In this example, the prevalence of breast cancer is 4%. Sensitivity is 90% and specificity is 96%.
Overall, this results in a positive predictive value of 8%.
This example assumes that out of 10,000 women who were screened, 436 had abnormal findings, 36 cancers were confirmed by biopsy, and the mammograms of 4 women appeared normal despite the presence of cancer.
test, the less likely an individual with a positive test will be free from disease and the greater the positive predictive value.
When the prevalence of disease in those without signs or symptoms is low, the positive predictive value will also be low, even using a test with high sensitivity and specificity. For such rare diseases where there will be few true positives, a large proportion of those with positive screening tests inevitably will be found not to have the disease upon further diagnostic testing (Box 2-1b). One way to increase the positive predictive value of a screening test is to target the screening test to those at high risk of developing the disease, based on considerations such as demographic factors, medical history, or occupation. For example, mammograms are recommended for women over age 40, because that population has a higher prevalence of breast cancer.
Because the ultimate purpose of screening is to save lives by detecting cancer sufficiently early for effective curative treatment to be administered, screening effectiveness must be measured in terms of reduction in cancer mortality. However, because the death rate from breast cancer in “healthy” women who qualify as participants in a screening trial is relatively low (around 1/10 of 1 percent per year) it requires many thousands of women