ORLANDO – When a woman with a suspicious ovarian mass presents to a primary care physician, the majority of these doctors would not refer the patients directly to a gynecologic oncologist, even though early management by such specialists is associated with improved outcome, Dr. Barbara A. Goff said.
In all, 52% of 414 internists and 40% of 591 family physicians who responded to a mailed survey indicated that they would refer a patient with a suspicious pelvic mass directly to a gynecologic oncologist. Overall, 98% indicated that they would refer or consult with another physician, but half would initially refer to an ob.gyn., Dr. Goff said at the annual meeting of the Society of Gynecologic Oncologists.
"It’s been shown in numerous studies that women who receive their care from gynecologic oncologists have a significantly higher likelihood of receiving NCCN guidelines [recommended] therapy, optimal cytoreduction, and better overall survival," Dr. Goff said. The National Comprehensive Cancer Network (NCCN) is a consortium of 21 leading cancer centers in the United States that regularly releases and updates clinical guidelines in oncology.
The survey findings suggest a need for greater awareness of the benefits of such direct referrals and, possibly, for incentives to get internists and family physicians to refer these women more often, said Dr. Goff, director of gynecologic oncology at the University of Washington, Seattle. "Promoting direct referral to gynecologic oncologists from primary care may be the best way to increase compliance."
A total of 596 ob.gyns. also responded to the vignette-based survey. Their answers differed from those of internists and family physicians when they were asked to consider how they would manage the same hypothetical patient. The scenario was a 57-year-old woman complaining of pelvic pain and bloating for 3 weeks, whose ultrasound reveals a 10-cm, complex, right adnexal mass with solid and cystic components and increased vascularity. Patient variables such as race and insurance status were changed in different versions of the survey.
About one-third of ob.gyns. (34%) indicated that they would perform surgery themselves. These ob.gyns. were significantly more likely to work in practices that were smaller and/or located in more remote places, according to a multivariate analysis. The other 66% responded that they would consult with or refer the woman to another physician, and 96% of these ob.gyns. would involve a gynecologic oncologist.
This combination of findings – that only about half of internists and family physicians would refer directly to a gynecologic oncologist, and about one-third of ob.gyns. would perform surgery themselves – may partially explain why many women with ovarian cancer in the United States do not receive comprehensive surgical care or get treated at a high-volume center, Dr. Goff said.
"Unfortunately, recent studies show that 30%-50% of women with ovarian cancer are not receiving care from gynecologic oncologists," Dr. Goff said. For example, 44% of 31,897 stage III/IV ovarian cancers were treated by a different type of physician (Gynecol. Oncol. 2010;117:18-22). Those women who were treated by a gynecologic oncologist had a 40% improvement in overall survival.
Dr. Goff also examined this phenomenon in her study that showed that 67% of 9,963 women with ovarian cancer who were admitted received comprehensive surgery (Cancer 2007;109:2031-42).
Dr. Goff and her associates looked for significant patient and physician factors that were associated with referral to a gynecologic oncologist. Private insurance was the only significant, unadjusted patient factor. Among the internists and family physicians, significant factors included female sex, internal medicine specialty, board certification, fewer years in practice, group practice, fewer patients seen per week on average, involvement in clinical teaching, and an urban practice location.
In a multivariate logistic regression, factors that were significantly associated with an internist or family physician’s not referring directly to a gynecologic oncologist included male sex, family physician specialty, Medicaid insurance, providers with a weekly average number of patients greater than 91, solo practice, and rural location.
The 12-page survey was mailed to 3,200 primary care physicians who were randomly sampled from the AMA master file. A $20 bill was included as an incentive. The response rate was 62%.
"This is an important and provocative paper," said invited study discussant Dr. Claes Trope, head of the national gynecologic oncology center at Oslo University Hospital.
The sample size is large, there seems to be no selection bias because a random sample of physicians was surveyed, and the "62% response rate is quite impressive," she said. An inability to judge whether physician responses completely reflected what the clinicians would actually do in practice, compared with a "politically correct" response, is a potential limitation, Dr. Trope added.