A Review of the Statistical Evidence on the Association Between Smoking and Lung Cancer

Abstract Leading investigators generally agree that a significant part of the observed increase in lung cancer mortality represents a real increase in the rate at which lung cancer is developing in the population. This increase cannot reasonably be attributed to genetic change in the human population and therefore must be due to environmental factors. Available evidence linking tobacco smoking to lung cancer is fairly extensive and impressive: (1) The increase in lung cancer mortality has been generally parallel to an increase in cigarette consumption; (2) In each of 14 case history studies there was a smaller percentage of non-smokers and a higher percentage of heavy smokers among lung cancer patients than among comparable controls; (3) Preliminary results of two population studies indicate higher mortality from lung cancer among smokers than among non-smokers and a still higher mortality among heavy smokers; and (4) At least one team of investigators has produced skin cancer in animals with condensates of tobacco smoke. There is disagreement whether the evidence at hand warrants a conclusion that smoking and lung cancer are causally related. The relative importance of smoking, air pollution, and occupational exposure to cancerigenic materials remains to be established.