To any woman having surgery for breast cancer, the words she most wants to hear in the recovery room are, “We got it all.” Having negative margins—which means there are no malignant cells within a few millimeters of the edge of the excised tissue—is associated with a lower risk that the cancer will recur, and thus a higher chance of survival. Having radiation after a lumpectomy or other breast-conserving surgery (that is, anything short of a mastectomy) is also associated with higher cancer-free survival.
And if you want to find surgeons who have the best track record on those important measures? You might as well throw darts at a printout of oncologists.
The difficulty that cancer patients face in finding excellent care has been a scandal for years, as NEWSWEEK described in a 2009 story. Now, a new study shows that for women diagnosed with ductal carcinoma in situ (DCIS), that difficulty poses a greater risk than anyone knew—making the refusal of medical-specialty groups, Medicare, and others to release data about how well different physicians adhere to guidelines, let alone how their patients fare, even more morally problematic.
Noninvasive breast cancer is usually treated with either breast-conserving surgery or mastectomy; breast-conserving surgery might or might not be followed by radiation. According to a paper released today in the Journal of the National Cancer Institute, however, health outcomes also depend on which surgeon treats you.
In an analysis led by Andrew Dick of the RAND Corporation, the two most important factors in keeping breast cancer from recurring after treatment for DCIS are negative tumor margins (specifically, no cancer cells closer than two millimeters from where tissue was removed) and having radiation after breast-conserving surgery. How important are these two factors? According to the researchers’ analysis of the medical records of 994 women with DCIS, breast cancer recurred in about 5 percent of women who had breast-conserving surgery with radiation, but 14 percent of those who did not have radiation. Among women with negative margins who got radiation, cancer returned in about 3 percent; among women who had positive margins and got radiation, cancer returned in about 15 percent. Among women with negative margins who did not get radiation, cancer returned in about 13 percent, compared with 25 percent of those with positive margins who didn’t get radiation.
That much wasn’t too surprising—without radiation therapy, women have a higher risk of the cancer returning; positive margins also raise that risk. More striking is that both these important determinants of outcomes varied markedly from surgeon to surgeon. How much difference does it make? The researchers estimate that if all surgeons were merely as good as the median in terms of leaving women with negative margins and treating with radiation, five-year recurrence rates could be cut by 22 percent.
“Results varied substantially by surgeon,” accounting for up to 35 percent of the difference in five-year recurrence rates, the researchers write. “The extent of variation [by surgeon] and its contribution to long-term health outcomes are troubling,” since these unexplained differences “could have profound implications for health outcomes.”
Now the rub: a woman trying to learn how her surgeon does on these crucial measures is out of luck. In an accompanying editorial, epidemiologist Beth Virnig and surgeon Todd Tuttle of the University of Minnesota ask rhetorically, “How should women select a provider knowing that up to 35 percent of the variation in outcomes is based on their choice of physician?” Not easily. There is no central repository for this information, Virnig told me, which is even more disturbing given that the choice of physician might be even more important than what treatment you get. “You might be better off getting breast-conserving surgery with this doctor than mastectomy—which ordinarily has a higher chance of cancer-free survival—with that doctor,” she says.
One way to help patients make an informed choice of physician would be to publish scores for all physicians performing breast-cancer surgery in a particular area, Virnig adds. But anyone who tries to pry out such information faces a Herculean task. Medicare refuses to make public any information having to do with how many cases of a particular disease a doctor sees, or how many operations a surgeon performs. When Consumers’ Checkbook, a nonprofit organization, tried to get Medicare to release such data, it was rebuffed, and lost in court in 2009. When NEWSWEEK tried to partner last year with the American Society of Clinical Oncology on a rudimentary database for patients—surveying ASCO members about how many cases of a particular cancer they’d seen in the last few years, whether they were board-certified, how long they had been in practice, and the like—we spent months on negotiations before being sent packing.
Consumers Union, publisher of Consumer Reports, made a significant breakthrough on this front when it persuaded the Society of Thoracic Surgeons (heart surgeons) last year to share at least some crucial data: 30-day mortality rates, complications such as serious infections, the number of procedures performed, and whether the patient was put on appropriate medications, Dr. John Santa of CU told me. But it took him two years to get the society to agree to even this much, and only 221 of its 950 member groups (a “group” can mean a single surgeon, or a practice consisting of several) will share their data.
Until the government releases physician data, or medical groups themselves do, patients will remain in the dark. For cancer surgery, the best database for now is one assembled by the American College of Surgeons’ Commission on Cancer. If you go here and choose “detail listing,” you can enter a town, state, or ZIP code, find cancer-treatment facilities nearby, and see how many cases of different forms of cancer (by type and stage) they have treated surgically on an annual basis. That way, at least you can avoid being operated on by someone who never sees a case like yours. But it is no substitute for the full range of information patients need to choose a physician to treat their cancer.