AI-empowered Nudge to Improve Colonoscopy Uptake (AINC)

We will conduct a two‐arm, parallel‐group, cluster‐randomized controlled trial toevaluate the effectiveness of an AI‐empowered nudge model in improving colonoscopy uptake(AINC) among high‐risk individuals aged 45 to 74. The AI‐empowered nudge model combinesdefault screening nudging with an AI chatbot on colorectal cancer screening. We will alsoevaluate the feasibility of this AINC model, and identify the facilitators and barriersto its real‐world adoption.The colonoscopy uptake rate is approximately 15% in China, and the proposed interventionis expected to increase this rate by 10%. Sample size calculation, based on detecting anincrease in colonoscopy uptake from 15% to 25% with 90% power (α=0.05, two‐sided), an ICCof 0.05, and 30 clusters per arm, indicates a need for 24 participants per cluster. Thereare 720 per arm, and 1440 in total. Allowing for 15% attrition, the final sample size isdetermined to be 1680 from 70 clusters. As a pragmatic trial in real world, the number ofparticipants each cluster depends on the population size of the respective villages orcommunities. All eligible participants in the participating villages or communities willbe included in the study.Participant recruitment will be conducted across 70 villages/communities in threerepresentative counties/cities in China, covering urban, suburban, and rural areas.Cluster randomisation will be performed at the level of villages or communities using astratified block design to ensure balanced allocation across the two trial arms.Stratification factors include geographic access to colonoscopy hospital and the size ofindividuals aged 45 to 74 for each cluster. Clusters with comparable levels of thesefactors will be grouped into blocks within each city and then randomly assigned withineach block to the AINC or control group. The random allocation sequence will be generatedby an independent statistician using a computer‐based random number generator in Rsoftware and implemented via a secure centralised system.The study procedure involves first identifying high‐risk individuals for CRC through aninitial risk assessment questionnaire and a fecal immunochemical test (FIT). Those whomeet the criteria will then receive the intervention corresponding to their village'sassigned study arm. Participants in the intervention group will receive an AI‐powerednudge for colonoscopy (AINC), featuring an AI chatbot providing real‐time personalizedresponses and a nudge environment with default screening option, followed by messagereminders once per two weeks. The control group will receive usual care. Colonoscopyuptake will be collected via the hospital information system at the 3‐month follow‐up.The primary analysis will follow the intention‐to‐treat (ITT) principle, while theper‐protocol (PP) analysis will serve as the secondary analysis. In the ITT analysis, allsubjects randomized to each group will be included. Between‐group comparisons forcontinuous and categorical variables will utilize t‐tests and chi‐square tests. Theprimary outcome (colonoscopy uptake) will be analyzed using Generalized EstimatingEquations (GEE), adjusting for cluster effects and relevant covariates to obtain robustestimates. Covariates include region, age, sex, smoking history, Body Mass Index, historyof bowel‐related symptoms or diseases, and family history. The timing of colonoscopyuptake will be analyzed using Kaplan‐Meier survival curves and log‐rank tests, and theintervention effects on the time‐to‐event will be quantified with a Cox proportionalhazards model. Subgroup analyses will be conducted to elucidate the effect heterogeneityacross populations stratified by pre‐specified characteristics, including region, age,sex, smoking history, Body Mass Index, history of bowel‐related symptoms or diseases, andfamily history.