|
In a study published in the Journal of the National Cancer Institute, researchers found that radiologists who read a large number of mammograms (around 5,000) per year are significantly more likely to accurately detect breast cancer, compared to those physicians who read relatively few mammograms. The researchers arrived at their findings by comparing radiologists in the United Kingdom who traditionally read large volumes of mammograms to U.S. radiologists who are only required to read 480 mammograms per year to keep their certification from the American Board of Radiology. While an accompanying editorial in the journal points to other factors that may also contribute to a radiologist’s accuracy record, the study suggests that volume does play a key role in detecting breast cancer with mammography. Countries such as the United Kingdom (U.K.) and Sweden that have centralized mammography screening programs have high specificity and sensitivity when it comes to reading mammograms, according to lead researcher Laura Esserman, MD, MBA, from the University of California at San Francisco, and her colleagues. High specificity equals a low percentage of false positive mammogram results (i.e., "indicating" breast cancer when it does not exist). High sensitivity refers to a high number of true positives, or accurate breast cancer diagnoses. To determine whether centralized mammography screening programs in the U.K. produce more accurate mammography interpretation than the decentralized system in the United States, Dr. Esserman and her colleagues evaluated the performance of 194 U.K. radiologists and 60 U.S. radiologists. In the U.K., radiologists are required to read 5,000 mammograms per year while U.S. radiologists are only required to read 480 mammograms each year. The U.S. radiologists were divided into three groups: those who read fewer than 100 mammograms per month, those who read between 101 and 300 mammograms per month, and those that read over 300 mammograms per month. All of the U.K. radiologists in the study had read over 300 mammograms per month. The researchers found that the U.S. radiologists who read a lower number of mammograms per month had a significantly lower accuracy rate compared to both the U.K. radiologists and the U.S. radiologists who read more than 300 mammograms per month. Thus, Dr. Esserman and her colleagues conclude that the U.S. mammography system may warrant changes so radiologists are encouraged to read a higher number of mammograms and thus improve their detection rates. However, an accompanying editorial by Dr. Joann Elmore of the University of Washington and Dr. Patricia Carney of Dartmouth Medical School in New Hampshire argue that there are multiple factors that may influence a radiologist’s breast cancer detection rate. These factors include enjoyment derived from interpreting mammograms, comfort level when reading mammograms, fear of being sued for malpractice, etc. Yet, the editorial authors acknowledge that mammogram volume logically plays a role in the accuracy of breast cancer detection. In the United States, mammography may only be performed and interpreted by a licensed physician who meets the MQSA (Mammography Quality Standards Act) minimum standards. Some radiologists have further training (fellowships) in breast imaging. Radiologists with fellowship training in breast imaging or women's imaging (or have equivalent practical experience) tend to have more expertise when it comes to interpreting mammograms and other breast imaging exams than non-fellowship trained radiologists. In general, a radiologist with at least five years of experience in breast imaging is considered to have significant expertise in the field. Previous studies of U.S. radiologists have also found that those who read a higher number of mammograms tend to more accurately detect breast cancer. Women may wish to inquire about the facility’s (or physician’s) mammogram volume when making an appointment for a mammogram. A radiologist who reads 50 mammograms per day (devoting approximately one half of his or her workload to mammography) reads approximately 5,000 mammograms per year.
|
