There was an interesting piece in the New York Times yesterday about mammograms (http://well.blogs.nytimes.com/2011/10/24/mammograms-role-as-savior-is-tested/?ref=health). A key question for public health experts to address is what proportion of women who obtained a screening mammogram and for whom the screening mammogram detected the cancer had their clinical outcome changed by the detection.
Why wouldn't the clinical outcome be changed by detection? The article states it quite well. Some cancers never would have amounted to anything. They are very small growths that never would cause a problem. Yet, once they are detected on a mammogram, they are treated. This is actually referred to as "overdiagnosis".
A second group is so aggressive that they can't be stopped.
A third group is more aggressive than the overdiagnosed group but not so aggressive that the screening really did anything.
It is described in the article as "only those that are aggressive enough to be potentially dangerous but that are found at just the right time to stop them from being dangerous" actually have their clinical course changed.
A difficulty is that we can't always tell which is which when the screening first occurs. This does suggest that we use a lot of resources for (and have a lot of concerns about) mammograms that don't always live up to the expectations that patients have.
It creates an interesting set of resource-related questions.
Lemon Zest, Turkish Apricot Scones
2 years ago