The goal of this clinical trial is to learn whether an artificial intelligence (AI)-supported communication tool can help improve communication between people receiving palliative care and their healthcare providers. The study will also evaluate a new questionnaire designed to measure the quality of healthcare communication. The main questions this study aims to answer are: 1. Can the new questionnaire accurately measure the quality of communication between participants and healthcare providers? 2. Does using the AI-supported communication tool, called NurseChat+, help participants feel more confident and prepared before meeting with their healthcare team? 3. Does NurseChat+ improve the quality of communication during palliative care visits? Researchers will compare participants who use NurseChat+ together with their usual care to participants who receive usual care alone to see whether the tool improves communication outcomes. The study will be conducted in three stages. First, researchers will develop and test a questionnaire that measures healthcare communication. Next, they will develop and refine the NurseChat+ tool. Finally, they will test the tool in adults receiving palliative care at National Taiwan University Hospital. Participants in the final stage of the study will be randomly assigned to one of two groups. One group will use NurseChat+ before their scheduled medical visits in addition to receiving usual care, while the other group will receive usual care only. Participants will complete questionnaires about their communication experiences during the study.
Effective health communication is essential to patient well-being, yet the field is constrained by major theoretical and methodological gaps. Existing theories tend to address isolated communication phenomena and lack an integrative framework spanning the full continuum of patient-provider interaction. As a result, many measurement instruments lack a clear theoretical foundation and demonstrate inconsistent psychometric quality. In parallel, most AI-supported communication tools are technology-driven rather than theory-driven, and few have undergone rigorous clinical evaluation. Palliative care represents a high-stakes clinical context in which communication quality is central to outcomes, making it an ideal setting to test theory-based generative artificial intelligence (GenAI) communication interventions. Grounded in the Goal-Power Theory of Communication developed by the applicant, this four-year project adopts a mixed-methods design to develop and evaluate a theory-based communication measurement instrument and a GenAI-supported communication aid, NurseChat+. The project comprises three sequential phases: (1) instrument development and validation, (2) intervention development and pilot testing, and (3) clinical evaluation using a randomized controlled trial (RCT). Phase 1 aims to develop and psychometrically validate a theory-based instrument for assessing health communication quality. Instrument development and validation will follow the MEASURE framework, including expert content validation and field testing with approximately 500 participants recruited from clinical settings within the National Taiwan University Hospital system. Phase 2 aims to develop and preliminarily validate the NurseChat+ using a user-centered design approach. A pilot usability and feasibility study will be conducted using a one-group pretest-posttest design with a minimum of 20 healthy adults to assess acceptability, usability, and preliminary effects on communication self-efficacy. Phase 3 will conduct a parallel-group, superiority RCT with an embedded mixed-methods design in palliative care outpatient and home care settings at National Taiwan University Hospital. Eligible adult patients receiving palliative care will be randomly allocated in a 1:1 ratio to the intervention group (NurseChat+ plus usual care) or the control group (usual care alone). The target sample size is 240 participants (120 per group). Intervention participants will use NurseChat+ prior to scheduled medical encounters.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
OTHER
Masking
SINGLE
Enrollment
240
Participants will use NurseChat+, a generative AI-supported communication tool, before scheduled palliative care visits to help prepare questions and concerns for discussion with healthcare providers. The intervention is provided in addition to usual care.
National Taiwan University Hospital
Taipei, Taiwan
Change from Baseline in Communication Self-Efficacy via the 5-item Perceived Efficacy in Patient-Physician Interaction Scale (PEPPI-5)
【Randomized Control Trial \& Pilot study】The PEPPI-5 is a 5-item self-report instrument used to measure a patient's perceived self-efficacy in communicating with their physician. Each item is scored on a 5-point Likert scale from 1 (not at all confident) to 5 (very confident). The total score ranges from 5 to 25, with higher scores indicating higher levels of communication self-efficacy. Outcome will be assessed as the change in scores from T0 to T1.
Time frame: Baseline (T0: within 12 hours prior to the scheduled consultation) and Post-test (T1: immediately following the medical consultation or the interaction with GenAI communication aid).
Quality of Health Communication via the Phase-One Developed Instrument
Evaluated using a newly developed and validated communication-quality instrument generated in Phase One of the study. This instrument is designed to comprehensively assess the specific communication quality, patterns, and clinical encounter effectiveness between healthcare providers and palliative care patients/families when aided by GenAI communication aid. The instrument is currently undergoing development and psychometric validation. The finalized total score will be structured such that higher scores will mathematically indicate a superior quality of health communication and better communication patterns.
Time frame: Post-test follow-up (T1: 24-72 hours after the consultation via telephone interview).
System Usability Scale (SUS) Score of NurseChat+
【Pilot study】The UEQ contains 26 pairs of contrasting adjectives used to evaluate the user experience of NurseChat+ across 6 scales (Attractiveness, Perspicuity, Efficiency, Dependability, Stimulation, and Novelty). Each item is scored on a 7-point scale ranging from -3 (horrible/extremely bad) to +3 (excellent/extremely good). Scores above 0.8 represent a positive evaluation.
Time frame: Post-test (immediately after the interaction with GenAI communication aid).
Inventory of Technology Acceptance Model (ITAM) Score
【Pilot study】The ITAM assesses user acceptance, perceived usefulness, and perceived ease of use regarding the NurseChat+ GenAI system. Items are scored on a 7-point Likert scale, ranging from 1(strongly disagree) to 7 (strongly agre), with higher total scores indicating a higher intention to adopt and greater acceptance of the technology.
Time frame: Post-test ( immediately after the interaction with GenAI communication aid)
Evaluation of User Experience via the User Experience Questionnaire (UEQ)
【Pilot study】The UEQ is a 26-item semantic differential scale used to provide a holistic assessment of the user experience of NurseChat+. It measures both pragmatic qualities (efficiency, perspicuity, dependability) and hedonic qualities (stimulation, novelty), along with overall attractiveness. Each item consists of a pair of contrasting adjectives rated on a 7-point scale ranging from 1 to 7. For analysis, responses are transformed to range from -3 (horribly bad) to +3 (extremely good), where 0 represents a neutral evaluation. Higher positive values reflect a superior user experience, and scale scores above 0.8 generally indicate a positive evaluation.
Time frame: Post-test (immediately after the interaction with GenAI communication aid)
Improvement in Symptom Burden via the Edmonton Symptom Assessment System (ESAS)
【Randomized Control Trial】The ESAS is a validated tool designed to assess 9 common symptoms in palliative care patients (pain, tiredness, drowsiness, nausea, appetite, shortness of breath, depression, anxiety, and wellbeing). Each symptom is rated on a visual analogue scale from 0 (no symptom) to 10 (worst possible symptom). The total distress score ranges from 0 to 90, with higher scores indicating a higher symptom burden.
Time frame: Baseline (T0: within 12 hours prior to the scheduled consultation) and Post-test (T1: post-consultation follow-up within 24-72 hours).
Patient Overall and Communication Satisfaction via the Chinese Patients' Satisfaction Scale (C-PSS)
【Randomized Control Trial】The C-PSS is a 27-item instrument validated in Taiwan's outpatient population, assessing 6 dimensions: efficiency, fairness, chin-chieh (kindness), professionalism, responsibility, and respect. Each item is rated on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). With reverse-scored items appropriately adjusted, total scores are aggregated, where higher total scores indicate greater patient communication and overall satisfaction with the clinical encounter.
Time frame: Post-test follow-up (T1: 24-72 hours after the consultation via telephone interview).
Patient Satisfaction with Decision Making via the Satisfaction with Decision (SWD) Scale
【Randomized Control Trial】The Chinese version of the SWD is a 6-item questionnaire designed to assess a patient's satisfaction with the healthcare decision-making process and its clinical outcomes. Each item is rated on a 5-point Likert scale. Total scores range from 6 to 30, with higher scores indicating a higher degree of satisfaction and agreement with the decisions made during the medical consultation.
Time frame: Post-test follow-up (T1: 24-72 hours after the consultation via telephone interview).
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.