Related Experiment Video
Updated: Aug 6, 2025
An R-Based Landscape Validation of a Competing Risk Model
Published on: September 16, 2022
Validation of the International Breast Cancer Intervention Study (IBIS) model in the High Risk Ontario Breast
Jennifer D Brooks1, Ashley Mah1, Rebecca A G Christensen1
1Dalla Lana School of Public Health, Division of Epidemiology, University of Toronto, Toronto, Ontario, Canada.
Purpose:
Women with a remaining lifetime risk of breast cancer of ≥25%, estimated using the International Breast Cancer Intervention Study (IBIS) model, were eligible for the High Risk Ontario Breast Screening Program. This study examined the performance of IBIS 10-year risk estimates in the program.
Methods:
This retrospective study included 7487 women aged 30 to 69 years referred to the High Risk Ontario Breast Screening Program between July 1, 2011, and December 31, 2016, with follow-up until December 31, 2018. Model calibration and discrimination were assessed. Analyses were conducted overall and stratified by age (< or ≥50 years). Different 10-year risk thresholds were compared with the current eligibility criteria.
Results:
Overall, IBIS overestimated the risk of breast cancer with an expected vs observed case ratio of 1.17 (95% CI = 1.04-1.35). Overestimation was highest in women aged 50 to 69 years (expected vs observed case ratio = 1.29, 95% CI = 1.03-1.69) and for those in the top quartile of risk. Overall discrimination was fair with a concordance statistic of 0.66 (95% CI = 0.63-0.70). Furthermore, when using different 10-year risk eligibility thresholds, most cases would have been missed in the 30 to 49 age group using the 8% 10-year risk threshold, whereas relatively few women aged 50 to 69 would have been ineligible at any of the thresholds examined.
Conclusion:
We found that IBIS overestimated the risk of breast cancer in this screening cohort but had adequate discrimination. Age-specific risk thresholds should be considered to optimize the program eligibility criteria.
