Environmental tobacco smoke and risk

Task 1
ARTICLE 1- Environmental tobacco smoke and risk of respiratory cancer and
chronic obstructive pulmonary disease in former smokers and never smokers in
the EPIC prospective study
Vineis, P., Airoldi, L., Veglia, F., Olgiati, L., Pastorelli, R., Autrup, H., Dunning, A., Garte,
S., Gormally, E., Hainaut, P. and Malaveille, C., 2005. Environmental tobacco smoke and
risk of respiratory cancer and chronic obstructive pulmonary disease in former smokers and
never smokers in the EPIC prospective study. Bmj, 330(7486), p.277.
The purpose of this paper is to find if there exists any association between
environmental tobacco smoke, plasma cotinine concentration, and respiratory cancer or
death. A nested case-control study was conducted on 303,020 volunteers chosen from the
European Prospective Investigation into Cancer (EPIC) cohort. The EPIC cohort consists of
more than 500,000 volunteers selected in 10 European countries, which includes men and
women, preferably within 35 to 74 years. Questionnaires were made to measure diet and
lifestyle of different countries. The participants chosen in this design had never smoked or
had stopped smoking at least 10 years ago. Although, 123479 participants among them gave
evidence of exposure to environmental tobacco smoke.
The proposed design analysed the relation between environmental tobacco smoke and
recently diagnosed cancers of bladder, lungs, oral cavity, pharynx, or larynx, or leukaemia. It
also took into account death due to respiratory diseases. People who were diagnosed with
cancer or died from the above conditions were treated as cases. Three controls were matched
per case, where matching was done for gender, age, smoking status, and months lapsed after

recruitment. The cohort was analysed using Cox’s proportional hazard models. In this design,
odds ratio and 95% confidence interval were computed for both conditional and
unconditional logistic regression models, both giving apparently identical results. Increased
odds ratio and hazard ratio were associated with tobacco smoke exposure. Differences in
risks between genders were insignificant. The study identifies that environmental tobacco
smoke poses a risk of developing lung cancer or other respiratory diseases, especially in ex-
smokers.

ARTICLE 2- Cigarette advertising and magazine coverage of the hazards of
smoking a statistical analysis
Warner, K.E., Goldenhar, L.M. and McLaughlin, C.G., 1992. Cigarette advertising and
magazine coverage of the hazards of smoking a statistical analysis. New England Journal of
Medicine, 326(5), pp.305-309.
This article focuses on the fact that the magazines that depends on revenues from
cigarette advertisements have lesser chance of publishing articles on hazards of smoking,
with special concern on magazines directed to women. The data was collected on 99 U.S.
magazines for years 1959 to 1986. In this study two hypotheses are considered accounting the
relation between dependence of the magazine on revenue generated from advertising
cigarettes, and the frequency at which the adverse health effects are published. The first
assertion being that acceptance of smoke advertising during a given year was negatively
related to the probability of publishing article on hazard issues. The second hypothesis being,
proportion of advertising revenues obtained from cigarette advertisement was inversely
related to the coverage of hazards. For both cases the dependent variable was set to 1 (if the
article was published in the given year) or 0 (otherwise). Few independent variables were
also added to the model. A logistic regression analysis was conducted to test the validity of
the hypotheses. The fixed effects model was used to confirm the results from the logistic
model. From the 1st analysis, for women’s magazine, presence of cigarette advertisement and
probability of publishing hazards were statistically significant. But significance disappeared
when covariates were included. From 2nd case, the hypothesis was strongly accepted. An 1%
increase in the revenue from cigarette advertisements decreases the probability of risk