Population-based service mammography screening: the Icelandic experience

Abstract

Objective

This study analyzes the efficacy of the Icelandic population-based service mammography screening.

Material and methods

Women aged 40–69 were invited for screening at 2-year intervals starting in November 1987. The study evaluates: (A) attendance and other screened performance parameters during 1998–2010; (B) trends in age-standardized and age-specific incidence rates during 1969–2010 and mortality rates during 1969–2010; and (C) distribution of risk factors and disease specific death rates according to mode of detection.

Results

(A) In the age group of 40–69, the average 2-year attendance was 62%, recall rate was 4.1%, needle biopsy rate was 1.3%, surgery rate was 0.6%, invasive cancer rate was 0.4%, and ductal carcinoma in situ (DCIS) rate was 0.06%. (B) The linear incidence trend after the start of screening decreased significantly in the age group 40–49, increased significantly in the age group 50–69, but decreased non-significantly in the age group 70–79. The decreased age-specific incidence in the 70–79 age group was, however, greater than the increased age-specific incidence at the ages 50–69. The mortality rate decreased 41% for all age groups and the linear mortality trend decreased significantly at ages 40–49, 50–69, and 70–79. In the age group 40–74 years, the age-specific mortality decreased by 6.9 cases per 2000 during a 10-year period. (C) Screen-detected women had significantly smaller tumors, more favorable tumor grade, fewer axillary metastases and, after correction for other risk factors, the likelihood of dying from cancer decreased 54% (hazard ratio: 0.46; 95% confidence interval: 0.31–0.69) for these patients compared to cases of nonparticipators.

Conclusion

The study results confirm acceptable rates of recalls and referrals for further diagnosis and treatment, and significantly decreased breast cancer mortality rate after starting screening.

Keywords:

• mammography
• screening
• breast cancer
• mortality
• incidence
• mode of detection
• risk factors

Acknowledgments

This work has been enabled by the technical assistance of many co-workers at the Icelandic Cancer Detection Clinic and the Icelandic Cancer Registry.

Disclosure

The authors report no conflicts of interest on this work.