Risk Assessment Models
Breast cancer risk models are tools for determining breast cancer risk in populations to guide screening and prevention decisions, but they have significant limitations that can lead to underestimation or misclassification of risk, particularly among women of color, younger women, and those with complex risk profiles.​
Risk Model Strengths
​Key strengths include:
​​​
-
Risk Stratification: Categorizes individuals into general risk levels, providing a starting point for identifying those who may require further evaluation.
​​
-
Data Integration: Merges diverse risk factors to create a broader understanding of risk trends, serving as a guide for further assessments.
​​
-
Support for Early Detection: Facilitates targeted prevention and screening efforts by identifying some higher-risk groups, even though personal risk factors may not be comprehensively addressed.
​​
-
Foundation for Personalized Care: Serves as a baseline for patient-provider discussions, encouraging tailored options beyond the model’s scope.
​​
-
Resource Prioritization: May assist in directing supplemental screening resources, such as MRIs or ultrasounds, to individuals with the highest risk based on aggregated data.
​​
-
Evolving Accuracy: Many models are being updated to include emerging data and technologies, enhancing their overall effectiveness.
Risk Model Limitations
​Key limitations include:
​​​
-
Focuses on averages, not individual complexity: All models focus on population averages, overlooking the unique combination of risk factors that make up personal risk.
​​
-
Biased or limited data inputs: Many models are based on predominantly white, postmenopausal populations, leading to reduced accuracy in predicting risk for Black, Hispanic, Asian, Indigenous, or premenopausal women.
​​
-
Incomplete risk factor inclusion: Some models omit crucial predictors such as breast density, prior breast biopsies, reproductive history, lifestyle factors (e.g., alcohol use), and hormonal exposures.
​​
-
Family history oversights: Several models inadequately capture extended family history, such as paternal lineage or second-degree relatives, or do not account for genetic mutations unless explicitly tested.
​​
-
Exclusion of prior breast cancer history: Most models are designed for women without a history of breast cancer and should not be used for survivors, leaving a gap in risk assessment for recurrence or new primary cancers.
​​
-
Outdated algorithms: Some models employ static algorithms that are not regularly updated to reflect evolving evidence or emerging risk factors, such as polygenic risk scores, newer imaging technologies, or social determinants of health.
​​
-
Variable inclusion of race/ethnicity: Some models include a crude adjustment for race, which may introduce bias rather than eliminate it, particularly when race is used as a proxy for genetic or structural factors.
​​
-
Inconsistent outputs: Different models may give different risk levels for the same individual, making it difficult for patients and providers to interpret and act on results with confidence.
Know the Model
Understand the strengths, limitations, and applications of breast cancer risk assessment models to make informed decisions and plan personalized care.