Use of multistage models to infer stage affected by carcinogenic exposure: example of lung cancer and cigarette smoking.

摘要: Statistical methods based on the Armitage-Doll mathematical model of carcinogenic process are presented for analyzing epidemiologic case-control studies cancer. These proposed to provide inferences regarding stage(s) in cancer at which exposure interest acts. An example these is given shows evidence that carcinogens cigarette smoke appear affect transition rates two separate stages development lung cancer, and relative magnitudes effects estimated. The data this analysis came from a European multi-center study results show that: (1) risk among continuing smokers compared nonsmokers same age decreases as started smoking increases, while rate stays fixed, result indicates effect an early stage process; (2) ex-smokers having duration with time since stopped, late process. Both shown be best described by hypothesis affects stages. estimated cigarettes' indicate largest proportion total lifetime due its effect, causes other than varies 23% men 1-10 cigarettes per day 6% those greater 30 day. findings imply preventive measures directed toward inducing stop would have potentially substantial payoff reducing future mortality.