Cardiovascular Kidney Metabolic Syndrome
Conditions
Keywords
artificial intelligence, Chatbot, electronic patient-reported outcomes, patient activation, randomized controlled trial, cardiovascular kidney metabolic syndrome
Brief summary
This randomized clinical trial aims to evaluate whether an AI chatbot-enabled, nurse-led electronic patient-reported outcome (ePRO) intervention can improve symptom management and self-care in adults with early-stage cardiovascular-kidney-metabolic (CKM) syndrome. The study will also assess the safety and feasibility of the intervention. Researchers will investigate whether the program reduces symptom burden, improves patient activation and health literacy, and lowers healthcare utilization such as emergency department visits and hospitalizations compared with usual ePRO monitoring alone. A total of 72 participants will be randomly assigned to either an intervention group, which receives weekly AI chatbot-supported symptom monitoring combined with nurse-led feedback and self-management support, or a control group receiving ePRO monitoring without feedback. Participants will complete PROMIS-29 assessments weekly and attend study evaluations at baseline, 3 months, and 6 months, while clinical outcomes and healthcare use are tracked through electronic medical records.
Detailed description
The trial is a two-arm, parallel-group randomized controlled study designed to evaluate an AI chatbot-enabled, nurse-led ePRO-based care model for individuals with early-stage cardiovascular-kidney-metabolic (CKM) syndrome. A total of 72 adults aged 40 years and older will be recruited from three hospitals in Northern Taiwan and followed for six months. Participants will be randomly assigned in a 1:1 ratio to either an intervention group or a control group using a computer-generated block randomization sequence with concealed allocation. Due to the nature of the intervention, participant blinding is not possible; however, outcome assessors and data analysts will remain blinded. The intervention group will receive a structured nurse-led program combined with an AI chatbot that delivers weekly PROMIS-29 symptom assessments, automated feedback, motivational self-management coaching, and risk-based alerts for clinical deterioration. Nurses will review alerts daily and provide tiered responses, including self-management guidance, teleconsultation, or referral to specialist care when necessary. The control group will receive standard care and weekly ePRO assessments without feedback or clinical integration. Primary outcomes include symptom burden measured by PROMIS-29 v2.1, while secondary outcomes include patient activation, health literacy, and healthcare utilization (emergency visits and hospitalizations). Data will be collected at baseline, 3 months, and 6 months, and analyzed using linear mixed-effects models under an intention-to-treat framework. Safety monitoring will be continuous, with predefined clinical thresholds triggering automated alerts and nurse-led follow-up actions to ensure timely intervention and participant safety.
Interventions
OTHERAI chatbot-enabled nurse-led program.
Participants in the intervention group will receive a six-month nurse-led, AI chatbot-enabled symptom monitoring and self-management program integrating weekly ePRO assessments. The intervention begins with a structured nurse-led intake session including CKM assessment, symptom review, goal setting, and training on chatbot use. Participants complete weekly PROMIS-29 v2.1 assessments via a smartphone-based chatbot, which provides automated, rule-based feedback, self-management guidance, and motivational prompts. A risk-stratification algorithm generates alerts when symptom deterioration is detected. Nurses review alerts daily and provide follow-up via messaging, teleconsultation, or referral depending on severity, with same-day escalation for high-risk symptoms.
OTHERUsual Care
Usual care based on standard outpatient follow-up, without access to the AI chatbot or nurse-led coordination
Sponsors
Taipei Medical University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
SUPPORTIVE_CARE
Masking
SINGLE (Outcomes Assessor)
Masking description
Outcome assessors and data analysts are all blinded to group assignments.
Intervention model description
Participants in the intervention arm will receive a 6-month AI chatbot-supported, nurse-led symptom monitoring and self-management program using ePROs. The intervention begins with a 45-60-minute nurse-led intake session involving clinical assessment, individualized goal setting, and orientation to the chatbot system, followed by device setup and scheduling. The AI chatbot delivers weekly PROMIS-29 (Traditional Chinese version) ePRO assessments via smartphone or tablet, with automated individualized feedback, reinforcement of self-management goals, and brief motivational interviewing-based prompts. Additional brief check-ins may be triggered in response to symptom deterioration or non-adherence. The system functions as a decision-support tool using predefined risk thresholds to enable symptom stratification and nurse-led escalation when required.
Eligibility
Sex/Gender
ALL
Age
40 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Adults ≥40 years * Classified as having stage CKM syndrome (stage 1-2) * Receiving outpatient care * Able to communicate in Mandarin/Taiwanese, and * Having access to a smartphone/tablet
Exclusion criteria
* End-stage kidney disease requiring dialysis or are listed for kidney transplantation * Have terminal illness (\<6 months life expectancy) * Have significant cognitive impairment * Currently enrolled in another interventional study targeting cardiovascular-kidney-metabolic syndrome or multimorbidity
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| PROMIS-29 v2.1 | Baseline, 3 months, and 6 months after randomization | Symptom burden will be evaluated using the PROMIS-29 v2.1 Traditional Chinese version (31). This instrument comprises 29 items across seven domains: Physical Function, Depression, Anxiety, Fatigue, Sleep Disturbance, Social Participation, and Pain (Intensity and Interference). Most items are rated on a 5-point Likert scale (1-5), with a single pain intensity item scored from 0 to 10. Domain scores are summed and converted to standardized T-scores ranging from 0 to 100, with higher scores indicating greater symptom severity |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| PAM-13 | Baseline, 3 months, and 6 months. | Patient activation (PAM-13), which evaluates individuals' knowledge, skills, and confidence in managing their health. Items are rated on a 5-point Likert scale (from strongly disagree to strongly agree, with an additional "not applicable" option). Raw scores are converted using standardized calibration tables to a 0-100 scale, with higher scores indicating greater activation |
| MMHLQ | Baseline, 3 months, and 6 months. | Health literacy will be assessed using the 20-item Mandarin Multidimensional Health Literacy Questionnaire. Health literacy will be assessed using the 20-item Mandarin Multidimensional Health Literacy Questionnaire (MMHLQ), which covers five domains: obtaining, understanding, evaluating, and applying health information, as well as communication and interaction. Each item is rated on a 4-point Likert scale ranging from very difficult to very easy. Scores are transformed to a 0-50 scale using the formula (Mean - 1) × (50/3), with higher scores indicating better health literacy. Levels are categorized as insufficient (0-25), limited (25-33), sufficient (33-42), and excellent (42-50). |
| Number of emergency department visit | Baseline, 3 months, and 6 months. | the total count of times a patient visits the emergency department within a specified period during intervention program |
| Hospitalizations | Baseline, 3 months, and 6 months | The number of times a patient is admitted to a hospital within a specified period during intervention program |
Countries
Taiwan
Contacts
CONTACTPei Shan Tsai, PhD in Nursing
Outcome results
None listed