Life-long physical activity and cancer risk among Finnish female teachers

This cohort study on Finnish female teachers of physical education (PE) and languages (L) was performed as an attempt to evaluate the importance of physical activity as a risk determinant of cancer, PE representing a physically active and L a less active subcohort. All the PE and L teachers graduating from 1920 onwards were obtained from four registers compiled from 1958–73 supplemented by an extension cohort drawn from the union membership register 1984–91 (in total 1,499 PE and 8,619 L teachers). Results of a questionnaire to representative samples of L and PE teachers did not reveal any major intergroup differences in social status, general health status, nutrition, maturation history, reproductive history, consumption of alcohol, smoking, or diet. However, the PE teachers reported higher life-long physical activity values than the L teachers. During the follow-up period 1967–91, the number of cancer cases totalled 108 for the PE and 513 for the L teachers. Expected numbers of cancer cases were calculated on the basis of national incidence figures, and the standardized incidence ratios (SIRs) were defined as ratios of observed and expected numbers of cases. The SIR for total cancers was 1.1 for the PE and 1.2 for the L teachers. In both teacher groups the SIRs for cancers of the breast, endometrium, ovary and colon were similarly elevated and the SIR for cervical cancer was reduced. There was an increased SIR for lung cancer (1.4) and skin melanoma (2.0) in the PE but not in the L teachers (0.5 and 0.9, respectively). For cancers of the stomach, rectum, pancreas and kidney, the SIR for the PE teachers was less than one-half of that for the L teachers. Similarly, the SIR for breast cancer at premenopausal age was lower among the PE than L teachers. None of the differences between the SIRs for the PE and L teachers were statistically significant. The results suggest that the potential protective effect of physical activity, if any, is so small that a larger cohort of physically active women would be needed to demonstrate it.