Does quitting smoking lower acute myeloid leukemia risk?

01

A ‘30 years’ headline leaves out the uncertainty

A Minnesota case-control study published in 2013 compared people with myeloid leukemia and population controls. In the AML cessation rows of its Table 3, people who had quit for at least 30 years had an adjusted odds ratio (OR) of 0.64, with a 95% confidence interval of 0.39–1.06, relative to current smokers. The comparison table keeps that reference group fixed. It is not a comparison with never smokers.

Odds means the probability of an outcome divided by the probability of its absence; an OR compares those ratios between two groups. So 0.64 is not a 64% probability of AML. The interval includes 1, the value representing no difference in odds. This estimate therefore remains imprecise: it neither proves that cessation has no benefit nor demonstrates equality with never smokers. The study's broad duration groups cannot identify a biological switch at the thirtieth anniversary. Cases and controls were selected by disease status, so their proportions also cannot supply your absolute cancer risk.

[2]

02

Why the overall conclusion can be stronger than one table

The Surgeon General's conclusion combines evidence rather than relying on this one study. Its AML discussion reports a pooled analysis of three case-control studies: quitting for more than 20 years, compared with continuing smoking, had an OR of 0.59 (95% confidence interval 0.45–0.78). That is a different analysis with different duration categories, not a contradiction to be solved by choosing whichever number sounds most reassuring.

The Minnesota investigators used recalled smoking histories assessed up to two years before questionnaire completion to reduce distortion from illness-related quitting. Recall and selection still matter: 230 potentially eligible myeloid-leukemia cases died soon after diagnosis and were not referred. That number covers the wider myeloid-leukemia recruitment, not AML alone. The study did not randomize quitting, and its estimates cannot determine the cause of one person's illness or the precise course of their risk.

[1][2]

03

AML risk and today's blood count answer different questions

AML is a cancer of blood and bone marrow with different subtypes; it is not a name for every leukemia. A study of new AML diagnoses does not establish outcomes for chronic myeloid leukemia, remission after treatment or recovery of blood-cell counts. NCI explains that blood and bone-marrow examinations are used in diagnosis: years without cigarettes cannot replace those clinical findings.

If a test result or persistent symptoms prompted this search, have them assessed by a qualified clinician rather than classifying them as harmless because you have quit. Keep the two questions separate: ‘What does this finding mean?’ and ‘Where can I obtain support to stop smoking?’ In England, the NHS provides a route to local stop-smoking services. This page needs no symptoms, laboratory values or smoking history from you.

[3][4]

04

What to keep in mind

• Reduced AML risk is not treatment of an existing leukemia.
• A duration estimate needs both its comparison group and confidence interval.
• Quit years cannot interpret a blood count.