Colorectal cancer is the second most common fatal malignancy in the United States and resulted in approximately 55,000 deaths in 1997. Because a reduction in cancer mortality with the use of screening sigmoidoscopy has been demonstrated in two case-control studies, screening of patients older than 50 years with flexible sigmoidoscopy has been recommended by the U.S. Preventive Services Task Force. Despite this recommendation, it is estimated that most patients are not being screened. Cooper and associates surveyed primary care physicians to identify characteristics of physicians who perform screening with flexible sigmoidoscopy and to identify barriers to initiation of screening.

The survey sample was obtained from the American Medical Association master file and included both family physicians and general internists from 10 states. A random sample of 1,000 physicians from each specialty was selected. All 2,000 physicians were mailed a questionnaire that asked if they screen for colon cancer with flexible sigmoidoscopy and if they do screen, whether they perform the tests themselves or refer patients to another physician. The physicians were asked to indicate reasons for screening or not screening.

The survey was initially mailed in the Spring of 1994 and, if there was no response within six weeks, a second survey was sent. If the second mailing got no response, a telephone call was made to verify the physician's address, and a third mailing was sent to eligible physicians who were willing to participate in the study.

From the original study sample of 2,000 physicians, 238 were eliminated because they were not practicing a primary care specialty. A total of 884 of the remaining eligible physicians responded, including 51 percent who were internists and 49 percent who were family physicians. Eighty-five percent were men, 76 percent were board certified and 89 percent were engaged in office practice.

Of all responders, 90 percent reported recommending flexible sigmoidoscopy to their patients. Forty-four percent performed the test themselves, and 46 percent referred their patients for the procedure. The overall use of screening did not correlate with gender, board certification status, year of medical school graduation or specialty. However, the physicians who performed sigmoidoscopy were more likely to be men, to have graduated after 1970, to be board certified and to specialize in family practice. The most frequently cited barrier to screening was poor patient acceptance (more than 40 percent of respondents). This was followed by lack of training, excessive time required, lack of equipment, lack of reimbursement and, finally, questionable effectiveness, which was cited by a minority of physicians.

The authors conclude that the majority of primary care physicians recommend colorectal cancer screening for their patients, although at least one half do not perform the procedure themselves. The authors suggest that targeting and training older, non–board-certified primary care physicians may be one way to increase flexible sigmoidoscopy screening for colorectal cancer. They also advocate improved patient education and population-based interventions as ways to improve the low rate of public acceptance of flexible sigmoidoscopy.