View details for DOI 10.1016/j.cardfail.2024.08.055

View details for PubMedID 39306232

View details for DOI 10.1016/j.jacc.2024.02.024

View details for PubMedID 38530685

Abstract

Limited real-world evidence exists for outcomes with contemporary guideline-directed medical therapy (GDMT) or GDMT with implantable cardioverter-defibrillator (ICD)/cardiac resynchronization therapy defibrillator (CRT-D) therapy for patients with heart failure with reduced ejection fraction (HFrEF) and left ventricular ejection fraction (LVEF) ≤35%.The present study aimed to assess survival associated with GDMT or GDMT with ICD/CRT-D therapy.This retrospective observational study included real-world de-identified data from January 1, 2016, to December 19, 2023, from 24 U.S. institutions per participating institutional agreements (egnite Database; egnite, Inc.). Patients with a diagnosis of HFrEF and an echocardiographic study documenting LVEF ≤35% were included for analysis.Of 43,591 patients with eligible index event of LVEF ≤35%, prescription history through ≥1 year preindex, and no ICD/CRT-D therapy preindex, mean ± standard deviation age at index was 71.2 ± 13.2 years; 14,805 (34.0%) patients were female. At 24 months, an estimated 99.1% (95% confidence interval [CI] 99.0%-99.2%), 89.9% (95% CI 89.7%-90.1%), 54.8% (95% CI 54.4%-55.2%), and 17.2% (95% CI 16.9%-17.5%), had ≥1, 2, 3, or all 4 GDMT classes prescribed, respectively; an estimated 15.7% (95% CI 15.3%-16.1%) had device placement. Of those without a device, by 24 months, an estimated 45.1% (95% CI 44.4%-45.7%) had a documented LVEF >35%. Counts of GDMT classes prescribed as well as ICD/CRT-D device therapy were associated with lower mortality risk in this population, even after adjustment for patient age, sex, and comorbidities.Both GDMT classes prescribed and device therapy were independently associated with lower mortality risk, even in the presence of more GDMT options for this more contemporary population.

View details for DOI 10.1016/j.hroo.2024.01.004

View details for PubMedID 38560378

View details for PubMedCentralID PMC10980920

Abstract

Large language models (LLMs) have recently shown impressive zero-shot capabilities, whereby they can use auxiliary data, without the availability of task-specific training examples, to complete a variety of natural language tasks, such as summarization, dialogue generation, and question answering. However, despite many promising applications of LLMs in clinical medicine, adoption of these models has been limited by their tendency to generate incorrect and sometimes even harmful statements.We tasked a panel of eight board-certified clinicians and two health care practitioners with evaluating Almanac, an LLM framework augmented with retrieval capabilities from curated medical resources for medical guideline and treatment recommendations. The panel compared responses from Almanac and standard LLMs (ChatGPT-4, Bing, and Bard) versus a novel data set of 314 clinical questions spanning nine medical specialties.Almanac showed a significant improvement in performance compared with the standard LLMs across axes of factuality, completeness, user preference, and adversarial safety.Our results show the potential for LLMs with access to domain-specific corpora to be effective in clinical decision-making. The findings also underscore the importance of carefully testing LLMs before deployment to mitigate their shortcomings. (Funded by the National Institutes of Health, National Heart, Lung, and Blood Institute.).

View details for DOI 10.1056/aioa2300068

View details for PubMedID 38343631

View details for PubMedCentralID PMC10857783

View details for Web of Science ID 001162343100065