“Tobacco use kills more than 5 million people per year,” according to the World Health Organization. In the United States, over 18% of deaths each year are attributed to smoking; about 50,000 of which is the result of second-hand smoke inhalation.
In January 1981, the British Medical Journal published a historically paramount study conducted between the years of 1966 to 1979, entitled “Non-smoking wives of heavy smokers have a higher risk of lung cancer”. During this study, 91,540 Japanese non-smoking wives were followed up for the period of time above mentioned (fourteen years) and mortality rates due to lung cancer were analyzed. A relationship was found between the amount that a husband smoked (the intensity of their addiction) and the mortality of their non-smoking wives due to lung cancer.
Several aspects of the study were interesting. The study was thorough in that it explored other possible cancers and illnesses relating to smoking and alcohol consumption, but the correlations were not statistically significant. The findings specified one risk – lung cancer in households with smoking husbands. Another important detail of the study was that spouses were questioned independently. The study was designed well in that it explored various illness and risk behavior possibilities in a huge number of subjects from varying backgrounds, which allows for the validity of trends found in the data. Also, the fact that it was designed with the use of independent questionnaires is vital, because this detail eliminates the possibility of bias due to the cultural pressures of male dominance, for example.
Ending with indisputably conclusive data on such a controversial and economically significant product requires a well-designed study that proves the epidemiological causality of the relationship between cigarette smoke and lung cancer. The aspects needed for a successful epidemiological causality study, as suggested by Mary Schneider, were included – a "dose-response relationship" between the severity of a husband's addiction and the development of lung cancer in his wife, almost 100,000 subjects, a strong statistical suggestion of the relationships between secondary smoke exposure and lung cancer. The results were indicative of the cause-and-effect relationship that researchers in the arena of disease are looking out for, although in this case it is suggestive as a risk rather than completely evidential. Takeshi Hirayama’s study on passive smoking has withstood the test of time, because after forty years his results are still reputable.
This study spurred greater awareness of the dangers of carcinogen-inhalation involved in cigarette smoking and the threats to health that second-hand smoke inflicts. This dangerous, addictive agent has affected so many lives regardless of if a person actively chooses to smoke or simply lives with a smoker. With 48.2 million smokers in the United States and 2.4 billion smokers worldwide, not only are these people put at risk but also those that live with or near them. This pushes the number of deaths due to cigarettes up much higher than the estimations might indicate.
Ginny, you highlight some important statistics on tobacco smoking and good points from the article. This was a fairly large study with historic importance. When thinking about causality, keep in mind that this was a prospective cohort study, not a randomized clinical study which has bit more weight on establishing causality. In science and epidemiology we typically talk about the Hill criteria for establishing causality. The Hill criteria are as follows: temporal relationship (did the exposure to the factor occur before the disease development?), strength of association (measured by relative risk or odds ratio), dose-response relationship (as the dose of exposure increases, does the risk of disease also increase?), replication of the findings in other studies, biological plausibility (does the relationship make sense biologically?), consideration of alternative explanations (ie. confounders), cessation of exposure (does the risk of disease decline when exposure to the factor is reduced or eliminated?), consistency with other knowledge, and specificity of the association (weakest of all the criteria). It might be helpful to think about the Hill criteria next time you read an article that tries to establish causality.
ReplyDelete