Effect of Patient-Directed Messaging on Colorectal Cancer Screening: A Randomized Clinical Trial

Abstract

Importance: Colorectal cancer (CRC) screening is underused in the US. Tailored message interventions have shown benefit for increasing screening uptake of mammography and Papanicolaou testing, but their role in CRC screening is less clear. Objective: To evaluate the effectiveness of a tailored message telephone intervention prior to scheduling of a screening or surveillance colonoscopy and its effect on CRC screening completion rates. Design, Setting, and Participants: This randomized clinical trial was conducted from July 2017 through August 2018 at the University of Pennsylvania Health System (UPHS), an urban academic medical center. Participants were asymptomatic patients aged 50 to 75 years who were eligible for CRC screening or surveillance, had been referred for colonoscopy, and did not have a scheduled colonoscopy appointment. Data analysis was conducted from January to September 2019. Interventions: Patients underwent block randomization in a 1:1:1 ratio to 1 of 3 study arms. Participants in the usual care group were contacted via a mailed letter and instructed to call to schedule a colonoscopy. In the generic message group, participants were contacted by telephone, completed an assessment, and received a uniform, nontailored message encouraging colonoscopy scheduling. Participants in the tailored message group were contacted by telephone, completed an assessment, and received a tailored message encouraging colonoscopy scheduling based on their identified assessment cohort. Main Outcomes and Measures: The primary outcome was colonoscopy completion rate within 120 days of enrollment. The secondary outcome was colonoscopy scheduling rate appointment within 120 days of enrollment. Results: A total of 600 participants (median [IQR] age, 56 [51-63] years; 373 women [62.2%]) were enrolled, including 200 participants randomized to usual care, 200 participants randomized to the generic message, and 200 participants randomized to the tailored message. The total sample included 12 Asian participants (2.0%), 324 Black participants (54.0%), and 227 White participants (37.8%), and 9 participants (1.5%) were of Latino or Hispanic ethnicity. Colonoscopy completion was significantly higher for both the tailored message group (69 participants [34.5%]) and the generic message group (64 participants [32.0%]) compared with the usual care group (37 participants [18.5%]) (P