Completeness of occupational history and occurrences of work-related diseases.

The lack of complete occupational histories causes underestimates or overestimates of occupational mortality and morbidity. The problem is emphasized in cohorts with high turnover and in cross-sectional study designs. In the study of the effect of selective turnover on occupational mortality and morbidity complete occupational histories were acquired via questionnaire. The sample comprised 1 200 current and 1 800 alive former workers and the closest relatives of 450 deceased workers and was formed from three exposure-level cohorts (defined primarily according to the physical demands of the work) of 15 714 metal workers hired in 1950-1976. The occupational histories reported in the questionnaires were compared with those collected from employers' personnel records. The coverage of occupational histories was the most accurate for current and former workers with the longest duration of exposure and the least accurate for dead workers and the oldest age groups. The use of complete occupational histories and the combining of the similar levels of exposure through the lifetime produced clear differences in the occurrences of musculoskeletal, respiratory, and cardiovascular diseases between the three levels of exposure. The errors in person-years were smaller than the errors in the duration of exposure; the coverage varied from 51 to 94%. However, misclassification clearly caused cancer cases to be located in classes with too short a period of latency. About 30% of the deceased workers had reportedly entered the industry before the first follow-up year.