medpundit |
||
|
Saturday, July 31, 2004MRI had a statistically significant better overall discriminating capacity than mammography. The sensitivity for detecting invasive breast cancer ranged from 18 percent for physical exam, 33.3 percent for mammography to 80 percent MRI. The corresponding specificities of the three procedures were 98 percent, 95 percent and 90 percent. What does that mean? The sensitivity of a test is the likelihood that a test will detect a disease if it's present. So, the lower the sensitivity, the higher the false negative rate. (Amazing, isn't it, how low the sensitivity is for mammograms, yet every day radiologists get sued for "missing" breast cancers.) In this category, MRI certainly seems to be superior. If there's a cancer there, it's more likely to see it. And that makes sense, for it's a better imaging technique. However, there is a draw back, and that's in specificity, the likelihood of a person without the disease having a negative test. That's 90% for MRI's, which doesn't sound bad, either, until you realize that it also means that 10 percent of women without breast cancer would have a postive MRI. That's a fairly high false-positive rate, especially for a test with a four digit price tag. And that's in a population of women with a higher-than-normal incidence of breast cancer. Those numbers for specificity and sensitivity change when the incidence of a disease within a population changes. In the general population, where the incidence of breast cancer is much lower than the study group, there would be many more false positives. Which is why mammography is still the screening method of choice in the general population. posted by Sydney on 7/31/2004 08:31:00 AM 0 comments 0 Comments: |
|