To Örebro University

BACKGROUND: Virtual patients (VPs) demonstrate effectiveness in improving clinical reasoning skills; however, traditional VP platforms often lack individualized feedback mechanisms. Advances in large language models (LLMs) enable automated analysis of student-VP interactions, providing scalable feedback on clinical performance. While artificial intelligence (AI)-enhanced social robotic VP platforms show promise for clinical reasoning training, no studies have examined whether AI-generated feedback integrated in such platforms improves clinical performance in standardized assessments.

OBJECTIVE: This study evaluated whether AI-generated postconsultation feedback integrated into social robotic VP interactions improves medical students' clinical performance, emphasizing medical history taking and communication.

METHODS: A quasi-experimental study with 115 sixth-semester medical students (N=157, 73.2% of eligible students) was conducted at Karolinska Institutet, Stockholm, Sweden, during spring 2025. Students were allocated by hospital site to receive (n=61, 53%) or not receive (n=54, 46.9%) AI-generated feedback following interactions with a Social AI-Enhanced Robotic Interface. All students completed 9 VP cases; the intervention group received approximately 1 page of structured feedback after each VP case. The feedback system used multiple LLMs following a 2-stage algorithm: assessing student-VP dialogues using an assessment rubric, then generating structured feedback on history-taking performance. Both groups participated in case-specific follow-up seminars led by consultant rheumatologists following each VP encounter. Clinical performance was assessed through an 8-minute objective structured clinical examination (OSCE)-based evaluation, with a standardized patient portraying axial spondylarthritis, evaluated by a blinded consultant rheumatologist using a 10-point rubric across 5 domains: communication at consultation start, generic medical history, targeted medical history, diagnostics and management reasoning, and communication at consultation end.

RESULTS: Students receiving AI-generated feedback achieved significantly higher total OSCE scores (mean 7.39, SD 0.86 vs mean 6.68, SD 1.04 points; mean difference 0.70; 95% CI 0.35-1.06; P<.001; Cohen d=0.74). Domain-specific analysis revealed significant improvement in generic medical history after Bonferroni correction (mean 2.46, SD 0.65 vs mean 2.03, SD 0.79 points; P=.004; r=0.27), while other domains showed no significant differences: communication at start (P=.13; r=0.14), targeted medical history taking (P=.60; r=0.05), diagnostics and management (P=.14; r=0.14), and communication at consultation end (P=.31; r=0.09). Pass rates were significantly higher in the feedback group (96.7% vs 79.6%; odds ratio 7.55, 95% CI 1.51-72.2; P=.006), with a number needed to assess of 6 students, that is, for every 6 students receiving feedback, 1 additional student passed the assessment.

CONCLUSIONS: AI-generated feedback following social robotic VP interactions significantly improved medical students' OSCE-based performance, particularly in generic medical history taking. These findings support integrating validated AI feedback systems as a supplement to expert-led teaching during VP simulations for clinical training and demonstrate the feasibility of scalable, automated feedback in medical education. The domain-specific improvements in generic medical history highlight the importance of targeted, competency-specific feedback design in VP platforms.

INTRODUCTION: Future physicians must develop the necessary knowledge and skills to effectively manage challenging patient communication. To design linked meaningful learning activities, perspectives from students and teachers need to be taken into account as well as students' previous knowledge. Aiming to better understand how this knowledge can be acquired through patient simulation, we examined students' perceptions, mirrored in teachers' reflections, and how these relate to the medical curriculum.

METHODS: Data were collected from students and teachers during the ninth semester of a medical program in Sweden. A narrative survey and individual interviews were conducted. Using thematic analysis, we analyzed nineteen student surveys, five student interviews, and four teacher interviews to explore perceptions of learning. The student themes were then used to identify corresponding subthemes in the teacher data through a mirroring process.

RESULTS: Self-knowledge was identified as a crucial skill in order to effectively convey empathy in challenging patient communications. Patient simulation was described as a valuable learning activity which, however, lacked integration with adjacent learning activities in the medical program. Learning progression was hindered by a disconnect between communication-related learning activities and those focused on medical knowledge.

CONCLUSION: Both self-knowledge and medical knowledge are perceived as essential for learning how to manage challenging patient communication. Our findings suggest that the learning of patient communication and medical knowledge should be integrated in the curriculum and reinforced across various settings. Patient simulation is a valuable method for developing self-knowledge and communication skills.

AIM: To investigate specialist nurses' experience of psychological safety in ad hoc teams during emergency care.

DESIGN: Interpretive descriptive qualitative study. METHODS: Semi-structured interviews with nine specialist nurses were conducted in Sweden from May to June 2024 and analysed using reflexive thematic analysis (Braun & Clarke).

RESULTS: Four themes were identified: Interpersonal skills: implications for psychological safety; Individuality and team dynamics; Confidence, competence and collaboration; and Organisational responsibility for promoting psychological safety.

CONCLUSION: Psychological safety in ad hoc emergency care teams is a fragile and multifaceted phenomenon, shaped by interpersonal skills, leadership and organisational culture. Supportive environments characterised by open communication and proactive leadership enable specialist nurses to collaborate confidently and safely, even under acute stress. Targeted efforts to strengthen these factors are essential for optimising teamwork and patient outcomes.

IMPLICATIONS FOR THE PROFESSION: The psychological safety implications for specialist nurses in ad hoc teams during emergency care are profound. Psychological safety fosters an environment that empowers nurses to leverage clinical expertise, collaborate in ad hoc teams and improve patient outcomes. Promoting psychological safety ensures specialist nurses feel respected, valued and secure, leading to better care and a more resilient workforce.

IMPACT: This qualitative study investigated specialist nurses' experience of psychological safety in ad hoc teams in acute care. The results will influence the awareness of nurses, specialist nurses, other professions, managers and organisations about the importance of feeling psychologically safe.

REPORTING METHOD: Presentation follows COREQ 32-item checklist.

PATIENT AND PUBLIC INVOLVEMENT: No patient or public involvement.

CONTRIBUTION TO GLOBAL COMMUNITY: Shows that psychological safety helps nurses perform in ad hoc emergency teams. Identifies key factors that affect teamwork and patient care.

Objective: To explore whether an AI-enhanced social robotic virtual patient (VP) platform reinforces empathetic behaviour patterns in medical students compared with a traditional computer-based platform.

Methods: Twenty-three sixth-semester medical students from Karolinska Institutet participated in semi-structured interviews following VP encounters with the Social AI-enhanced Robotic Interface (SARI) and, as a comparator, the computer-based Virtual Interactive Case system (VIC). Additionally, 178 students evaluated the VP platforms in empathetic training quantitatively using categorical nominal variables and a visual analogue scale (VAS), with a score of 0 indicating full preference for SARI and 10 full preference for VIC. Interview data were thematically analysed, and quantitative preferences were compared using the Fisher's exact test with Monte Carlo simulation and the Wilcoxon signed-rank test.

Results: Thematic analysis yielded five major themes wherein students consistently reported that SARI facilitated greater empathetic engagement through multimodal interaction, ability to express emotions, and real-time communication adaptability. Quantitative analysis demonstrated a higher preference for SARI versus VIC (78% versus 6%; OR: 190.4; 95% CI: 76.8-472.0; p < 0.001), which remained consistent across subgroups of interest, i.e., female and male students, with and without prior experience in VPs, and students first exposed to SARI or first exposed to VIC. VAS data also showed a preference for SARI versus VIC (median: 2.00; IQR: 1.00-4.00; W: 738.5; r: 0.70; p < 0.001).

Conclusion: Our AI-enhanced social robotic VP platform was superior to a traditional computer-based VP platform in fostering empathetic engagement in medical students through enhanced authenticity and interactivity, supporting its potential to supplement clinical rotations.

Objective/Background: Integration of virtual patient (VP) cases has traditionally complemented clinical encounters during medical students clinical placements at Karolinska Institutet (KI). This study aimed to assess the educational value of clinical reasoning (CR) training by comparing two platforms: a novel large language model (LLM)-enhanced social robotic VP platform and a conventional computer-based platform within the context of rheumatology.

Methods/summary of work: A qualitative study involved 23 third-year medical students from KI during clinical placements in rheumatology. Each student completed nine VP cases using two distinct platforms: an LLM-enhanced social robotic platform and a computer-based semi-linear platform. Following each case completion, students participated in seminars with consultant rheumatologists to discuss the clinical cases. In-depth interviews assessed students’ self-perceived acquirement of CR skills using the two platforms following all VP cases and their corresponding seminars. Thematic analysis was employed to identify themes and sub-themes.

Results/Summary of results: Thematic analysis revealed three principal themes: authenticity, VP application, and strengths and limitations. Students perceived the social robotic platform more authentic and engaging, particularly through its capacity for interactive communication and emotional expression, collectively delivering a realistic clinical experience. The platform demonstrated effectiveness in facilitating active learning processes, hypothesis formation, and adaptive thinking skills. However, notable limitations were identified, including the absence of physical examination capabilities and instances of mechanical dialogue patterns.

Conclusion: In our setting of undergraduate medical education placements within rheumatology, an LLM-enhanced social robotic VP platform offered a more authentic and interactive learning experience compared to a conventional computer-based platform. Despite some limitations, the social robotic platform shows promise in training CR skills, communication, and adaptive thinking. AI-enhanced Social robotic VPs may prove useful learning modalities for exposing medical students to diverse, highly interactive patient simulations.

BACKGROUND: Virtual patient (VP) simulations can be used to practice clinical reasoning (CR) in controlled learning environments. Traditional computer-based VP platforms often lack the authenticity and interactivity required for effective CR training. Artificial intelligence (AI)-enhanced social robotic VPs can enhance realism and engagement; however, quantitative evidence comparing them with conventional VP platforms remains limited.

OBJECTIVE: We compared medical students' experience of an AI-enhanced social robotic versus a conventional computer-based VP platform regarding the extent to which the design characteristics of the respective platform facilitate CR skill training.

METHODS: This observational crossover cohort study involved 178 sixth-semester medical students at Karolinska Institutet, Stockholm, Sweden (response rate: 42.3%; 178 of 421 invited students; Spring 2024-Spring 2025), who experienced both a large language model-enhanced social robotic VP platform supporting dialogue (social artificial intelligence-enhanced robotic interface [SARI]) and a conventional computer-based VP platform (virtual interactive case [VIC]) during their clinical rotation within rheumatology. Platform order was determined by clinical rotation scheduling. VP design was evaluated using a validated questionnaire across 5 domains: authenticity, professional approach, coaching quality, learning effects, and overall judgment. Students' CR training preferences were assessed using categorical responses and a Visual Analogue Scale, where a lower score favored SARI and a score of 5 indicated equal preference between platforms.

RESULTS: SARI outperformed VIC across all 5 VP design domains. Students rated SARI higher for authenticity (median 4.0, IQR 3.5-4.5 vs 3.0, IQR 2.5-3.5; P<.001), professional approach (median 4.5, IQR 4.0-4.8 vs 4.0, IQR 3.5-4.5; P<.001), coaching quality (median 4.3, IQR 4.0-4.7 vs 4.0, IQR 3.7-4.7; P<.001), learning effect (median 4.4, IQR 4.0-5.0 vs 4.0, IQR 3.5-4.5; P<.001), and overall judgment (median 5.0, 4.0-5.0 vs 4.0, IQR 4.0-5.0; P<.001). Students strongly preferred SARI for CR training (72% vs 14%; odds ratio [OR] 27.1, 95% CI 14.3-53.7; P<.001), with Visual Analogue Scale scores confirming this preference (median 3.0, IQR 2.0-5.0; P<.001). Preferences were consistent across most subgroups (sex, prior VP experience, and platform order); in 2 subgroups, the difference was not significant, that is, students with prior VP experience (62% vs 38%; OR 2.6; 95% CI 0.8-8.9; P=.11) and students first introduced to VIC (55% vs 45%; OR 1.5; 95% CI 0.7-2.9; P=.33).

CONCLUSIONS: Our findings provide the first quantitative evidence that AI-enhanced social robotic VPs offer superior design characteristics than conventional computer-based platforms for CR training in medical education. These results support the use of AI-driven social robots for VP simulations to better prepare medical students for real clinical encounters, and warrant future research on objective CR skill outcomes and long-term transfer to clinical practice. Unlike previous qualitative studies examining each platform separately, this study provides the first quantitative comparison of design characteristics between AI-enhanced social robotic and conventional computer-based VPs.

This paper emphasizes adapting frameworks for interprofessional collaboration (IPC) from high-income settings to culturally distinct, resource-constrained regions. We adapted and validated the Interprofessional Education Collaborative (IPEC) competency framework within a Balkan healthcare context facing systemic, educational, and financial challenges. Frameworks such as IPEC may not fully capture contextual nuances in lower-middle-income countries (LMICs). Therefore, we examined how the IPEC framework could be culturally adapted for healthcare workers in the Balkans. A convergent mixed-methods design was employed, involving two interprofessional workshops and a follow-up survey. Healthcare workers from Bosnia and Herzegovina, Montenegro, and Kosovo participated in reflective discussions and prioritized competencies from the original 38-item IPEC instrument. The adapted instrument (IPEC-21B) was evaluated for psychometric reliability and contextual relevance and proved to be both valid and practical due to its shorter format. The instrument measured attitudes toward IPC and encouraged critical dialogue about hierarchy, communication, and shared roles, positioning it as both an assessment tool and a catalyst for professional learning. The results demonstrate how an adapted survey on IPC foster systemic thinking and professional development provides a tool for increased awareness. By retaining all core IPEC domains, IPEC-21B provides a tool for increasing awareness and developing interprofessional competencies in a way that aligns with local cultural and educational needs in settings with limited traditions of interprofessional practice. Tailoring an IPC instrument to local contexts enhances healthcare workers’ understanding of collaboration and may improve patient outcomes and care equity. The participatory adaptation process emphasizes embedding dialogue and shared understanding into interprofessional education and assessment in LMICs.

Purpose: To explore how stroke survivors experience and prefer to participate in clinical reasoning processes in the subacute phase of stroke rehabilitation.

Methods: An explorative qualitative design was used. Individual interviews were conducted with 10 stroke survivors (4 women and 6 men, mean age 68 years) 4 weeks after their stroke, and follow-up interviews were conducted with 6 of them after 10 weeks. The interview settings were the patient's home during their home rehabilitation, an inpatient and an outpatient rehabilitation unit. A reflexive thematic analysis was performed.

Results: Four themes were identified: discharge as a critical point for participation, describing a stressful time with varying involvement; supportive actions and context as crucial for participation, describing collaboration with the stroke team, the team's consideration of the stroke survivor's resources and needs, and a supportive home environment; the importance of goals and follow-up, describing goals as motivational and an unstructured use of goals; and difficulties in participation, describing a lack of dialogue with the stroke team and undetected resources and needs.

Conclusions: The stroke survivors experienced changes in their participation in the clinical reasoning process as their rehabilitation progressed. They moved from perceiving themselves as passengers at the time of their hospital discharge to gradually seeing themselves as the driver of their rehabilitation process. Some person-centered attributes, such as respectful relationships and a health focus, were incorporated into the clinical reasoning, while others, such as a holistic view and shared goal-setting, required further emphasis for improved person-centeredness in stroke rehabilitation.

BACKGROUND: Leadership is widely recognized as essential for fostering collaborative healthcare teams and improving patient outcomes. However, there is limited research on how leadership supports the development of nonclinical skills in healthcare settings in many low- and middle-income countries, including those in the Balkan region. This study addresses that gap by examining how leadership roles and practices enhance non-technical skills (NTSs)-such as communication, teamwork, and role clarity-among healthcare workers in the Balkans while also considering sustainable development, organizational values, cultural influences, and social dynamics.

METHODS: A qualitative approach was employed, drawing on data collected from three workshops conducted between 2018 and 2022 in university hospital clinics in Bosnia and Herzegovina, Kosovo, and Montenegro. Data sources included observations of group discussions, focus groups, and semi-structured interviews with healthcare leaders. Reflexive thematic analysis was used to identify patterns and develop key themes.

RESULTS: Four key themes emerged regarding the role of leadership in the development of NTSs: (1) defining roles and responsibilities, (2) fostering communication and teamwork, (3) promoting readiness for change, and (4) developing leadership competencies. The participants noted that clear role definitions enhanced team coordination, inclusive communication reduced misunderstandings, supportive leadership eased resistance to change, and mentorship served as a valuable mechanism for leadership development.

CONCLUSION: Leadership plays a key role in strengthening NTSs in Balkan healthcare contexts by promoting communication and teamwork within culturally and hierarchically complex environments. Role clarity, open dialogue, and shared accountability emerged as key factors for effective team performance and patient safety. These findings highlight the need for leadership development and the implementation of formal training initiatives-such as structured mentorship programs-to foster collaborative and resilient healthcare systems in low- and middle-income countries.