FROM JAMA
Can outreach improve the globally low rates of adherence to colorectal cancer screening? Yes, according to two recent studies in JAMA; the studies found that both patient-focused and physician-focused outreach approaches can result in significantly better patient participation in colorectal cancer (CRC) screening.
The first study ( JAMA. 2017;318[9]:806-15 ) compared a colonoscopy outreach program and a fecal immunochemical test (FIT) outreach program both with each other and with usual care. The results of the pragmatic, single-site, randomized, clinical trial showed that completed screenings were higher for both outreach groups, compared with the usual care group.
The primary outcome measure of the study was completion of the screening process, wrote Amit Singal, MD , and his coauthors. This was defined as any adherence to colonoscopy completion, the completion of annual testing for patients who had a normal FIT test, or treatment evaluation if CRC was detected during the screening process. Screenings were considered complete even if, for example, a patient in the colonoscopy arm eventually went on to have three consecutive annual FIT tests rather than a colonoscopy.
A total of 5,999 patients eligible for screening were initially randomized to one of the three study arms. Across all study arms, approximately half were lost to follow-up. These patients were excluded from the primary analysis but were included in an additional intention-to-screen analysis. A total of 2,400 patients received a colonoscopy outreach mailing; 2,400 received FIT outreach, including a letter, the home FIT testing kit and instructions; 1,199 received usual care. Patients in both intervention arms also received up to two phone calls if they didn’t respond to the initial mailing within 2 weeks. Mailings and phone calls were conducted in English or Spanish, according to the patients’ stated language preferences (those whose spoke neither language were excluded from the study).
Of the patients in the colonoscopy outreach group, 922 (38.4%) completed the screening process, compared with 671 (28.0%) in the FIT outreach group and 128 (10.7%) in the usual care group.
Compared with the group receiving usual care, completion of the screening process was 27.7% higher in the colonoscopy outreach group and 17.3% higher in the FIT outreach group. Screening process completion was 10.4% higher for the colonoscopy outreach group, compared with the FIT outreach group (P less than .001 for all).
Dr. Singal, who is with the department of internal medicine at UT Southwestern Medical Center, Dallas, and his colleagues also performed several post-hoc secondary analyses. In one, they used a less-stringent definition of screening process completion in which biennial FIT testing was considered satisfactory. When this definition was applied, the colonoscopy outreach group had 0.5% lower screening process completion than the FIT outreach group. The chances of a patient receiving any screening during the study period was highest in the FIT group (65%), with 51.7% of those in the colonoscopy outreach group and 39% of those in the usual care group receiving any screening.
“FIT has lower barriers to one-time participation but requires annual screening and diagnostic evaluation of abnormal results,” wrote Dr. Singal and his colleagues.
Strengths of the study, said Dr. Singal and his colleagues, included the fact that the study took place at a “safety net” institution with a racially and socioeconomically diverse population. Also, the study design avoided volunteer bias, and offered a pragmatic head-to-head comparison of colonoscopy and FIT.
The second study took place in western France, and targeted outreach to physicians rather than patients (JAMA. 2017;318[9];816-84). When physicians were given a list of their own patients who were not up to date on CRC screening, investigators saw a small, but significant, uptick in patient participation in FIT screening.
One year after the reminders went out, FIT screening had been initiated in 24.8% of patients whose physicians had received the list, compared with 21.7% of patients of physicians who had received a more generic notice and 20.6% of patients whose physicians received no notification, according to first author Cedric Rat, MD, and his colleagues.
The study examined which notification approach was most effective in increasing FIT screening among the physicians’ patient panels: sending general practitioners (GPs) letters that included a list of their own patients who had not undergone CRC screening, or sending them generic letters describing CRC screening adherence rates specific to their region. A usual care group of practices received no notifications in this 3-group randomized cluster design.
Patients in the patient-specific reminders group had an odds ratio of 1.27 for participation in FIT screening (P less than .001) compared to the usual care group. The odds ratio for the generic reminders group was 1.09, a nonsignificant difference.
Between-group comparison showed statistical significance for both the 3.1% difference between the patient-specific and generic reminders groups, and for the 4.2% difference between the patient-specific and usual care groups (P less than .001 for both). There was no significant difference between the generic reminders group and the usual care group.
Dr. Rat, professor of medicine at the Faculty of Medicine, Nantes, France, and his colleagues enrolled GPs in a total of 801 practices that included patients aged 50participating GPs caring for 33,044 patients who met study criteria.
Physician characteristics that were associated with higher FIT participation included younger age and an initially smaller number of unscreened patients. Patients with low socioeconomic status and those with a higher chronic disease burden were less likely to participate in FIT screening.
Also, Dr. Rat and his colleagues noted that the busiest practices actually had higher CRC screening rates. The investigators hypothesized that a recent physician pay-for-performance grant for CRC completion might be more appealing for some busy physicians.
This was the largest study of CRC screening participation to date, according to Dr. Rat and his coauthors, and showed the small but detectable efficacy of an inexpensive intervention that, given complete patient records, is relatively easy to effect. Though the effect size was smaller than the 12% difference the investigators had anticipated seeing for the patient-specific reminders group, the study still showed that targeting physicians can be an effective public health intervention to increase CRC screening rates, said Dr. Rat and his colleagues.
None of the investigators in either study reported conflicts of interest.
