Occupation and industry classifications categorize occupations and industries into clearly defined groups. As such they provide a common basis for collecting, presenting, and comparing of labour statistics. Occupational classifications group people based on job and tasks performed, and are commonly used in sociology and population studies. Industry classifications group people based on the sector of economic activity in which they are employed and are mainly used for economic analysis.Although not primarily developed for use in epidemiological studies, occupation classifications, and to a lesser extent industry classifications, are often used in this field. Population-based epidemiological studies frequently include questions about job title and specific tasks, after which the information is coded using either national or international classifications.Reviewing the literature in the British Medical Journal, American Journal of Epidemiology, and International Journal of Epidemiology (published between 1995 and2000) indicated that information on occupation in epidemiological studies (n = 129), was mostly used as an indicator for social class (38%). In 27% of the studies, occupation was studied directly in relation to disease, and in 24% occupation was used to infer occupational exposure. In the remaining 11% of the studies occupation was treated as a confounding factor or used to describe the study population.Although widely applied in epidemiological studies, only limited methodological information is available on the use of job and industry classifications. The lack of a theoretical basis may hamper full exploitation of occupation information within epidemiology and limits the potential to optimize its reliability.In this paper we review the potential for occupation and industry classifications in epidemiology. The main classifications available are reviewed and different methods for coding occupation and industry are discussed. In addition, we will address issues of reliability of the coding process. Options for analysis Labour statisticsFor population-based studies, a classification can be used to describe the occupational profile of the study sample, which subsequently can be compared with that of the national population from which the study sample is drawn. Occupation and industry classifications are used in epidemiological studies to classify study subjects according to their job and subsequently to study risk by job, to infer social class indicators, or to infer exposure to specific agents through job-exposure matrices. However, documentation on methodological aspects concerning the use of occupation and industry classifications is sparse within epidemiology. This paper reviews the diverse applications of occupation and industry classifications in population-based epidemiological studies. The different classifications in use are discussed, and criteria are given for choosing a classification in an epidemiological study. Finally, the reliability of coding for occupation and industry is reviewed. A further ...

Despite evidence that smoking restrictions do not adversely affect restaurant sales,lA the industry and smokers often resist increased smoking restrictions in public establishments. Although effective smoking bans will eliminate exposure to environmental tobacco smoke (ETS) in restaurants, some claim that less restrictive policies are also effective. Previous measurements in restaurants and bars indicated high levels of ETS, with considerable variability among establishments. Siegel5 found 1.5 to 2.0 times higher ETS levels in restaurants than in offices or residences with at least 1 smoker. Levels in bars were 3 times higher than in restaurants. In 20 restaurants and bars in Vancouver, British Columbia, 6-hour measurements were taken from July to October 1995. Five establishments were nonsmoking restaurants, 11 were restaurants with a nonsmoking section (regulations required at least 40% nonsmoking seats) (restricted smoking), and 4 were bars with food service and unrestricted smoking. Measurement devices were placed on dining tables in the nonsmoking areas (where applicable), as close as possible to the smoking areas. Room dimensions were measured, the restaurant layout and ventilation system were sketched, and every 5 minutes the numbers of customers (a surrogate for the amount of cooking) and buming cigarettes were counted. The restaurants would not allow us to measure ventilation with a tracer gas. Van-couver, British Columbia. Andrea 't Mannetje is with

Nonsmokers who are heavy coffee drinkers may have a small excess risk of bladder cancer. Although these results cannot be attributed to confounding by smoking, the possibility of bias in control selection cannot be discarded. On the basis of these results, only a very small proportion of cancers of the bladder among nonsmokers could be attributed to coffee drinking.

No abstract