News and Views
Promoting Colorectal Cancer Screening: Which Interventions Work?
Article first published online: 22 JUN 2009
Copyright © 2009 American Cancer Society, Inc.
CA: A Cancer Journal for Clinicians
Volume 59, Issue 4, pages 215–217, July/August 2009
How to Cite
Pinkowish, M. D. (2009), Promoting Colorectal Cancer Screening: Which Interventions Work?. CA: A Cancer Journal for Clinicians, 59: 215–217. doi: 10.3322/caac.20023
- Issue published online: 9 JUL 2009
- Article first published online: 22 JUN 2009
Although colorectal cancer screening is highly effective in reducing mortality, only 60% of adults are up to date with the tests recommended by current guidelines. Consequently, researchers and clinicians are eager to find ways to improve screening rates. A recent study by John Z. Ayanian, MD, MPP, and colleagues assessed and compared the effectiveness of 2 strategies for increasing colorectal screening: (1) mailing reminders to patients who were overdue for colorectal cancer screening, and (2) issuing electronic medical record reminders to their physicians (Archives of Internal Medicine 2009;169:364–371).
Dr. Ayanian, professor of medicine and health care policy at Harvard Medical School, Boston, Massachusetts, and colleagues conducted a randomized controlled trial that included patients and physicians at 11 ambulatory care centers in the Harvard Vanguard Medical Associates (HVMA) network, a multispecialty group practice in the Boston area. HVMA has used an electronic medical record (EMR) system since 1997. The EMR system allowed the investigators to identify 59,181 patients aged 50 to 80 years who had visited a primary care physician (PCP) at HVMA at least once in the previous 18 months. Of this group, 37,321 (63%) patients had been screened for colorectal cancer according to the HVMA standard of care—with either flexible sigmoidoscopy in the previous 5 years plus fecal occult blood testing (FOBT) in the previous year or colonoscopy in the previous 10 years. The remaining group of 21,860 patients who were overdue for colorectal cancer screening, along with their 110 PCPs, were included in this study.
Patients were randomized to receive educational mailings or no mailings. Each person in the mailing group received a package with a cover letter from the chief medical officer of HVMA reminding them that they were overdue for screening. The package also included a pamphlet about colorectal cancer screening options, an FOBT kit, and telephone number they could call to set up an appointment for flexible sigmoidoscopy or colonoscopy. Two mailings were made as follows: one during the first month of the study and a second 6 months later to patients who were still overdue for screening at that time. The participating physicians were randomized to a control group or to receive electronic medical record reminders that appeared on screen during office visits with patients who were overdue for screening. These reminders enabled 1-click orders for endoscopy.
The primary outcome included patients' completion of an FOBT, flexible sigmoidoscopy, or colonoscopy during the course of the 15-month study. The secondary outcome was detection of colorectal adenomas.
Compared with patients who received no mailings, those who did get mailings had higher screening rates (44.0% vs 38.1%; P < .001). Older patients were more likely than younger patients to get screened in response to the mailings. Mailings increased screening rates by 3.7% among persons aged 50 to 59 years; 7.3% in those aged 60 to 69 years; and 10.1% in patients aged 70 to 80 years (Ptrend = .01). Screening rates did not differ by sex of the patients. The mailing had the strongest impact on FOBT use, which was used and submitted to the laboratory by 25.4% of the group that got the mailing compared with 20.4% of the group that did not get the mailing (P = .001).
Overall, electronic reminders to physicians had no significant effect on whether their patients underwent screening (41.9% for the electronic-reminder group vs 40.2% for the no-reminder group; P = .47). But among patients who had made 3 or more visits with their PCPs during the study period, screening was slightly (but not significantly) higher among those patients whose physicians did get reminders (59.5% vs 52.7%; P = .07). Of note, one-third of the patients in this study made no visits to their PCP during the 15-month study period. Adenoma detection rates were higher (not statistically significant) in the groups that received patient mailings (5.7% vs 5.2%; P = .10) and whose physicians received reminders (6.0% vs 4.9%; P = .09).
On the basis of these findings, the investigators concluded that mailed reminders are an effective means of increasing colorectal cancer screening rates in patients who are overdue for these tests and that electronic reminders to physicians may increase screening rates among their patients who make frequent visits. There was no significant interaction between the patient and physician interventions.
“Studies of reminders to patients and physicians with regard to colorectal cancer screening that were conducted 5 to 10 years ago did not directly compare the effectiveness of patient and physician reminders,” said Dr. Ayanian, who adds that the reminders to doctors in prior studies were based on paper records, not electronic records, as was the case in the current study. “Our ability to conduct the study was based on the existence of an electronic medical record system that allowed us to identify patients who were overdue for colorectal cancer screening. The size of our study was substantially larger than earlier studies,” he added.
“The results of this study add to the body of evidence about what kinds of intervention are effective for getting people screened,” said Michael Pignone, MD, MPH, associate professor of medicine in the School of Medicine at the University of North Carolina (UNC) in Chapel Hill, who was not involved in this study. “But I don't think that a study like this is generalizable to a system that does not have the ability to identify and then to mass communicate with people who are not up-to-date with screening,” he said. “That doesn't mean I don't think it's an important study. It means that more systems need to develop that kind of capability,” said Dr. Pignone, who is also codirector of the Program on Medical Practice, Sheps Center for Health Services Research at UNC and associated with the UNC Lineberger Comprehensive Cancer Center.
Dr. Pignone also notes that although the absolute difference in screening rates between the group of patients that got mail reminders and who did not was modest—about 6%—it is an important difference. “You can't judge the effect in isolation,” he said, “…although 6% doesn't sound like a big difference, you need to ask how much effort you gave to get the 6% difference.” Dr. Pignone explains that once the infrastructure is established, the cost per person for this intervention is low. Once patients who are overdue for screening are identified and matched to their addresses, “you can send one message to 10,000 people with the push of a button,” he said.
Dr. Ayanian said that the patients included in this study were selected because they had not taken earlier opportunities for screening. “Before our study, the screening rate was 63% in that population. We were testing these approaches [mailings and electronic reminders] to increase screening in the 37% of patients who were overdue,” he said, adding that the 63% screening rate in the HVMA population is a relatively good one compared with other groups in the country, but it is still far from optimal. “With our mailed reminders,” he said, “we improved it from 63% to 79%, but that leaves 20% that are not reached by current screening practices.” Dr. Ayanian said that other forms of public education and outreach must be tested as means of reaching patients. “Some may choose not to be screened, but the key is to make sure all patients are well informed,” he said.
Dr. Pignone said that interventions like those described in this study “catch the low-hanging fruit.” For patients who are not screened in response to this type of intervention, a slightly more expensive one, perhaps using so-called patient navigators who call patients directly, might be effective.
Looking at screening barriers in isolation is a mistake, said Dr. Pignone, because barriers often interact. He used the example of a doctor whose colon cancer screening counseling skills were weak. This physician will be further handicapped if he is unaware of which of his patients are actually overdue for screening, whether the insurance carrier will pay for the person to be screened, and what the copay will be. Ideally, interventions will address these multiple barriers, said Dr. Pignone.
Dr. Ayanian said that other studies have found that patient barriers to screening include privacy concerns, discomfort, and the bowel-cleansing process. Time away from work is a major concern for many people, and he notes that consultation with the gastroenterologist before the procedure has been eliminated by some medical groups to streamline the screening process. “That removes one time barrier, but substantial time and effort is still required on the day before and the day of the procedure,” he said.
The patchwork nature of the primary care that many people now receive in the United States was also identified by Dr. Ayanian as a barrier to effective screening. “So-called ‘minute clinics’ are reasonable for simple, urgent problems like UTI [a urinary tract infection] and minor injuries, but they are not designed to provide chronic-disease care or preventive services that are more complex than flu shots. We need a primary care system that provides a medical home for people to return to over time to get their screening services,” he said. 1