Effect of Text Messaging on Bowel Preparation and Appointment Attendance for Outpatient Colonoscopy: A Randomized Clinical Trial

IMPORTANCE: Outpatient colonoscopy is important for colorectal cancer screening. However, nonadherence and poor bowel preparation are common. OBJECTIVE: To determine if an automated text messaging intervention with a focus on informational and reminder functions could improve attendance rates and bowel preparation quality for outpatient colonoscopy. DESIGN, SETTING, AND PARTICIPANTS: This randomized clinical trial was conducted in an endoscopy center at an urban academic medical center. Adult patients scheduled for outpatient colonoscopy between January and September 2019 were enrolled by telephone call (early phase) or by automated text message (late phase). Data were analyzed from October 2019 to January 2020. INTERVENTIONS: After enrollment, patients were randomized in a 1:1 ratio to usual care (ie, written instructions and nurse telephone call) or to the intervention (ie, usual care plus an automated series of 9 educational or reminder text messages in the week prior to scheduled colonoscopy). MAIN OUTCOMES AND MEASURES: The primary outcome was appointment attendance rate with good or excellent bowel preparation. Secondary outcomes included appointment attendance rate, bowel preparation quality (poor or inadequate, fair or adequate, and good or excellent), and cancellation lead time (in days). RESULTS: Among 753 patients included and randomized in the trial (median [interquartile range] age, 56 [49-64] years; 364 [48.3%] men; 429 [57.2%] Black), 367 patients were randomized to the intervention group and 386 patients were randomized to the control group. There was no significant difference in the primary outcome between groups (patients attending appointments with good or excellent bowel preparation: intervention, 195 patients [53.1%]; control, 210 patients [54.4%]; P = .73), including when stratified by early or late phase enrollment groups. Similarly, there were no significant differences in secondary outcomes. CONCLUSIONS AND RELEVANCE: This randomized clinical trial found no significant difference in appointment attendance or bowel preparation quality with an automated text messaging intervention compared with the usual care control. Future work could optimize the content and delivery of text message interventions or identify patient subgroups that may benefit from this approach. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03710213