An Interactive Virtual Nephrologist to Engage Patients With Advanced CKD About Transplantation.

RATIONALE & OBJECTIVE: Despite advantages in survival and quality of life with kidney transplantation (KT) compared to other treatments for kidney failure requiring dialysis, many patients do not receive sufficient information about KT or have concerns about it. New artificial intelligence capabilities enable the creation of virtual agents to supplement human efforts to engage patients about treatment options. STUDY DESIGN: Single arm, open label trial with the primary outcome assessed pre- and post-intervention. SETTING & PARTICIPANTS: Adults (≥18 and <75 years) with eGFR <25 mL/min/1.73 m2 in a large health system. INTERVENTIONS: Our group developed the virtual transplant nephrologist - an interactive, video-based platform to provide personalized education about KT. Meetings could occur anytime from a smartphone with a weblink. Participants first selected the demographic profile of their virtual nephrologist. Participants then self-tailored the encounter, choosing among 11 topics, pausing, changing the topic, and requesting more information. OUTCOME: The primary outcome was the difference in participants' intentions to speak to their human doctor about KT, before versus after the intervention, assessed on a 5-point Likert-like scale. RESULTS: Participants had a median age of 66 years; 59% were female and 50% were Black. Female and Black participants demonstrated preferences to select a gender (p<0.001) or race (p<0.001) concordant virtual nephrologist, respectively. Participants spent a median 12.6 minutes reviewing a median of 11 topics in the encounter. The primary outcome of intention to speak to a human clinician about KT increased significantly after the virtual encounter (mean increase 0.30; p<0.001). LIMITATIONS: The clinical significance of the outcome is uncertain. CONCLUSIONS: Virtual clinicians offer a promising path to education and engagement about complex treatments such as KT. This customizable approach to providing high quality information about KT using a virtual clinician warrants more definitive testing in a randomized trial.