interventions should focus on the first week. Fetched 2026-11-20 (HTTP 203). https://pubmed.ncbi.nlm.nih.gov/14678060/
  2. CDC. Adult Smoking Cessation — United States, 2022. MMWR 2024;73(29) What we read there: of 28.8 million us adults who smoked, approximately one half (53.3%) made a quit attempt, but only 8.8% quit smoking. recent successful cessation means having formerly smoked and stayed quit for at least 6 months during the past year. no e-cigarette figures. Fetched 2026-11-20 (HTTP 200). https://www.cdc.gov/mmwr/volumes/73/wr/mm7329a1.htm

speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam