Intended for healthcare professionals

Letters Passive smoking

Paper does not diminish conclusion of previous reports

BMJ 2003; 327 doi: (Published 28 August 2003) Cite this as: BMJ 2003;327:501
  1. Allan Hackshaw, deputy director (allan.hackshaw{at}
  1. Cancer Research UK and UCL Cancer Trials Centre, University College London, London NW1 2ND

    Editor–I am writing on behalf of members of the 2002 working group on involuntary smoking and cancer for the International Agency for Research on Cancer (IARC).1 We concluded that environmental tobacco smoke causes lung cancer among never smokers. The paper by Enstrom and Kabat2 does not diminish this conclusion or those of previous reports.35

    Enstrom and Kabat's paper was based on one of the 25 US states (California) in the American Cancer Society's prevention study. The relative risk of lung cancer in never smoking women married to ever smokers was reported as 0.99 (95% confidence interval 0.72 to 1.37), based on only 177 cases, whereas the IARC meta-analysis, based on 46 studies and 6257 cases, yielded an estimate of 1.24 (95% confidence interval 1.14 to 1.34).1 The estimate of Enstrom and Kabat is consistent with both an increased risk of lung cancer (the confidence interval includes the IARC estimate of 1.24) and no effect. Adding the result from Enstrom and Kabat to the IARC analysis reduces the pooled estimate to 1.23.

    The observed relative risk of 0.99 is based on the smoking status of husbands in 1959, but many would have quit by 1998, particularly in California. Table 8 in the paper confirms this; in 1959 63% of ever smoking husbands were current smokers compared with 26% in 1998. This exposure misclassification would mask the association between exposure to environmental tobacco smoke and lung cancer.

    IARC's classification of environmental tobacco smoke as a human carcinogen was based on the full scope of evidence; observational studies, carcinogenic components of environmental tobacco smoke, experimental models, and biomarker studies. Additionally, active smoking is an established cause of lung cancer, and knowledge of mechanisms of carcinogenesis implies no risk free level of exposure to tobacco smoke. Enstrom and Kabat's conclusions are not supported by the weak evidence they offer, and, although the accompanying editorial alluded to “debate” and “controversy,” we judge the issue to be resolved scientifically, even though the “debate” is cynically continued by the tobacco industry.


    • Competing interests None declared.

    • Members of the IARC Working group: Patricia Buffler (USA), Richard Doll (UK), Elizabeth Fontham (USA), Yu-Tang Gao (China), Prakash Gupta (India), Allan Hackshaw (UK), Elena Matos (Argentina), Jonathon Samet (USA), Michael Thun (USA), Kurt Straif (France), Paolo Vineis (Italy), H-Erich Wichmann (Germany), Anna Wu (USA), David Zaridze (Russia).


    View Abstract