1. Home>
2. Specialties>
3. General Medicine>
4. Summary and Comment

INTERPRETING MAMMOGRAMS: ONE SIZE DOESN'T FIT ALL.

A two-part study of mammography proves the maxim that no test result should be interpreted in isolation of clinical data.

The researchers first studied 28,271 California women who had a first screening mammogram between 1985 and 1992, of whom 238 were found to have breast cancer during the next one to two years. Allowing 13 months for detection of breast cancer, the sensitivity of the screening mammogram was 90 percent overall, ranging from 77 percent among women aged 30 to 39 to more than 91 percent for women over 50. When analysis was restricted to invasive cancers (excluding ductal carcinoma in situ), sensitivities were 58, 75, 92, 93, and 87 percent, respectively, for women in their 30s, 40s, 50s, 60s, and older. Sensitivity was only 69 percent among women under 50 with a family history of breast cancer, possibly because of faster-growing tumors. Notably, 59 percent of younger women who later presented with breast cancer did so within 13 months, versus only 39 percent of older women.

The second paper incorporated specificity data and offers guidelines for interpreting abnormal mammograms. Because the risk of breast cancer increases with age, the probability of breast cancer in a woman whose first mammogram is read as "additional evaluation needed" (as were 93 percent of abnormal mammograms in this study) is only about 1 percent for women in their 30s, increasing to 7 percent for those over age 70. For tests read as "suspicious for malignancy," the risk rises to 9 percent for women in their 30s and 63 percent over age 70.

Comment: These analyses suggest that clinicians should neither be overly reassured by negative mammograms among young women, nor overly alarmed by marginally abnormal results since the test's diagnostic performance is weak in this population. In older patients, however, both positive and negative test results are more likely to be accurate.

— TH Lee

Published in Journal Watch General Medicine July 26, 1996

Citation(s):

Kerlikowske K et al. Effect of age, breast density, and family history on the sensitivity of first screening mammography. JAMA 1996 Jul 3 276 33-38.

• Medline abstract (Free)

Kerlikowske K et al. Likelihood ratios for modern screening mammography: risk of breast cancer based on age and mammographic interpretation. JAMA 1996 Jul 3 276 39-43.

• Medline abstract (Free)