Messing K, Dumais L, Courville J, Seifert A M, Boucher M
Centre pour l'étude des interactions biologiques entre la santé et l'environnement (CINBIOSE), Université du Québec à Montréal, Canada.
J Occup Med. 1994 Aug;36(8):913-7.
In the epidemiological approach to occupational cancers, large bodies of data must be analyzed to find rare cases of cancer. The exposure status of workers must therefore be assessed. Inaccuracies will lead to bias toward the null value in certain cases. Job title has often been used as a proxy for exposure status. This study was undertaken to examine content (ie, tasks and activities) associated with job title among men and women in a large Québec municipality. Occupational accident reports were studied for 1589 accidents, and 113 men and women workers were interviewed about job content. Women and men did not seem to have the same accident rates. From interview data, it appeared that women and men with the same job title did not perform the same tasks. Thus, they might have different exposures. The data reported here support caution in using job title to estimate exposure for both genders if the job-exposure matrix has not previously been validated separately by gender. In addition, it may be unwise to adjust relationships between job title and cancer incidence for gender, thus treating gender as a confounder when it may be a proxy for specific exposures.
在职业性癌症的流行病学研究方法中，必须分析大量数据以找出罕见的癌症病例。因此，必须评估工人的接触状况。在某些情况下，不准确的数据会导致偏向零值的偏差。工作头衔常常被用作接触状况的替代指标。本研究旨在调查魁北克一个大型城市中男性和女性与工作头衔相关的工作内容（即任务和活动）。研究了1589起职业事故报告，并就工作内容对113名男女工人进行了访谈。男性和女性的事故率似乎不同。从访谈数据来看，拥有相同工作头衔的男性和女性执行的任务并不相同。因此，他们可能有不同的接触情况。如果之前没有按性别分别验证工作接触矩阵，那么这里报告的数据支持在使用工作头衔来估计两性的接触情况时要谨慎。此外，将性别作为工作头衔与癌症发病率之间关系的调整因素可能是不明智的，因为性别可能是特定接触的替代指标，却将其当作混杂因素来处理。
