Objectives: This study presents a systematic review aimed at evaluating the effectiveness of virtual patients in enhancing clinical reasoning skills in medical education. A hybrid methodology was used, combining human reviewers and ChatGPT to assess the impact of conversational virtual patients on student learning outcomes and satisfaction. Methods: Various studies involving conversational virtual patients were analyzed to determine the effect of these digital tools on clinical competencies. The hybrid review process incorporated both human assessments and AI-driven reviews, allowing a comparison of accuracy between the two approaches. Results: Consistent with previous systematic reviews, our findings suggest that conversational virtual patients can improve clinical competencies, particularly in history-taking and clinical reasoning. Regarding student feedback, satisfaction tends to be higher when virtual patients’ interactions are more realistic, often due to the use of artificial intelligence (AI) and natural language processing (NLP) in the simulators. Furthermore, the study compares the accuracy of AI-driven reviews with human assessments, revealing comparable results. Conclusions: This research highlights AI’s potential to complement human expertise in academic evaluations, contributing to more efficient and consistent systematic reviews in rapidly evolving educational fields.

Background: Virtual simulated patients (VSPs) powered by generative artificial intelligence (GAI) offer a promising tool for training clinical interviewing skills; yet, little is known about how different system- and user-level variables shape students’ perceptions of these interactions. Objective: We aim to study psychology students’ perceptions of GAI-driven VSPs and examine how demographic factors, system parameters, and interaction characteristics influence such perceptions. Methods: We conducted a total of 1832 recorded interactions involving 156 psychology students with 13 GAI-generated VSPs configured with varying temperature settings (0.1, 0.5, 0.9). For each student, we collected age and sex; for each interview, we recorded interview length (total number of question–answer turns), number of connectivity failures, the specific VSP consulted, and the model temperature. After every interview, students provided a 1-10 global rating and open-ended comments regarding strengths and areas for improvement. At the end of the training sequence, they also reported perceived improvement in diagnostic ability. Statistical analyses assessed the influence of different variables on global ratings: demographics, interaction-level data, and GAI temperature setting. Sentiment analysis was conducted to evaluate the VSPs’ clinical realism. Results: Statistical analysis showed that female students rated the tool significantly higher (mean rating 9.25/10) than male students (mean rating 8.94/10; Kruskal-Wallis test, H=8.7; P=.003). On the other side, no significant correlation was found between global rating and age (r=0.02, 95% CI –0.03 to 0.06; P=.42), interview length (r=0.04, 95% CI –0.2 to 0.10; P=.18), or frequency of participation (Kruskal-Wallis test, H=4.62; P=.20). A moderate negative correlation emerged between connectivity failures and ratings (r=–0.26, 95% CI –0.41 to –0.10; P=.002). Temperature settings significantly influenced ratings (Kruskal-Wallis test, H=6.93; P=.03; η²=0.02), with higher scores at temperature 0.9 compared with 0.1 (Dunn’s test, P=.04). Concerning learning outcomes, self-perceived improvement in diagnostic ability was reported by 94% (94/100) of students; however, final practical examination scores (mean 6.67, SD 1.42) did not differ significantly from those of the previous cohort without VSP training (mean 6.42, SD 1.56). Sentiment analysis indicated predominantly negative sentiment in GAI responses (median negativity 0.8903, IQR 0.306-0.961), consistent with clinical realism. Conclusions: GAI-driven VSPs were well-received by psychology students, with student gender and system-level variables (particularly temperature settings and connection stability) shaping user evaluations. Although participants perceived the training as beneficial for their diagnostic skills, objective examination performance did not significantly differ from the previous cohort. However, lack of randomization limits the generalization of the results obtained, and further experiments are required.

This study presents a systematic review to evaluate the effectiveness of virtual patients in enhancing clinical reasoning skills in medical education, utilizing a hybrid methodology that combines human reviewers and ChatGPT. By analyzing various studies involving conversational virtual patients, the review examines the impact of these digital tools on student learning outcomes and satisfaction. Consistent with previous systematic reviews, our findings suggest that conversational virtual patients can improve clinical competencies, particularly in history-taking and clinical reasoning. Regarding student feedback, satisfaction tends to be higher when virtual patients’ interactions are more realistic, often due to the use of artificial intelligence (AI) and natural language processing (NLP) in the simulators. Furthermore, the study compares the accuracy of AI-driven reviews with human assessments, revealing comparable results. This research highlights AI’s potential to complement human expertise in academic evaluations, contributing to more efficient and consistent systematic reviews in rapidly evolving educational fields.

Background: Health care institutions face high systemic risk due to the inherent uncertainty and complexity of their operations. This often leads to stressful incidents impacting the well-being of health care professionals, which can compromise the effectiveness of health care systems. Enhancing resilience among health care professionals is essential for maintaining high-quality care and ensuring patient safety. The role of middle managers is essential to ensure the response capacity of individuals and teams. Objective: This study aims to develop a web-based intervention aimed at middle management to enhance individual, team, and organizational resilience. Methods: An observational study was conducted in 3 phases: design, validation, and pilot study. The study was initiated in February 2022 and concluded in June 2023. Phase 1 involved designing the content for the web-based tool based on a comprehensive review of critical elements around resilience. Phase 2 included validation by an international panel of experts who reviewed the tool and rated it according to a structured grid. They were also encouraged to highlight strengths and areas for improvement. Phase 3 involved piloting the tool with health care professionals in Ecuador to refine the platform and assess its effectiveness. A total of 458 people were invited to participate through the Institutional Course on Continuous Improvement in Health Care Quality and Safety offered by the Ministry of Public Health of Ecuador. Results: The tool, eResiliencia, was structured into 2 main blocks: individual and team resilience and organizational resilience. It included videos, images, PDFs, and links to dynamic graphics and additional texts. Furthermore, 13 (65%) of the 20 experts validated the tool, rating content clarity at an average of 4.5 (SD 0.7) and utility at an average of 4.7 (SD 0.5) out of 5. The average overall satisfaction was 9.3 (SD 0.6) out of 10 points, and feedback on improvements was implemented. A total of 362 health care professionals began the intervention, of which 218 (60.2%) completed preintervention and postintervention questionnaires, with significant knowledge increases (P<.001). Of the 362 health care professionals, 146 (40.3%) completed the satisfaction questionnaire, where overall satisfaction was rated at an average of 9.4 (SD 1.1) out of 10 points. Conclusions: The eResiliencia web-based platform provides middle managers with resources to enhance resilience among their teams and their components, promoting better well-being and performance, even under highly stressful events. Future research should focus on long-term impacts and practical applications in diverse clinical settings.

Background: Long-term care hospitals have been considered an efficient response to the health care needs of an increasingly aging population. These centers are expected to contribute to better hospital bed management and more personalized care for patients needing continuous care. The evaluation of their outcomes is necessary after a sufficient period to assess their impact. Hospitals for Acute and Chronic Long-Term Extended Stay (HACLES) emerged in Spain in the late 20th century as a response to the aging population and the increase in chronic diseases. Objective: This study aimed to analyze the profile of patients treated in a HACLES, particularly analyzing gender differences, and evaluate the cost savings associated with using these centers. Methods: A retrospective study was conducted based on data from patients 65 years old or older admitted to a HACLES between 2022 and 2023. Gender, age, household cohabitation data, diagnosis and comorbidity, daily medication intake, and degree of dependency were obtained to describe the profile of patients who attended the HACLES. Data coded in SIA-Abucasis (version 37.00.03; Consellería Sanitat, Generalitat Valenciana; a digital medical record system used in the Valencian region) were reviewed, and descriptive statistics and comparison tests were used. The direct cost savings of HACLES admissions were calculated by comparing the daily cost of a general hospital bed with that of a HACLES bed. Results: Data from 123 patients with a mean age of 77 years were analyzed. Most (n=81, 65.9%) had a cohabiting family member as their primary caregiver. Palliative care was the most frequent reason for admission (n=75, 61%). The mortality rate (odds ratio [OR] 61.8, 95% CI 53.2-70.5) was similar between men and women (OR 54.1, 95% CI 47.8-71.5 vs OR 59.7, 95% CI 42.2-66.0; P=.23). The cognitive assessment, using the Pfeiffer scale, improved at discharge (mean 3.2, SD 3.2 vs mean 2.5, SD 3.1; P=.003). The length of stay was significantly larger for patients who returned home compared with patients discharged to other facilities (mean 89.8, SD 58.2 versus mean 33.1, SD 43.1 days; P<.001). The direct cost savings were estimated at US $42,614,846 per 1000 admissions. Conclusions: Patients typically treated in HACLES are older, with a high level of cognitive impairment and physical dependency, and a significant proportion are in palliative care, highlighting the importance of adapting care to the individual needs of the admitted patients. The HACLES model contributes to the sustainability of the public health system.