Greater Interpretive Volume Leads to Greater Accuracy in Quebec Breast Cancer Screening Program
In a study reported in the Journal of the National Cancer Institute, Théberge et al evaluated the association between radiologist interpretive volume and breast cancer screening accuracy in the Quebec Breast Cancer Screening Program. They found that accuracy increased with increasing volume, with the improvement reflecting reduced rates of false-positive results.
Study Details
The study involved 1,315,327 screening mammograms performed by 340 radiologists in the Quebec Breast Cancer Screening Program between 2000 and 2006. The program is a population-based program that invites women aged 50 to 69 years to receive biannual screening mammography in accredited facilities. All ductal carcinoma in situ or invasive breast cancer diagnoses made in the 2 years after screening were identified by linking program data with other provincial databases.
A screening mammogram was classified as abnormal if the patient was referred for assessment and was otherwise classified as normal. The association of volume with sensitivity, false-positive rate, and accuracy (sensitivity/false-positive rate) was assessed by multivariable Poisson regression, with all models being adjusted for potential confounders, including participant characteristics, radiologists, and facilities.
Among the 7,915 women diagnosed with breast cancer, 87% had cancer detected during screening and 13% during the 1-year interval after screening. Among women without breast cancer, 9.4% of screening examinations were false-positive.
Accuracy
Compared with the 159 radiologists (47%) who maintained a total interpretive volume of ≥ 500 mammograms per year, the 29 radiologists (8.5%) who consistently had volume < 500 mammograms per year exhibited a 20% reduction in sensitivity (adjusted sensitivity ratio = 0.80, 95% confidence interval [CI] = 0.66–0.98) and a 91% increase in false-positive rate (adjusted false-positive ratio = 1.91, 95% CI = 1.20–3.04), yielding a 58% reduction in screening accuracy (adjusted accuracy ratio = 0.42, 95% CI = 0.24–0.74).
Accuracy progressively increased with increasing volume (P = .0005 for trend), with, for example, radiologists with volume of ≥ 4,000 having a 32% increase in accuracy compared with those having volume of 500 to 999 mammograms (adjusted accuracy ratio = 1.32, 95% CI = 1.13–1.54). The greatest increase in accuracy with increasing volume was with increasing volume up to approximately 3,000 mammograms per year. Accuracy increased by 0.19 per 100 additional mammograms per year at an annual volume of 1,000 mammograms and by approximately 0.05 per 100 additional mammograms at an annual volume of 3,000 mammograms per year.
False-Positive Rates
False-positive rates progressively decreased with increasing volumes (P = .001 for trend), with, for example, radiologists with volume of ≥ 4,000 having a 24% reduction in false-positive rate compared with those having volume of 500 to 999 mammograms (adjusted false-positive rate ratio = 0.76, 95% CI = 0.65–0.89).
Overall, there was little change in sensitivity according to volume (P = .68 for trend).
The investigators concluded, “The minimum annual volume of 500 mammograms required in North America is justified; radiologist accuracy may be compromised if interpretive volume is consistently less than this requirement. Raising interpretive volume may help to reduce the frequency of false positives without loss of sensitivity. Possible gains in accuracy may be greater with increases in volume of up to approximately 3,000 mammograms interpreted annually.”
Jacques Brisson, MD, ScD, of Centre Hospitalier Universitaire de Québec, is the corresponding author for the Journal of the National Cancer Institute article
The study was supported by the Ministry of Health and Social Services of Quebec.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
