Confounding Effects Of Cigarette Smoking Research Articles

Reevaluation of the Confounding Effect of Cigarette Smoking on the Relationship between Alcohol Use and Lung Cancer Risk, with Larynx Cancer Used as a Positive Control

Background. The effect of smoking on lung cancer risk has been well documented, while the effect of alcohol remains controversial. We examined the hypothesis that the apparent association between alcohol intake and lung cancer risk is fully due to the confounding effect of cigarette smoke.Methods. Our sample of hospitalized patients included 2,953 male and 1,622 female lung cancer cases; 521 male and 159 female larynx cancers cases; and 8,169 male and 4,154 female controls, admitted to participating hospitals between 1981 and 1994. All controls had been diagnosed with non-smoking-related diseases. Larynx cancer was used as a positive control for lung cancer. Relative risks were estimated through odds ratios, adjusted through multiple logistic regression.Results. Although the odds ratios for alcohol had been significantly elevated prior to adjustment for smoking (OR = 2.4, 95% CI = 2.0–2.8), alcohol had no effect on lung cancer following this adjustment (OR = 1.2, 95% CI = 1.0–1.4). By contrast, the effect of alcohol on larynx cancer remained high even after adjustment for smoking (OR = 5.6, 95% CI = 3.7–8.6).Conclusion. The often-reported association between alcohol and lung cancer risk can be fully explained by the confounding effect of cigarette use.

Adjustment for the confounding effect of cigarette smoking in an historical cohort mortality study of workers in a fiberglass manufacturing facility.

In 1986 a statistically significant lung cancer SMR based on U.S. white male national mortality rates was reported for male fibrous glass workers for follow-up through 1982 of a cohort of U.S. man-made mineral fiber workers. The Newark, Ohio, plant of Owens-Corning, which comprised 38% of the fibrous glass workers in that cohort, also exhibited a statistically significant lung cancer standardized mortality ratio based on U.S. white male mortality rates. A case-control study of the Newark workers demonstrated that a history of cigarette smoking and not exposure to respirable glass is the most important factor in lung cancer risk for workers at the Newark plant. We provide an estimate of the extent of confounding by cigarette smoking for the Newark plant nationally based lung cancer standardized mortality ratio with data not previously available and which suggests that adjusting for the confounding effect of cigarette smoking could reduce the lung cancer standardized mortality ratio to a non-statistically significant level.