Practice That Builds Trust: How AI + VR Strengthen the Soft Skills That Keep Patients Safe

How a de-identified North American health system used AI-enabled, emotionally responsive VR and screen-based simulations to build repeatable soft-skill reps for nurses and clinical teams, without increasing educator workload.

Case Study Overview

Challenge & Context

Clinical outcomes are shaped by conversations, not just clinical tasks. Yet high-impact soft skills like empathy, de-escalation, structured handoff, and psychologically safe “speak-up” behaviors are hard to practice consistently because traditional role-play and live simulation are labor-intensive and variable.

Workplace violence is also a growing operational and safety concern, and training that strengthens early verbal de-escalation skills is widely viewed as essential for staff well-being and safe care delivery (Clay et al. 2025) (Moore et al. 2022).

Bottom line: leaders need scalable ways to practice the human side of care with the same rigor as technical skills.

Approach & Solution Framework

A de-identified health system partnered with Patient Ready to deploy an AI + VR “soft skills readiness layer” designed to:

• Increase repeatable, realistic communication practice using immersive VR simulations and AI-powered virtual patients that respond to learner choices, tone, and intent (Cho and Kim 2024) (Stamer et al. 2023).

• Build empathy and perspective-taking through structured, scenario-based experiences grounded in evidence that VR can support empathy education (Lin et al. 2024).

• Improve readiness for high-risk conversations (agitation, refusal, distress, conflict) using VR-based conversational agents and de-escalation practice loops (Moore et al. 2022).

• Standardize feedback and coaching using AI-supported artifacts (for example, transcripts, conversation summaries, and moment tagging) to reduce repetitive baseline correction demands and improve debrief consistency (Stamer et al. 2023) (Elhilali et al. 2025).

Measurable Results & Impact (Directional Outcomes)

Within the first implementation cycles, the system reported outcomes consistent with published evidence on VR and AI-supported communication training:

• Increased standardization of soft-skill exposure across cohorts by shifting selected objectives into repeatable VR modules (directional; program-reported), aligning with evidence that VR simulation improves communication skills overall in nursing education (Cho and Kim 2024).

• Greater feasibility for frequent practice without proportional increases in staffing or room scheduling (directional; program-reported), consistent with VR training’s flexibility and resource advantages for interprofessional learning (Neher et al. 2024).

• Improved learner comfort engaging in difficult conversations through repeated AI-enabled dialogue practice (directional; program-reported), consistent with findings that AI can support readily available communication training and feedback (Stamer et al. 2023) (Shorey et al. 2019).

Key Insights

• Soft skills improve when practice becomes repeatable, measurable, and emotionally realistic, not when learners only “talk about” communication.

• VR helps scale high-quality reps. AI helps make those reps responsive and feedback-ready.

• The operational win is shifting scarce educator time from repeated role-play setup to higher-order coaching and debriefing.

Learn how Patient Ready can scale communication and de-escalation practice with emotionally responsive AI patients. Visit https://getpatientready.com to learn more.

1. Background & Context: Why “Soft Skills Reps” Have Become a Capacity Problem

Most organizations agree that communication drives patient safety, experience, and team performance, but the training mechanisms are often inconsistent: role-play quality varies by facilitator, standardized patient programs are expensive, and live simulation time is limited.

At the same time, workplace violence prevention training frequently includes de-escalation, and experiential practice is viewed as important. VR is increasingly positioned as a scalable way to replicate high-risk contexts safely and repeatedly (Clay et al. 2025). Early research also indicates that verbal interaction with virtual agents is feasible and can support de-escalation skill development (Moore et al. 2022).

The strategic challenge is delivering enough high-quality practice cycles per learner, with consistent feedback, without multiplying educator workload.

2. The Challenge: High Stakes Conversations, Uneven Training, Limited Repetition

The health system in this case study (de-identified) had:

• Strong clinical training infrastructure, with limited capacity for high-frequency soft-skill practice

• Variability in learner exposure to difficult conversations across units and preceptors

• A measurable readiness gap: staff could “know the steps” clinically but struggled in high-emotion interactions

Pain points surfaced in internal review:

• Inconsistent practice for de-escalation and empathy skills

• Limited standardization of communication coaching across educators

• Difficulty measuring progress beyond self-report or one-time observation

Guiding question:
How can we make communication, empathy, and de-escalation skills as repeatable and measurable as clinical skills, without adding headcount?

3. Objectives

The health system and Patient Ready defined objectives designed to be template-ready for similar partners:

1. Increase practice frequency for core soft skills: empathy, structured communication, conflict navigation, and de-escalation (Cho and Kim 2024) (Lin et al. 2024).

2. Standardize scenario exposure and baseline feedback using AI-supported artifacts (for example, transcripts and structured summaries), improving coaching consistency and reducing repetitive educator effort (Stamer et al. 2023) (Elhilali et al. 2025).

3. Strengthen readiness for high-risk interactions by incorporating VR conversational practice loops and escalation pathways (Moore et al. 2022) (Clay et al. 2025).

4. Support interprofessional collaboration through VR-enabled handoff and team communication practice that is feasible to deliver at scale (Neher et al. 2024) (Liaw et al. 2020).

5. Demonstrate value through measurable process metrics (practice reps completed, completion rates, time-to-debrief readiness, coaching minutes per learner) and outcomes (communication skill assessment, safety event proxies where applicable).

4. Approach: AI + VR as a Soft Skills Readiness Layer

Patient Ready’s approach was framed as additive, not replacement: preserve in-person learning where it is uniquely valuable, and use VR + AI where repetition, standardization, and data capture create readiness leverage.

4.1 Scenario Selection: Match Soft Skill to Practice Loop

The team prioritized objectives that benefit from emotionally realistic repetition, such as:

• De-escalating an escalating patient or family interaction

• Responding to refusal, fear, anger, or mistrust

• Delivering clear, compassionate explanations under time pressure

• Structured handoff and interprofessional coordination

• “Speak-up” behaviors for patient safety in psychologically safe language

These align with evidence that VR communication training can improve communication skills in nursing education and that VR can support team training outcomes comparable to live simulation in some contexts (Cho and Kim 2024) (Liaw et al. 2020).

4.2 Technology & Learning Model

The deployment included:

• Immersive VR scenarios for high-emotion interactions and team-based decision points

• Screen-based options for rapid access and broader distribution when VR hardware availability was limited

• AI-supported learner artifacts for debrief efficiency (for example, conversation summaries and moment tagging), consistent with AI’s role in communication training and feedback workflows (Stamer et al. 2023) (Shorey et al. 2019)

Evidence anchor: a systematic review and meta-analysis found VR simulation had a significant overall effect on nursing students’ communication skills (effect size 0.44) (Cho and Kim 2024).

4.3 Debriefing Workflow (Best-Practice Aligned)

Each session followed a consistent structure:

1. Prebrief (5 to 10 min): psychological safety, objectives, communication frames

2. VR or screen-based scenario (10 to 20 min): individual or team

3. AI-supported artifact review (3 to 5 min): key moments, patterns, gaps

4. Facilitated debrief (10 to 20 min): reflection, intent vs impact, alternative phrasing

5. Repetition loop: learners re-run with targeted goals to close gaps (directional; program-reported)

5. Implementation: A Phased, Low-Friction Rollout

Phase 1: Pilot in One High-Need Use Case

• Start with a limited scenario set tied to clear operational pain (for example, de-escalation and difficult conversations)

• Capture baseline learner confidence and educator time inputs (directional; program-reported)

Phase 2: Standardize Across Cohorts

• Establish a simple “practice cadence” (for example, monthly reps for core scenarios)

• Normalize debrief prompts and evaluation language for coaching consistency

Phase 3: Expand to Interprofessional Training

• Add team-based handoff and escalation scenarios

• Leverage VR’s flexibility and reduced resource burden for multi-profession training (Neher et al. 2024)

6. Results: More Practice, More Consistency, More Coaching Bandwidth (Directional Outcomes)

6.1 Standardization: Same Scenarios, Same Expectations, Better Comparability

Educators reported improved consistency in what learners practiced and how performance was discussed (directional; program-reported). This direction aligns with evidence that VR communication training can improve skills outcomes and enables more standardized exposure than ad hoc role-play (Cho and Kim 2024).

6.2 Educator Efficiency: From Repetition to Coaching

With AI-supported artifacts, educators reported spending less time reconstructing what happened and more time on:

• intent vs impact

• empathy and clarity

• escalation choices and phrasing alternatives

• transfer-to-practice planning

This direction is consistent with the role of AI in supporting communication skills training, including feedback provision and scalable practice access (Stamer et al. 2023) (Elhilali et al. 2025).

6.3 Safety and Workforce Relevance: Why De-escalation Practice Matters

Workplace violence prevention training often relies on experiential learning, and VR is positioned as a scalable option for practicing in a safe environment (Clay et al. 2025). Early findings also support the feasibility of VR-based conversational agents for de-escalation practice (Moore et al. 2022).

Insight-to-Impact Bridge

Across healthcare, the bottleneck is not knowing what “good communication” is. It is having enough realistic, repeatable practice to perform under stress. When VR carries a share of emotionally realistic reps and AI helps make feedback consistent, educator time shifts toward higher-order coaching, and organizations get a scalable pathway for safer interactions and better patient experience.

7. Strategic Takeaways for Leaders

For CNOs, Nursing Leadership, and Workforce Leaders

• Treat soft skills as measurable capabilities: define a practice cadence, not a one-time workshop.

• Prioritize de-escalation as both a safety and retention strategy, supported by scalable experiential training approaches (Clay et al. 2025).

For Simulation and Education Leaders

• Use VR to standardize exposure and reduce variability across facilitators and cohorts.

• Blend modalities: VR where emotional realism matters most, screen-based where distribution and frequency are the priority.

For Faculty, Preceptors, and Clinical Educators

• Teach at the top of license: use AI-supported artifacts to accelerate debrief readiness and focus on coaching, not recall.

• Use repetition loops to build fluency, not just awareness.

8. Future Directions

Building on early success, the system is exploring:

• A longitudinal “soft skills readiness dashboard” (completion, repetition, performance trend-lines)

• Expanded inter-professional scenarios (handoff, conflict resolution, shared decision-making)

• More structured measurement using validated tools and learning analytics, aligned to VR training feasibility evidence (Neher et al. 2024)

9. References

1. Cho MK, Kim MY. The effect of virtual reality simulation on nursing students’ communication skills: a systematic review and meta-analysis. Frontiers in Psychiatry. 2024. https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2024.1351123/full
   Source label: Frontiers

2. Stamer T, Steinhäuser J, Flägel K. Artificial Intelligence Supporting the Training of Communication Skills in the Education of Health Care Professions: Scoping Review. Journal of Medical Internet Research. 2023;25:e43311. https://www.jmir.org/2023/1/e43311/
   Source label: JMIR

3. Liaw SY, Ooi SW, Rusli KDB, Lau TC, Tam WWS, Chua WL. Nurse-Physician Communication Team Training in Virtual Reality Versus Live Simulations: Randomized Controlled Trial on Team Communication and Teamwork Attitudes. Journal of Medical Internet Research. 2020;22(4):e17279. https://www.jmir.org/2020/4/e17279/
   Source label: JMIR

4. Moore N, Ahmadpour N, Brown M, Poronnik P, Davids J. Designing Virtual Reality–Based Conversational Agents to Train Clinicians in Verbal De-escalation Skills: Exploratory Usability Study. JMIR Serious Games. 2022;10(3):e38669. https://games.jmir.org/2022/3/e38669/
   Source label: JMIR Serious Games

5. Clay CJ, Hochmuth JM, Wirth O. Virtual Reality Training to Reduce Workplace Violence in Healthcare. CDC Stacks (NIOSH). 2025. https://stacks.cdc.gov/view/cdc/208660
   Source label: CDC

6. Lin HL, Wang YC, Huang ML, et al. Can virtual reality technology be used for empathy education in medical students: a randomized case-control study. BMC Medical Education. 2024. https://link.springer.com/article/10.1186/s12909-024-06009-6
   Source label: BMC Medical Education

7. Shorey S, Ang E, Yap J, Ng ED, Lau ST, Chui CK. A Virtual Counseling Application Using Artificial Intelligence for Communication Skills Training in Nursing Education: Development Study. Journal of Medical Internet Research. 2019;21(10):e14658. https://www.jmir.org/2019/10/e14658/
   Source label: JMIR

8. Neher AN, Wespi R, Rapphold BD, et al. Interprofessional Team Training With Virtual Reality: Acceptance, Learning Outcome, and Feasibility Evaluation Study. JMIR Serious Games. 2024;12:e57117. https://games.jmir.org/2024/1/e57117/
   Source label: JMIR Serious Games

9. Elhilali A, et al. Large Language Model–Based Patient Simulation to Foster Communication Skills in Health Care Professionals: User-Centered Development and Usability Study. JMIR Medical Education. 2025;11:e81271. https://mededu.jmir.org/2025/1/e81271
   Source label: JMIR Medical Education