Frailty in emergency general surgery: Low-risk procedures pose similar risk as high-risk procedures for frail patients.

BACKGROUND: The association of frailty on postoperative outcomes after elective and emergency general surgery procedures has been widely studied. However, this association has not been examined in the geriatric population stratified by emergency general surgery procedural risk. METHODS: A retrospective cohort study was performed using the 2012 to 2017 American College of Surgeons-National Surgical Quality Improvement Program database. We identified geriatric patients (age ≥65 years) undergoing an emergency general surgery procedure within 48 hours of admission stratified by the procedural risk. Frailty was accessed using Modified 5-item Frailty Index, and the patients were divided into 4 groups Modified 5-item Frailty Index = 0, 1, 2, and ≥3. Multivariable logistic regression was used to assess the impact of increasing Modified 5-item Frailty Index score on postoperative complications, failure-to-rescue, and readmissions. RESULTS: In the study, 16,911 low risk procedure emergency general surgery patients were grouped as (33.3%) Modified 5-item Frailty Index = 0, (45.1%) Modified 5-item Frailty Index = 1, (18.7%) Modified 5-item Frailty Index = 2, and (2.9%) Modified 5-item Frailty Index ≥3 respectively. After multivariable analyses, increasing Modified 5-item Frailty Index score (versus Modified 5-item Frailty Index = 0) was associated with complications (odds ratio [95% confidence interval]; Modified 5-item Frailty Index = 2: 2.1 [1.3-3.5], Modified 5-item Frailty Index ≥ 3: 2.2 [1.2-4.2]), failure-to-rescue (Modified 5-item Frailty Index = 2: 2.3 [1.3-4.0], Modified 5-item Frailty Index ≥ 3: 2.3 [1.2-4.6]), readmission (Modified 5-item Frailty Index = 2: 1.4 [1.2-1.7], Modified 5-item Frailty Index ≥ 3: 1.5 [1.1-2.1]). In addition, 30,305 high-risk patients undergoing procedure emergency general surgery were grouped as (24.1%) Modified 5-item Frailty Index = 0, (44.9%) Modified 5-item Frailty Index = 1, (24.0%) Modified 5-item Frailty Index = 2, and (7.0%) Modified 5-item Frailty Index ≥3, respectively. After multivariable analyses, increasing Modified 5-item Frailty Index score (versus Modified 5-item Frailty Index = 0) was associated with complications (odds ratio [95% confidence interval]; Modified 5-item Frailty Index = 2: 1.2 [1.2-1.3], Modified 5-item Frailty Index ≥3: 1.7 [1.5-2.0]), failure-to-rescue (Modified 5-item Frailty Index = 2: 1.3 [1.2-1.5], Modified 5-item Frailty Index ≥3: 1.5 [1.3-1.7]), readmission (Modified 5-item Frailty Index = 2: 1.3 [1.2-1.4], Modified 5-item Frailty Index ≥3: 1.6 [1.4-1.9]). CONCLUSION: Increasing levels of frailty in geriatric emergency general surgery patients are associated with higher levels of postoperative complications, failure-to-rescue, and readmission. Clinicians should consider frailty in assessing the risk of even low-risk surgeries in this population.

Artificial intelligence in virtual standardized patients: Combining natural language understanding and rule based dialogue management to improve conversational fidelity.

INTRODUCTION: Advances in natural language understanding have facilitated the development of Virtual Standardized Patients (VSPs) that may soon rival human patients in conversational ability. We describe herein the development of an artificial intelligence (AI) system for VSPs enabling students to practice their history taking skills. METHODS: Our system consists of (1) Automated Speech Recognition (ASR), (2) hybrid AI for question identification, (3) classifier to choose between the two systems, and (4) automated speech generation. We analyzed the accuracy of the ASR, the two AI systems, the classifier, and student feedback with 620 first year medical students from 2018 to 2021. RESULTS: System accuracy improved from ∼75% in 2018 to ∼90% in 2021 as refinements in algorithms and additional training data were utilized. Student feedback was positive, and most students felt that practicing with the VSPs was a worthwhile experience. CONCLUSION: We have developed a novel hybrid dialogue system that enables artificially intelligent VSPs to correctly answer student questions at levels comparable with human SPs. This system allows trainees to practice and refine their history-taking skills before interacting with human patients.