Adequate bowel preparation is critical for successful colonoscopy, yet inadequate preparation remains a significant clinical challenge, occurring in 20-30% of procedures. In hospitalized patients undergoing therapeutic colonoscopy, suboptimal preparation leads to increased costs, prolonged hospital stay, and potential procedure cancellation or rescheduling. Current standard care involves resident ward rounds twice daily.

This cluster-randomized crossover trial aims to evaluate whether increasing the frequency of structured resident ward rounds from 2 to 4 times per day can improve bowel preparation quality in hospitalized patients scheduled for therapeutic colonoscopy. The enhanced ward round intervention includes standardized checklist review, medication verification, dietary compliance confirmation, adverse event screening, and timely intervention when needed.

Three hospital wards will be randomly assigned to different sequences of intervention and control periods using a crossover design with washout periods. The primary outcome is adequate bowel preparation quality assessed by Boston Bowel Preparation Scale (BBPS ≥6 with each segment ≥2), evaluated by blinded endoscopists. Secondary outcomes include procedure quality metrics (cecal intubation rate, examination duration), safety endpoints (electrolyte disturbances, aspiration events), health economics measures (length of stay, total costs), and healthcare worker burden (nursing workload, night-time call frequency).

Subgroup analyses will examine intervention effects across age groups, cognitive function levels, prior colonoscopy experience, and comorbidity burden to identify populations most likely to benefit from enhanced monitoring.

This pragmatic trial addresses a clinically relevant question using a real-world implementation strategy designed to minimize workflow disruption. Results will inform evidence-based policies regarding optimal ward round frequency for colonoscopy preparation in hospital settings.

Background and Rationale Adequate bowel preparation is essential for high-quality colonoscopy. The Boston Bowel Preparation Scale (BBPS) is a validated tool for assessing preparation quality, with scores ≥6 (and each segment ≥2) considered adequate. Studies report inadequate preparation rates of 20-30% in routine practice, leading to missed lesions, prolonged procedures, and increased costs.

Hospitalized patients face unique challenges for bowel preparation including comorbidities, cognitive impairment, medication complexity, and communication barriers. While outpatient preparation often relies on written instructions and telephone reminders, inpatients theoretically benefit from direct medical supervision. However, the optimal frequency and structure of ward rounds during bowel preparation remain undefined.

Current practice at our institution includes twice-daily resident ward rounds (morning and afternoon). We hypothesize that increasing ward round frequency to four times daily (adding midday and evening rounds) with a standardized intervention package will improve preparation quality by:

1. Enhanced medication compliance monitoring
2. Real-time dietary adherence verification
3. Early detection and management of adverse events
4. Improved patient education and anxiety reduction
5. Timely adjustment of preparation regimens Study Design This is a single-center, cluster-randomized, crossover trial with three hospital wards serving as clusters. Each ward will alternate between intervention and control periods in a balanced sequence with washout periods to minimize carryover effects.

Design Features:

• Cluster Unit: Hospital ward (to avoid contamination within same care team)
• Crossover Design: Each ward serves as its own control across time periods
• Washout Periods: 1-week washout between intervention periods
• Blinded Outcome Assessment: Endoscopists remain blinded to patient allocation

Timeline Structure:

• Phase 0 (Weeks 0-1): Protocol training and system testing
• Phase 1 (Weeks 2-5): First intervention period (4 weeks)
• Washout 1 (Week 6): Transition period
• Phase 2 (Weeks 7-10): Second intervention period (4 weeks)
• Washout 2 (Week 11): Transition period
• Phase 3 (Weeks 12-15): Third intervention period (4 weeks)
• Weeks 16-17: Data cleaning and database lock Randomization: An independent biostatistician will generate a randomization sequence using Williams design for 3×3 crossover. Ward assignments will be concealed in sealed envelopes and revealed to ward supervisors only at the start of each period.

Interventions

Control Group (Standard Care):

• Ward rounds: 2 times/day (morning 7:30-9:00, afternoon 15:00-17:00)
• Standard clinical assessment during rounds
• Verbal confirmation of preparation instructions

Intervention Group (Enhanced Ward Rounds):

• Ward rounds: 4 times/day (morning 7:30-9:00, midday 12:00-13:00, afternoon 15:00-17:00, evening 21:00-22:00)

• Duration: ≥10 minutes per patient per round

• Standardized intervention package at each round:

  1. Medication Verification: Confirm PEG dose, timing, completion rate; assess tolerance (nausea, bloating)
  2. Dietary Compliance Check: Review food/fluid intake in past 4 hours; correct any deviations (e.g., family-provided meals, solid foods)
  3. Adverse Event Screening: Assess for electrolyte imbalance symptoms (weakness, palpitations, muscle cramps), aspiration risk (vomiting with cough), dehydration signs
  4. Clinical Decision Support: Prescribe antiemetics for severe nausea; adjust PEG rate for bloating; order electrolyte testing if indicated
  5. Patient Education Assessment: (First and pre-procedure evening rounds only) Three-question quiz on preparation instructions; re-educate if <2/3 correct Adherence Monitoring: Electronic time-stamped check-in via mobile app at each ward round, with mandatory completion of standardized checklist (cannot submit if <10 minutes elapsed).

Standardized Elements (Both Groups): To isolate the effect of ward round frequency, all other aspects remain identical:

• Preparation Regimen: Split-dose PEG (2L at D-1 18:00, 1L at D-day 6:00)
• Diet Protocol: Low-residue diet at D-2, clear liquids only at D-1
• Patient Education: Written handbook + 8-minute video within 24h of admission
• Rescue Protocol: Additional 500ml PEG or enema on examination morning if residual stool reported Outcome Measures

Primary Outcome:

• Adequate bowel preparation rate: Proportion of patients achieving BBPS total score ≥6 AND each segment score ≥2
• Assessment: During colonoscopy by blinded endoscopist
• Timeframe: At time of colonoscopy procedure

Key Secondary Outcomes:

1. BBPS total score (continuous, 0-9 scale)
2. Cecal intubation rate and time (minutes)
3. Total examination duration (minutes)
4. Adenoma detection rate (ADR)
5. Preparation-related adverse events (electrolyte abnormalities, aspiration, vomiting requiring medication)
6. Procedure rescheduling/repeat preparation rate
7. Length of hospital stay (days)
8. Total hospitalization costs (RMB)
9. Patient satisfaction score (5-point Likert scale)
10. Night-time nurse call frequency (21:00-7:00, normalized per patient)
11. Nursing workload assessment score

Subgroup Analyses (Pre-specified):

• Age: <60 / 60-74 / ≥75 years
• Cognitive function: Normal / mild impairment / moderate-severe impairment
• Prior colonoscopy: Yes / No
• Charlson Comorbidity Index: 0-2 / 3-4 / ≥5 Statistical Analysis Plan The primary analysis will use a generalized linear mixed model (GLMM) with logit link for the binary primary outcome, including fixed effects for treatment, period, and sequence, and random effects for ward. An intention-to-treat approach will be employed. Sample size of 268 evaluable patients (134 per group) provides 80% power to detect an absolute increase from 75% to 90% in adequate preparation rate, assuming ICC=0.03 and two-sided α=0.05.

Sensitivity analyses will include per-protocol population (excluding participants with <75% adherence to assigned ward round frequency), complete case analysis, and varying ICC assumptions. Subgroup effects will be assessed through treatment-by-subgroup interaction terms.

Safety Monitoring

The independent DSMB will conduct one interim analysis at 50% enrollment. Pre-specified stopping rules include:

• Significant increase in serious adverse events (SAEs) in intervention group (RR>2.0, p<0.01)
• Overwhelming efficacy (preparation rate difference >20%, p<0.001) All SAEs will be reported to the IRB within 24 hours. Quality Control
• BBPS rater training: All endoscopists complete standardized training with 20 reference videos; inter-rater reliability (Kappa) >0.75 required
• Calibration meetings every 4 weeks to review borderline cases
• Electronic data capture (REDCap) with real-time validation rules
• Weekly data monitoring by research coordinator with 10% source verification

Inclusion Criteria:

1. Age ≥18 years
2. Hospitalized patients (inpatients)
3. Scheduled for therapeutic colonoscopy (polypectomy, endoscopic mucosal resection [EMR], endoscopic submucosal dissection [ESD], or biopsy for diagnosis)
4. Time from admission to scheduled colonoscopy ≥48 hours (allowing adequate preparation time)
5. Patient or legal representative able to provide written informed consent
6. Able to understand and follow bowel preparation instructions (with assistance from caregivers if needed)

Exclusion Criteria:

1. Emergency gastrointestinal conditions: bowel obstruction, acute lower gastrointestinal bleeding requiring urgent intervention, suspected or confirmed perforation
2. Emergency or same-day colonoscopy (scheduled <24 hours after admission)
3. History of total colectomy or current colostomy/ileostomy
4. Severe cognitive impairment (Mini-Cog score 0-1 or MMSE <18) without available caregiver
5. Severe cardiopulmonary disease: New York Heart Association (NYHA) Class IV heart failure, chronic kidney disease (CKD) stage 4-5 (eGFR <30 ml/min/1.73m²) unable to tolerate PEG-based preparation
6. Known or suspected pregnancy, currently breastfeeding
7. Previous severe allergic reaction to PEG or bowel preparation agents
8. Current participation in another interventional clinical trial
9. Unable to provide informed consent and no legal representative available
10. Patient or legal representative declines participation

Individual participant data will not be shared publicly due to institutional privacy policies and Chinese regulations on patient data protection. Aggregate results will be published in peer-reviewed journals and presented at scientific conferences. Researchers interested in collaboration or data access may contact the principal investigator to discuss potential data sharing arrangements under appropriate data use agreements.

Endoscopists performing colonoscopy and assessing BBPS scores are blinded to patient group assignment. Due to the nature of the ward round intervention, participants and care providers cannot be blinded. Data analysts will remain blinded to group allocation until after primary analysis completion.