[Skip to Content]
[Skip to Content Landing]
November 1, 1985

Sensitivity and Specificity in Silicone Breast Models

Author Affiliations

Columbia Presbyterian Medical Center New York

JAMA. 1985;254(17):2409. doi:10.1001/jama.1985.03360170049013

To the Editor.—  The use of silicone breast models,1 the better to train physicians in the detection of breast lumps, is well meaning but of uncertain value in clinical care. However, this letter is intended to point out that certain techniques will actually make early detection of breast cancers less effective. The problem with currently available models is that they place "tumors" in a uniform matrix designed to simulate the normal breast. The ability to detect lesions of varying size, depth, and consistency in such a model is irrelevant, and the fact that fingertips (as opposed to pads) proved to be more effective in the present study illustrates this point. Successful breast examination requires a method that will not only emphasize abnormal masses but will also tend to subtract the multiple and variable masses that make up the normal breast. For that reason experienced examiners have always used finger