Audit procedures in the National Breast Screening Study: mammography interpretation

Area of Science:

• Radiology
• Oncology
• Public Health

Background:

• Mammography screening programs are vital for early breast cancer detection.
• Ensuring consistent interpretation of mammograms is essential for program quality and patient outcomes.
• Previous assessments of inter-reader agreement in screening mammography have varied.

Purpose of the Study:

• To evaluate the agreement in mammography interpretation between multiple screening center radiologists and a reference radiologist.
• To assess the quality assurance of a national breast cancer screening program through systematic audit procedures.
• To quantify inter-reader agreement using Kappa statistics in a large-scale screening setting.

Main Methods:

• A systematic audit of mammography interpretations was conducted within a national breast cancer screening program.
• Agreement was assessed using Kappa statistics between five screening center radiologists and one reference radiologist.
• Data included mammograms from 739 women without known breast cancer and 204 women with detected breast cancer.

Main Results:

• Moderate agreement was observed between center radiologists and the reference radiologist.
• Kappa statistics ranged from 0.409 (women without known cancer) to 0.493 (all women), indicating agreement beyond chance (p < .0001).
• This represents a systematic, simultaneous, and ongoing measure of agreement in a screening program.

Conclusions:

• The study demonstrates moderate, statistically significant agreement in mammography interpretation within the national screening program.
• These findings support the implemented audit procedures as a valuable tool for quality assurance.
• Continuous monitoring of radiologist agreement is recommended for maintaining high standards in breast cancer screening.