A randomized, open-label behavioral study designed to evaluate the feasibility, acceptability, and early impact of a large language model (LLM)-based artificial intelligence (AI) tool to support patient understanding following an initial oncology consultation.
Age ≥18 years.
Histologically or cytologically confirmed diagnosis of localized prostate cancer
Newly diagnosed and planning to undergo or recently completed a radiation or surgical oncology consult at University of California, San Francisco (UCSF)
Eastern Cooperative Oncology Group (ECOG) performance status of 0-2
Prior definitive treatment for prostate cancer (e.g., surgery, radiation, or systemic therapy)
Cognitive, sensory, or language impairment that would interfere with the ability to understand, participate in, or complete the study procedures
Inability or unwillingness to participate in two study-related teleconferences
Current enrollment in another interventional clinical trial that may interfere with study participation
Parallel
AI-guided COUNSEL Intervention
BehavioralAn AI-guided, interactive educational tool that allows participants to interact with a large language model (LLM) will be delivered via teleconference.
Questionnaires
BehavioralParticipant and clinicians will complete questionnaires
Mean differences on EORTC QLQ-INFO25 - Patient-Perceived Cancer-Related Information
Patient-perceived quality, breadth, and depth of cancer-related information received following an oncology consultation, as measured by the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-INFO25) survey. The EORTC QLQ-INFO25 comprises four multi-item scales and eight single items to assess the level and quality of information received by cancer patients. Scores for all scales and single items range from 0-100. The survey uses a 4-point Likert scale (1 = "not at all' to 4 = "very much") and some yes/no questions. Higher scores represent a greater patient-perceived level of shared decision-making. Mean differences between treatment groups will be reported with 95% confidence intervals, along with Hedges' g effect sizes. A Mann-Whitney U test will be used for sensitivity analysis.
Mean differences on SDM-Q-9 and SDM-Q-Doc - Patient- and Provider-Perceived Shared Decision Making
Perceived shared decision making will be assessed from both patient and clinician perspectives using the 9-item Shared Decision Making Questionnaire (SDM-Q-9) and the Shared Decision Making Questionnaire-Physician Version (SDM-Q-Doc), respectively. Each item is rated on a 6-point Likert scale from 0 (completely disagree) to 5 (completely agree), with total scores ranging from 0 to 45. Higher scores indicate greater perceived levels of shared decision making from the patient or clinician perspective. Mean differences between groups will be reported using Welch's t-test, along with the 95% confidence intervals.
Mean differences on NCCN Distress Thermometer - Patient-Reported Distress
Patient-reported distress will be assessed using the National Comprehensive Cancer Network (NCCN) Distress Thermometer at Day 2 and Day 21. The Distress Thermometer is a single-item measure rated on a 0-10 scale, where 0 = "No distress" and 10 = "Extreme distress. Mean differences will be reported with 95% confidence intervals, along with Hedges' g effect sizes and 95% confidence intervals. Descriptive statistics will also be reported.
Patient-Provider Concordance in Shared Decision Making
Concordance between patient and provider perceptions of shared decision making will be assessed by comparing SDM-Q-9 and SDM-Q-Doc scores within each study arm. Concordance will be evaluated using an intraclass correlation coefficient (ICC). Bland-Altman plots will be used to visualize treatment group differences in concordance. ICC estimates and Bland-Altman plots will be reported.
Click on the lead sponsor to view all of their trials.
This trial is not recruiting at the moment. You can still explore other options:
