In contrast to effective treatment, the sensitivity of the screening test, that is, its ability to detect early cancer, may or may not be an important factor in determining the benefit from a screening program. If a screening test is made more sensitive (for example, by reducing the cut-point for defining "abnormal"), it is likely that the test will detect more cancers. However, if these additional cancers are either more
Disease Detectable by Detectable by Begins Screening Test Patient
Lead-Time figure 12.1. Natural history of cancer. Critical Point, point at which treatment becomes less effective; could be between A and B, between B and C, or after C.
benign (and would never cause problems) or more malignant (and thus have already metastasized), then the extra sensitivity would not have made a contribution to improving health outcomes. The operative question is not how many more cancers are found by a more sensitive test but rather whether screening has moved detection for at least some potentially fatal cancers back to a more treatable stage. If this has not occurred, then the more sensitive test has not been a useful addition to the screening program. This rationale includes such strategies as screening more frequently (i.e., reducing the screening interval), which may increase sensitivity but may or may not improve health outcomes.
For this reason, the sensitivity of a screening test may not be related to its ability to improve health outcomes. For example, screening for cervical cancer with the Pap smear probably has a fairly low sensitivity,7,8 yet screening every 3 years apparently reduces cervical cancer mortality by more than 80%.9 Developments in the technology of screening tests that seek to improve screening programs by increasing the sensitivity of the screening test may increase sensitivity without improving health outcomes. Such approaches may increase the cost of screening without providing additional health benefit.
Was this article helpful?