Szymczak W
Zakładu Epidemiologii Srodowiskowej, Instytutu Medycyny Pracy, Lodzi.
Med Pr. 1999;50(6):517-36.
The role of data derived from epidemiological studies in the quantitative assessment of cancer risk has been constantly growing. That results from the fact that epidemiological studies are performed directly on humans so that it is possible to avoid bias very often associated with methods of extrapolation between animals and humans. Moreover, a slow but continuous increase in the number of well designed epidemiological studies with results which can be used to estimate dose-response relationship is being observed. In this paper, the methodological foundations of cancer risk modelling on the basis of epidemiological studies are discussed, and the dose-response relationship based on epidemiological study carried out in an occupational cohort of workers employed in a pulp and paper plant was estimated. The epidemiological study was performed by the team of the Department of Epidemiology, The Nofer Institute of Occupational Medicine, Lódź. Lung cancer was adopted as the effect of occupational exposure. In oder to estimate the increase in cancer risk induced by occupational exposure, the risk for the whole occupational life was calculated by employing the methods of competitive risks, essential relative risks were modelled using Poisson means for data divided into groups, and a Cox proportional-hazards model was used for single data. The analysis indicated, that the data from epidemiological studies render it possible to carry out a more thorough investigation as compared to data obtained from animal experiments. As to the cohort under study, a certain effect of age at which workers enter into exposure on lung cancer risk was observed. A risk 'outfit' the workers bring in at the beginning of their employment is clearly visible in older age groups. A general conclusion drawn from the study may be formulated as follows: epidemiological studies play a major role in assessing the effect of working conditions on the increase in the morbidity risk, however, they are not a panacea for all the problems emerging in the cancer risk assessment in exposed persons. Therefore, information obtained from epidemiological and toxicological studies should supplement and reinforce one another, and data from both sources should be used to obtain the best possible characteristics of risk cancer in humans.
流行病学研究得出的数据在癌症风险定量评估中的作用一直在不断增强。这是因为流行病学研究直接在人体上进行，从而有可能避免常常与动物和人类之间的外推方法相关的偏差。此外，正在观察到设计良好的流行病学研究数量在缓慢但持续地增加，其结果可用于估计剂量反应关系。本文讨论了基于流行病学研究的癌症风险建模的方法基础，并估计了在一家纸浆和造纸厂工作的职业队列中进行的流行病学研究的剂量反应关系。该流行病学研究由罗兹诺费职业医学研究所流行病学系的团队进行。将肺癌作为职业暴露的影响。为了估计职业暴露引起的癌症风险增加，采用竞争风险方法计算整个职业生命期的风险，使用泊松均值对分组数据建模基本相对风险，并对单个数据使用Cox比例风险模型。分析表明，与从动物实验获得的数据相比，流行病学研究的数据使得能够进行更全面的调查。对于所研究的队列，观察到工人开始接触时的年龄对肺癌风险有一定影响。在年龄较大的组中，工人在开始就业时所带来的风险“组合”清晰可见。从该研究得出的一般结论可表述如下：流行病学研究在评估工作条件对发病风险增加的影响方面发挥着主要作用，然而，它们并非解决暴露人群癌症风险评估中出现的所有问题的万灵药。因此从流行病学和毒理学研究中获得的信息应相互补充和加强，并且应使用来自这两个来源的数据以获得人类癌症风险的最佳可能特征。
