"Balance Sheet" Weighs Benefits vs. Risks in Breast Screening
Tuesday, Dec. 01, 2015
An evaluation of mortality reduction and over-diagnosis rates among women in the Norwegian Breast Cancer Screening Program (NBCSP) found that the benefit-harm ratio between the two rates averages out to about one life saved for every two women over-diagnosed, according to research presented Monday.
In her presentation Solveig Hofvind, Ph.D., a researcher at the Cancer Registry of Norway, said the study was designed to estimate a "balance sheet" of the benefits and harms—a subject of continuing debate—to women targeted by the NBCSP.
Dr. Hofvind and colleagues used data from published studies using individual level data from the NBCSP to evaluate the reduction in breast cancer mortality compared to over-diagnosis.
Dr. Hofvind and colleagues determined that for every 10,000 women screened that between 53 and 61 lives were saved from breast cancer, while between 45 and 126 women were over-diagnosed. Furthermore, 1,590 showed false positive results with a non-invasive assessment, while 410 showed a false positive result with an invasive procedure.
They determined that the benefit-harm ratio between mortality reduction and over-diagnosis varied from 0.4 to 1.4 depending on mortality reduction and over-diagnosis estimates. The 0.4 ratio includes the lowest mortality and highest over-diagnosis estimates and represents about 2.4 women over-diagnosed per life saved from breast cancer, while the 1.4 ratio includes the highest mortality and lowest over-diagnosis estimates and represents less than one woman over-diagnosed per life saved.
Trying to come up with such a benefit/harm ratio needs to be approached with extreme caution, she said, because of the variability in the study results used for the estimates and because of different ways in which over-diagnosis can be measured.
"So which ratio do we expect and which ratio are we willing to accept?" she asked. And she also wondered "who we are" since a women in a physician's office worried about breast cancer might have a different perspective than a physician or health policy maker.
And considering the number of variables associated with the estimates used to come up with such a ratio she questioned whether attempting to come up with this kind of balance sheet is a valid exercise.
"Is it even appropriate to estimate a ratio between lives saved from breast cancer death and epidemiologic over-diagnosis?" she asked.
Research Evaluates False Negatives
In another study presented Monday, Hannah Perry, M.D., a radiology resident at Beth Israel Deaconess Medical Center in Boston, pointed out that published data regarding breast cancer and false negative mammograms doesn't reflect current U.S. practice, but instead relies on data from programs that follow a European screening model or use film-screening mammography.
"So we wanted to try to evaluate the false negative rate in screening mammography in an American academic institution," she said. "And going deeper, we wanted to identify any characteristics of these false negative cases that may help improve the sensitivity of our breast screening programs."
In the study, Dr. Perry and colleagues evaluated patients diagnosed with breast cancer during an 18-month period from 2011-2012 who had a negative mammogram within 15 months prior to diagnosis. Cases were classified as either true negative or false negative based on the consensus findings of five dedicated breast imagers.
Of the 125 patients who met the inclusion criteria, 97 cases were classified as true negative while 28 were false negative. Dr. Perry and her colleagues determined that the 22 percent false negative rate was within previously published rates of approximately 13 to 35 percent.
"In looking at the cases that were false negative, all were considered to be from reader error based on reader reviews of the five breast imagers," Dr. Perry said. "We concluded there were no distinguishing mammographic features among the false negative cancers.
"Moving forward we are going to be looking at more cases and trying to see if we can find any information about the distinguishing characteristics of these false negative cancers," she said.