Mortality reduction in the Swedish mammography screening program.

Abstract

29 Background: The introduction of a nationwide mammography screening program in Sweden in 1986 was associated with a 50% persistent increase of breast cancer incidence (including carcinoma in situ). Whether screening has been instrumental in a recent decline in the breast cancer mortality remains an open question. Methods: We compared the 18 years breast cancer mortality for two groups of women; those diagnosed in the first 6 years of the screening program and those diagnosed during a 6 year period that included the last 4 years before screening started and the first 2 years of the screening program. The test group (n + 328,927) included women aged 40 to 69 at the first invitation to screening. On average the women in the age-matched control group (n + 317,404) were born 4 years earlier then the women in the test group. The numbers of incident invasive breast cancers in each county were obtained from the Swedish Cancer Registry (in-situ cancers were excluded) and the corresponding breast cancer deaths were obtained from the Swedish Causes of Death Registry. Results: After 18 years follow up, the mortality reduction was 14% in the test group. However, most of this difference (10%) occurred in the first five years after diagnosis reflecting improved medical treatment of breast cancer. Most of the 14% difference also disappeared if we adjusting for period effects. Conclusions: About 83% of women in the test group were also included in the control group making the two groups almost identical with respect to risk variables. By screening the control group at the end of a 6-year period, the results are adjusted for differences due to lead-time bias and by studying mortality instead of survival overdiagnosis (length time bias) is also adjusted for. We conclude that organized mammography screening is not an important cause for the reduction in the breast cancer mortality in Sweden.