Audit of a breast cancer screening programme using clinical examination and lightscanning.

To ascertain the accuracy of a breast cancer screening programme which relied on clinical examination and lightscanning, women attending between 1984 and 1989 were retrospectively followed-up. Three thousand, seven hundred and twenty women, median age 40 years, made 5,150 visits for screening. Two hundred and sixteen women (4.2%) were positive on screening but only six were found to have breast cancer. The sensitivity of clinical examination was 46%. The likelihood that a lesion found on clinical examination was a breast cancer rose with age, from 1.1% in women under 45 years to 33% in women over 54 years of age. Lightscanning had a sensitivity of 8%, but did not make an independent contribution to diagnosis. Sixty-five asymptomatic women without clinical abnormalities had false positive lightscans requiring subsequent investigation. This screening programme has demonstrated a demand for breast cancer screening from women of all ages. However the yield of clinical examination was low, and lightscanning proved to be an unsatisfactory screening test. Routine audit must be an integral part of all screening programmes.