Legler, Sean R.

Abstract Objective We evaluated the quality and adoption of a large language model (LLM)-based summarization tool for ongoing hospital care. Materials and methods An AI summarization tool was created to provide a “Patient Story”, specialty-specific “Recent Notes”, and “Recent Events” over the previous 24 or 72 hours. We conducted a pragmatic mixed-methods quality assessment at three tertiary-care academic hospitals utilizing (1) Provider Documentation Summarization Quality Instrument (PDSQI-9), (2) utilization analytics, and (3) qualitative end user feedback. The PDSQI-9 included whether summaries were accurate, cited, comprehensible, organized, succinct, non-stigmatizing, synthesized, thorough, and useful. 512 users were given access, from whom 52 participants submitted 205 surveys (10.2% response rate). Results 52 respondents submitted an average of 4.3 surveys (range 1–8). Users rated the tool favorably across all PDSQI-9 domains, with a combined average score of 4.68 (range 4.56–4.78, S.D. 0.67) across the eight domains scored on a 5-point modified Likert scale. Utilization metrics demonstrated strong uptake with frequent views. Positive qualitative feedback revealed cognitive offloading for complex patients and effective summarization of medical problems. Critical feedback showed a need to cross-reference narrative notes to current-state data and lack of detailed specialty-specific summarization. Discussion Generative artificial intelligence has emerged as a potentially transformative technology for generating succinct, verifiable summaries of ongoing care. In this pragmatic implementation, end-users indicated high perceived quality across PDSQI-9 domains. Conclusions An LLM-based summarization tool for ongoing hospitalization care was rated of high quality by diverse clinicians in real-world settings, demonstrated a favorable safety profile, and showed sustained utilization.

Introduction Optimal hospital bed utilization requires innovative patient care models. We studied a novel hospitalist model utilizing telemedicine to facilitate collaboration with affiliated emergency departments (EDs) and support medical triage and care of ED patients with high likelihood of hospital admission. Methods Telehospitalists based at a tertiary care facility collaborated with four community EDs in the same healthcare network between January 1, 2022, and April 30, 2023. Telehospitalists supported ED clinicians in medical care decisions and facilitated patient disposition. Emergency department length of stay (LOS) and disposition were evaluated, as were hospital LOS, 30-day readmission, and in-hospital mortality. For patients discharged from the ED, 7-day ED readmission and subsequent hospitalization were evaluated. Results Telehospitalists discussed 550 “admit-likely” patients with ED clinicians: 105 patients (19.1%) discharged from the ED and avoided admission; 322 patients (58.5%) were admitted to local or nearby community hospitals; 123 patients (22.4%) transferred to the tertiary care facility. Emergency department LOS differed significantly among disposition groups, including patients discharged home (10.2 h), admitted to local hospitals (12.6 h), and transferred to tertiary care hospitalist services (14.9 h; p < 0.001). Hospital LOS and in-hospital mortality were not significantly different among disposition groups. Patients admitted locally had lower 30-day readmission compared to those transferred to tertiary care facility (odds ratio = 0.59 [0.36, 0.99], p = 0.048). Discussion Telehospitalists as triage clinicians is an innovative approach to support local ED clinicians and patients. Telehospitalists optimized hospital bed utilization and healthcare system resources by facilitating safe discharges to home and expediting tertiary care transfers when necessary.

Introduction: The association of patient satisfaction with hospital ACS care and risk for long-term outcomes is incompletely described. This is particularly relevant to the Gulf region, where the ACS burden is high and literature on patient satisfaction and outcomes is lacking. The aim was to compare determinants of 1-year mortality based on patient satisfaction with ACS care in the Gulf region. Methods: In a prospective registry of 3566 ACS patients from Bahrain, Kuwait, Oman, and UAE (Gulf COAST registry), we grouped patients by their self-reported overall satisfaction with ACS care reported at 1-month following ACS (low satisfaction [n=1654] vs. high satisfaction [n=1912]; Table). We examined associations of baseline characteristics and risk for 1-year mortality in logistic regression models adjusted for Global Registry of Acute Coronary Events (GRACE) score. Results: Mortality at 1-year following ACS was higher in patients reporting low satisfaction (8.8%; n=146/1654) vs. high satisfaction (5.4%; n=103/1912) ( P <.0001). Women, compared with men, had higher 1-year mortality risk in both satisfaction groups, and statistically significant risk in the low satisfaction group (aOR 1.45 [1.01–2.07]; P <.05). Lack of employment (vs. any employment) was associated with higher 1-year mortality risk in both the low satisfaction (aOR 6.44 [2.58–16.03]) and high satisfaction (aOR 2.58 [1.35–4.93]) groups with borderline significant interaction ( P interaction = .08). In both satisfaction groups, lack of education, hypertension and prior angina were associated with higher 1-year mortality risk, whereas dyslipidemia and diabetes showed no association. Conclusions: Low patient satisfaction with ACS care was associated with higher 1-year mortality risk. Socioeconomic factors such as employment and education were associated with higher 1-year mortality, independent of patient satisfaction, and may guide mortality reduction interventions.