COLONOSCOPY PREPARATION GUIDED BY A PERSONALIZED DIGITAL PLATFORM IS ASSOCIATED WITH LOWER RATES OF INADEQUATE PREPARATION AND REDUCED NEED FOR REPEATED COLONOSCOPIES COMPARED TO DIGITAL WRITTEN INSTRUCTIONS

Purpose: Inadequate bowel preparation leads to missed colonic lesions and incomplete colonoscopy. We evaluated whether implementing a new personalized digital platform could improve bowel preparation and related outcomes.
Methods: The personalized digital platform (DP) is a patient-tailored animated video of bowel preparation instructions, individually produced according to age, sex, comorbidities and language. Before scheduled ambulatory colonoscopy, a link to the digital platform system was sent to random patients (DP-link group), while the control group received standard written instructions. We then performed a retrospective analysis of prospectively collected data regarding colonoscopy outcomes.
Results: Out of 2447 colonoscopies, the DP-link was randomly sent to 913 (37.3%) patients and standard written prep instructions to 1534 (62.7%) (control group). Lower rates of inadequate bowel preparation (Boston bowel preparation scale (BBPS) <6) were observed in the DP-link group compared to the control group (21.2% vs. 27.6%, P<0.001). Among DP link receivers, 470/913 (51.5%) watched >75% of the video (“full-users”) while 443/913 (48.5%) did not (“partial-users”). BBPS<6 rate was lowest among full-users, higher in partial-users and highest in the control group (19.1%, 23.5% and 27.6%, respectively, P<0.001). Cecal intubation rates were higher in the DP-link group (96.5% vs. 94.5%, P=0.026) and propofol dose was lower (233.7±97.5 vs. 244.3±105.9 mg, P=0.020). Logistic regression analysis adjusted for age, sex, prep protocol and morning/afternoon colonoscopy, showed that receiving the DP-link and DP full-usage were each independently associated with reduced rates of BBPS<6 [(OR 0.695, P<0.001) and (OR 0.627, P<0.001), respectively]. Significantly reduced BBPS< 6 rates were maintained upon similar logistic regression in selected population subgroups [male sex (OR 0.727, P=0.013), female sex (OR 0.630, P=0.004), 50–70-year-old patients (OR 0.602, P<0.001), and both morning (OR 0.726 P=0.015), and afternoon colonoscopy (OR 0.639, p=0.004)].
Conclusions: A personalized digital platform for pre-colonoscopy preparation guidance leads to lower rates of inadequate preparation and higher procedure completion rates.