After Positive Fecal Test, Here's What Gets Patients to Colonoscopy

— Even with interventions, follow-up colonoscopy completion rates still disappoint in meta-analysis

A meta-analysis favored patient navigation and patient or clinician reminders for increasing colonoscopy completion after stool-based tests for colorectal cancer (CRC) screening turned up positive.

After a positive fecal immunochemical test (FIT) or fecal occult blood test (FOBT), colonoscopy completion was 28% higher with patient-level interventions (RR 1.28, 95% CI 1.17-1.41) and 45% higher with system-level interventions (RR 1.45, 95% CI 1.19-1.76), reported Michael Pignone, MD, MPH, of Duke University School of Medicine in Durham, North Carolina, and colleagues.

"Pooled data suggest that 69.1% of patients in intervention groups completed follow-up colonoscopy after an abnormal FIT/FOBT result, compared with 47.8% in control groups," the researchers wrote in JAMA Internal Medicineopens in a new tab or window. "Despite these interventions, completion rates remained below the recommended 80% targetopens in a new tab or window set by the U.S. Multi-Society Task Force on CRC."

Individual interventions were associated with increased colonoscopy completion on the order of 24% to 68%, but analyses showed moderate to high heterogeneity in studies that included automated referral to gastrointestinal (GI) consult:

• Patient navigation: risk ratio [RR] 1.24 (95% CI 1.15-1.33)
• Patient reminders: RR 1.32 (95% CI 1.04-1.67)
• Clinician reminders: RR 1.36 (95% CI 1.12-1.65)
• Auto referral to GI consult: RR 1.68 (95% CI 1.09-2.60)

"While individual interventions demonstrate benefit, further research is needed to evaluate the incremental effectiveness of multicomponent strategies and the cost-effectiveness of achieving the 80% completion rate," according to Pignone and co-authors.

The U.S. Preventive Services Task Forceopens in a new tab or window and the American Cancer Societyopens in a new tab or window recommend CRC screening for average-risk adults starting at age 45 via colonoscopy or other proven methods such as FIT or FOBT.

Roughly 9 in 10 patients undergoing a fecal test for CRC screening have a negative result, but the strategy involves a two-step approach for those with a positive test. This group has the highest risk for advanced adenomas or early-stage CRCs, but various studies -- including this one -- have shown only about half of patients complete a subsequent colonoscopy.

In an editor's noteopens in a new tab or window accompanying the study, JAMA Internal Medicine editors Cary Gross, MD, and Ilana Richman, MD, MHS, said the meta-analysis results "can be seen as a direct challenge to the current approach to stool-based screening, which clearly results in inadequate screening efficacy."

Prior researchopens in a new tab or window has shown that healthcare system problems (including a lack of coordinated care or staffing issues) and patient cognitive factors, health literacy, or fear of a cancer diagnosis can stand in the way of colonoscopy completion for those with a positive fecal test.

"To paraphrase former Surgeon General C. Everett Koop, MD, screening tests do not work in patients who do not complete them," said Gross and Richman, both of Yale University School of Medicine in New Haven, Connecticut.

Compared with the sub-50% rate among the control groups in the meta-analysis, patient navigationopens in a new tab or window yielded the highest colonoscopy follow-up rates (72.5%), followed by patient reminders (67.7%), clinician reminders (65.7%), and automated referrals to GI consult (60.2%). Overall, the patient-level interventions appeared to perform slightly better than the system-level interventions (71.1% vs 65.4%).

"Various strategies to promote adherence to follow-up colonoscopy are effective," the editorialists concluded. "Given that clinics have different capacities, priorities, and constraints, clinicians and clinical leaders should choose an approach that is sensible for their own clinical setting, while also recognizing that it is important and effective to do something, rather than nothing."

The systematic review and meta-analysis from Pignone and colleagues included 21 interventional studies -- 10 nonrandomized and 11 randomized -- involving 19,613 patients (41.3% women) who ranged in age from 50 to 75 years.

Fifteen of the studies were U.S.-based while the remaining six were conducted in the U.K., Italy, Malaysia, South Korea, China, and Israel. Ten studied patient navigation, four patient reminders, three studied clinician reminders, three automated referrals to GI consult, and one study tested multicomponent interventions involving clinician reminders and automated referrals.

Limitations of the meta-analysis noted by the researchers included the significant heterogeneity of the various study designs and the inclusion of nonrandomized interventional studies, which can be more susceptible to bias. Though when broken down by study type, the researchers found no difference in effect of the interventions between randomized trials (RR 1.37, 95% CI 1.20-1.57) and nonrandomized interventional studies (RR 1.35, 95% CI 1.16-1.57). In addition, there were no significant differences between randomized and nonrandomized studies across system-level and patient-level interventions.

Gross and Richman pointed to other limitations too, including that few of the studies directly compared the various interventions and that newer stool-based tests such as FIT-DNA were not included.