University and medical education and the risk of cancer in Sweden.

Socio-economic gradients are known to exist in cancer but we want to focus on the university-educated population and specifically on physicians to find out whether their special educational background on cancer causation helps them to avoid cancers. The analysis was based on the latest update of the Swedish Family-Cancer Database, in which the educational level was obtained from the national census of 1970 for those aged 30 years or older and the follow-up for cancer was extended up to year 2000. We determined standardized incidence ratios (SIRs), adjusted for several variables, for cancer among men and women in four educated groups and compared them with those with less than 9 years of education. Total cancer risks did not differ much, but at individual sites, the university-educated population showed consistent, increasing or decreasing trends. The educated group showed high SIRs for melanoma and skin cancer and for female breast cancer. At all these sites, SIRs for in situ tumours exceeded those for invasive tumours; the highest SIR was 4.81 for male MD, PhDs for in situ melanoma. SIR for in situ breast cancer for female physicians was 1.95. SIR for non-Hodgkin's lymphoma for male MD, PhDs was 2.20 but their risk of stomach cancer was only 0.26. Tobacco-related cancers were decreased among the educated group. Cancer risks for physicians were not different from those of their academic colleagues. Some of the increased risks were probably due to lead-time bias, caused by early diagnosis.