Same quitting interval, different outcomes
Roura et al. (2014), original Table 2: adjusted HRs (95% confidence intervals), compared with current smokers. Years already quit at entry; observed new diagnoses, not clearance or personal probabilities.
| Years since quitting | CIN3/CIS: HR; cases | Invasive cancer: HR; cases |
|---|---|---|
| 10–19 years | 0.5 (0.4–0.8); 46 | 0.8 (0.5–1.3); 21 |
| At least 20 years | 0.5 (0.3–0.7); 28 | 0.4 (0.2–0.8); 10 |
Check which outcome a headline calls ‘cancer’
The 2014 EPIC study followed 308,036 women across ten European countries and recorded 261 new invasive cervical cancers and 804 CIN3/CIS cases over a median nine years. CIN3 means severe cervical precancerous change; CIS means carcinoma in situ. The study grouped these together, separately from invasive cervical cancer (ICC). An association with fewer new lesions is not a measured reversal of an existing lesion.
Its Table 2 did not give identical results for these two outcomes. Among those reporting 10–19 years since quitting at entry, the CIN3/CIS estimate was 0.5, but the invasive-cancer estimate was 0.8 with a confidence interval spanning 1. Calling both ‘half the cancer risk after ten years’ loses information. The longest category was at least 20 years, not a deadline at exactly year twenty.
What this comparison can and cannot establish
The table compares groups with current smokers, not with never smokers. The adjusted hazard ratio (HR) compares the rate of new diagnoses during follow-up among people who have not yet developed that outcome; 1 is the reference, not a 1% personal chance. A lower estimate is not a forecast for an individual, and an interval crossing 1 neither proves equal risk nor makes quitting pointless.
Smoking history was recorded at entry and was not updated during follow-up. Screening attendance was not collected, and CIN3/CIS recording was incomplete across centres. Those limitations matter because people who quit may differ in other ways. The paper's separate nested case-control analysis adjusted for HPV antibodies; the whole cohort table did not. Its odds ratios must not be substituted for the HRs here.
There was no HPV-clearance stopwatch
The study did not repeatedly test cervical HPV DNA to measure clearance. Antibody testing in the smaller analysis was not proof of a current cervical infection, and a negative antibody result did not establish that infection was absent. HPV is common; an infection or a cancer diagnosis is not a moral verdict. Do not turn a smoking record into a negative HPV test, a healed-lesion claim or a guarantee about progression.
Keep quitting support and cervical decisions distinct
If you have a screening result, a cervical lesion or an existing cancer diagnosis, take the actual report and your question to the professional responsible for that care. Do not cancel follow-up or select a vaccine, test or interval from a cessation-years table. Vaccination, screening and follow-up address different clinical tasks; eligibility and arrangements depend on local guidance and personal circumstances.
For separate help stopping smoking in England, the NHS links to local stop-smoking services; confirm access with the service. This page needs no HPV result, sexual history, screening report or quit date entered online. A useful question for the appointment is ‘What does my existing result require, independently of the benefit of stopping smoking?’
What to keep in mind
Sources
The central claims on this page were checked against the sources below.
- U.S. Department of Health and Human Services: Smoking Cessation (2020), Chapter 4: Cervix and synthesis of cancer evidence
Sources checked: 2026-10-07
- International Journal of Cancer / Universitat de Barcelona repository: Roura et al. (2014), Smoking and the risk of cervical cancer in EPIC, pp. 453–466; original article in author's 2018 dissertation
Sources checked: 2026-10-07
- National Cancer Institute: Cervical Cancer Causes, Risk Factors, and Prevention; updated August 2, 2024
Sources checked: 2026-10-07
- NHS: England: local stop-smoking services
Sources checked: 2026-10-07
General population education, not interpretation of an HPV or cervical test, treatment, vaccine eligibility, screening schedule or individual prognosis.