medpundit |
||
|
Saturday, October 29, 2005Addressing a major unknown in the longstanding debate over mammograms, a new study sponsored by the National Cancer Institute found that the screening test contributed to a pronounced drop in the death rate from breast cancer. The study, being published today in The New England Journal of Medicine, concludes that 28 to 65 percent of the sharp decrease in breast cancer deaths from 1990 to 2000 was due to mammograms. The rest was attributed to powerful new drugs to treat breast cancer. That's quite a point spread. Maybe 28% reduction, maybe 65%. It all depends on how you parse the numbers: " An unprecedented statistical assault indicates that mammography screening has made a significant contribution to the decline in breast cancer mortality in recent years -- an issue of contentious running debate. Seven teams of statisticians -- using the same data sets but different statistical models -- reached substantial agreement that mammography screening has played a role in the 24% decline in the death rate between 1990 and 2000. Depending on the model, early detection by mammography screening contributed between 28% and 65% of the decrease, with the rest attributed to adjuvant therapy. The median contribution of mammography was 46%.... Statistical gymnastics of this sort are beyond the scope of this poor blog, but a 28% to 65% spread in results does not inspire confidence. P.S. I do believe that mammograms help reduce the mortality of breast cancer in that they help detect some cancers that can be treated successfully when found early. I'm just not convinced that this study is the slam-dunk the media reports make it out to be. posted by Sydney on 10/29/2005 12:40:00 AM 0 comments 0 Comments: |
|