Artificial Intelligence (AI), Bowel Preparation Quality, Colon Polyps and Adenomas, Colorectal Cancer Screening
Conditions
Keywords
colorectal cancer, colon polyp, colonoscopy, bowel preparation, Boston Bowel Preparation Score (BBPS), artificial intelligence, AIChatbot, large language model, patient satisfaction, cancer screening
Brief summary
Traditional pre-colonoscopy counselling requires significant time from healthcare workers to explain procedures, limiting efficiency and patient turnover. Inadequate bowel preparation exacerbates this issue, leading to repeat procedures and increased costs. However, no study has yet evaluated the effectiveness of AI in improving the Boston Bowel Preparation Scale (BBPS) for colonoscopy preparation. By addressing this gap, AI chatbots could provide personalized guidance, reduce healthcare worker burden, improve preparation quality, and enhance patient experience.This research attempts to evaluate the effectiveness of using Artificial intelligence (AI) chat bot to improve bowel preparation, anxiety level and patient's satisfaction among colonoscopy patients in Hospital Tuanku Muhriz (HCTM), compared to conventional instructions
Interventions
PROCEDUREAI chatbot study arm
Intervention arm will have to undergo bowel preparation using standard polyethylene glycol solution (Fortrans)
PROCEDURETraditional counselling study arm
Intervention arm will have to undergo bowel preparation using standard polyethylene glycol solution (Fortrans)
Sponsors
National University of Malaysia
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
DIAGNOSTIC
Masking
SINGLE (Caregiver)
Masking description
Endoscopist performing the colonoscopy will be masked
Intervention model description
Prospective, single-masked, randomized control trial
Eligibility
Sex/Gender
ALL
Age
18 Years to 75 Years
Healthy volunteers
Yes
Inclusion criteria
* All scheduled colonoscopy with indication * Adequate digital literacy * Adequate language literacy with Malay and English language
Exclusion criteria
* Patients with memory impairment due to previous stroke, dementia or Alzheimer's disease * Diagnosed with clinical anxiety
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| To determine the effectiveness of artificial intelligence (AI) chat bot in improving bowel preparation among colonoscopy patients in Hospital Tuanku Mukhriz (HCTM), compared to conventional instructions | 1 year | Patients in interventional arm will be receiving counselling via interaction with AI chatbot.Each interaction session were limited to 15 minutes maximum. The quality of bowel preparation will be graded by the masked endoscopists using Boston Bowel Preparation Score (BBPS). Each of the three colon segments (right, transverse and left) will be assigned scores (0 - 3) . 0 - unprepared colon while 3 indicates the best possible preparation.A total score will range from 0 to 9, where higher scores indicate better bowel preparation. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| To determine the effectiveness of AI chat bot in relieving anxiety among colonoscopy patients in HCTM, compared to conventional instructions. | 1 year | Anxiety level for both arms will be measured before and after colonoscopy using the validated Depression Anxiety and Stress Scale 21 (Anxiety Scale). Is a self-reporting scale that measure the the anxiety.DASS interpretation ;Normal 0-7, Mild 8-9, Moderate 10-14, Severe 15-19 and Extremely severe 20+ |
| To determine the effectiveness of AI chat bot in improving satisfaction among colonoscopy patients in HCTM, compared to conventional instructions. | 1 year | Patients from both arms will be interviewed after colonoscopy using the validated Patient Satisfaction Questionnaire (PSQ).PSQ are a set of questionnaires that help researchers to understand patient experiences and identify areas of improvement.Patients will be evaluated on multiple domains- general satisfaction, technical quality, interpersonal aspect, clarity of communication,access to care and confidence in healthcare provider. Each domain will be given scores according to the Likert scale ranging from 1 to 5. 1 means very dissatisfied while 5 means very satisfied. |
Countries
Malaysia
Contacts
Primary ContactNabil Mohammad Azmi, Doctor in General Surgery
Outcome results
None listed