RESUMO
BACKGROUND: Several SGLT2i (sodium-glucose transport protein 2 inhibitors) and GLP1-RA (glucagon-like peptide-1 receptor agonists) reduce cardiovascular events and improve kidney outcomes in patients with type 2 diabetes; however, utilization remains low despite guideline recommendations. METHODS: A randomized, remote implementation trial in the Mass General Brigham network enrolled patients with type 2 diabetes with increased cardiovascular or kidney risk. Patients eligible for, but not prescribed, SGLT2i or GLP1-RA were randomly assigned to simultaneous virtual patient education with concurrent prescription of SGLT2i or GLP1-RA (ie, Simultaneous) or 2 months of virtual education followed by medication prescription (ie, Education-First) delivered by a multidisciplinary team driven by nonlicensed navigators and clinical pharmacists who prescribed SGLT2i or GLP1-RA using a standardized treatment algorithm. The primary outcome was the proportion of patients with prescriptions for either SGLT2i or GLP1-RA by 6 months. RESULTS: Between March 2021 and December 2022, 200 patients were randomized. The mean age was 66.5 years; 36.5% were female, and 22.0% were non-White. Overall, 30.0% had cardiovascular disease, 5.0% had cerebrovascular disease, and 1.5% had both. Mean estimated glomerular filtration rate was 77.9 mL/(minâ§1.73 m2), and mean urine/albumin creatinine ratio was 88.6 mg/g. After 2 months, 69 of 200 (34.5%) patients received a new prescription for either SGLT2i or GLP1-RA: 53.4% of patients in the Simultaneous arm and 8.3% of patients in the Education-First arm (P<0.001). After 6 months, 128 of 200 (64.0%) received a new prescription: 69.8% of patients in the Simultaneous arm and 56.0% of patients in Education-First (P<0.001). Patient self-report of taking SGLT2i or GLP1-RA within 6 months of trial entry was similarly greater in the Simultaneous versus Education-First arm (69 of 116 [59.5%] versus 37 of 84 [44.0%]; P<0.001) Median time to first prescription was 24 (interquartile range [IQR], 13-50) versus 85 days (IQR, 65-106), respectively (P<0.001). CONCLUSIONS: In this randomized trial, a remote, team-based program identifies patients with type 2 diabetes and high cardiovascular or kidney risk, provides virtual education, prescribes SGLT2i or GLP1-RA, and improves guideline-directed medical therapy. These findings support greater utilization of virtual team-based approaches to optimize chronic disease management. REGISTRATION: URL: https://www.clinicaltrials.gov; Unique identifier: NCT06046560.
Assuntos
Diabetes Mellitus Tipo 2 , Inibidores do Transportador 2 de Sódio-Glicose , Humanos , Feminino , Masculino , Idoso , Diabetes Mellitus Tipo 2/tratamento farmacológico , Inibidores do Transportador 2 de Sódio-Glicose/uso terapêutico , Pessoa de Meia-Idade , Educação de Pacientes como Assunto , Receptor do Peptídeo Semelhante ao Glucagon 1/agonistas , Hipoglicemiantes/uso terapêutico , Guias de Prática Clínica como Assunto , Doenças Cardiovasculares , Telemedicina , Fidelidade a Diretrizes , Resultado do TratamentoRESUMO
Introduction: Image-based machine learning holds great promise for facilitating clinical care; however, the datasets often used for model training differ from the interventional clinical trial-based findings frequently used to inform treatment guidelines. Here, we draw on longitudinal imaging of psoriasis patients undergoing treatment in the Ultima 2 clinical trial (NCT02684357), including 2,700 body images with psoriasis area severity index (PASI) annotations by uniformly trained dermatologists. Methods: An image-processing workflow integrating clinical photos of multiple body regions into one model pipeline was developed, which we refer to as the "One-Step PASI" framework due to its simultaneous body detection, lesion detection, and lesion severity classification. Group-stratified cross-validation was performed with 145 deep convolutional neural network models combined in an ensemble learning architecture. Results: The highest-performing model demonstrated a mean absolute error of 3.3, Lin's concordance correlation coefficient of 0.86, and Pearson correlation coefficient of 0.90 across a wide range of PASI scores comprising disease classifications of clear skin, mild, and moderate-to-severe disease. Within-person, time-series analysis of model performance demonstrated that PASI predictions closely tracked the trajectory of physician scores from severe to clear skin without systematically over- or underestimating PASI scores or percent changes from baseline. Conclusion: This study demonstrates the potential of image processing and deep learning to translate otherwise inaccessible clinical trial data into accurate, extensible machine learning models to assess therapeutic efficacy.
RESUMO
Background: Subject screening is a key aspect of all clinical trials; however, traditionally, it is a labor-intensive and error-prone task, demanding significant time and resources. With the advent of large language models (LLMs) and related technologies, a paradigm shift in natural language processing capabilities offers a promising avenue for increasing both quality and efficiency of screening efforts. This study aimed to test the Retrieval-Augmented Generation (RAG) process enabled Generative Pretrained Transformer Version 4 (GPT-4) to accurately identify and report on inclusion and exclusion criteria for a clinical trial. Methods: The Co-Operative Program for Implementation of Optimal Therapy in Heart Failure (COPILOT-HF) trial aims to recruit patients with symptomatic heart failure. As part of the screening process, a list of potentially eligible patients is created through an electronic health record (EHR) query. Currently, structured data in the EHR can only be used to determine 5 out of 6 inclusion and 5 out of 17 exclusion criteria. Trained, but non-licensed, study staff complete manual chart review to determine patient eligibility and record their assessment of the inclusion and exclusion criteria. We obtained the structured assessments completed by the study staff and clinical notes for the past two years and developed a workflow of clinical note-based question answering system powered by RAG architecture and GPT-4 that we named RECTIFIER (RAG-Enabled Clinical Trial Infrastructure for Inclusion Exclusion Review). We used notes from 100 patients as a development dataset, 282 patients as a validation dataset, and 1894 patients as a test set. An expert clinician completed a blinded review of patients' charts to answer the eligibility questions and determine the "gold standard" answers. We calculated the sensitivity, specificity, accuracy, and Matthews correlation coefficient (MCC) for each question and screening method. We also performed bootstrapping to calculate the confidence intervals for each statistic. Results: Both RECTIFIER and study staff answers closely aligned with the expert clinician answers across criteria with accuracy ranging between 97.9% and 100% (MCC 0.837 and 1) for RECTIFIER and 91.7% and 100% (MCC 0.644 and 1) for study staff. RECTIFIER performed better than study staff to determine the inclusion criteria of "symptomatic heart failure" with an accuracy of 97.9% vs 91.7% and an MCC of 0.924 vs 0.721, respectively. Overall, the sensitivity and specificity of determining eligibility for the RECTIFIER was 92.3% (CI) and 93.9% (CI), and study staff was 90.1% (CI) and 83.6% (CI), respectively. Conclusion: GPT-4 based solutions have the potential to improve efficiency and reduce costs in clinical trial screening. When incorporating new tools such as RECTIFIER, it is important to consider the potential hazards of automating the screening process and set up appropriate mitigation strategies such as final clinician review before patient engagement.
RESUMO
AIM: Describe the rationale for and design of Diabetes Remote Intervention to improVe use of Evidence-based medications (DRIVE), a remote medication management program designed to initiate and titrate guideline-directed medical therapy (GDMT) in patients with type 2 diabetes (T2D) at elevated cardiovascular (CV) and/or kidney risk by leveraging non-physician providers. METHODS: An electronic health record based algorithm is used to identify patients with T2D and either established atherosclerotic CV disease (ASCVD), high risk for ASCVD, chronic kidney disease, and/or heart failure within our health system. Patients are invited to participate and randomly assigned to either simultaneous education and medication management, or a period of education prior to medication management. Patient navigators (trained, non-licensed staff) are the primary points of contact while a pharmacist or nurse practitioner reviews and authorizes each medication initiation and titration under an institution-approved collaborative drug therapy management protocol with supervision from a cardiologist and/or endocrinologist. Patient engagement is managed through software to support communication, automation, workflow, and standardization. CONCLUSION: We are testing a remote, navigator-driven, pharmacist-led, and physician-overseen management strategy to optimize GDMT for T2D as a population-level strategy to close the gap between guidelines and clinical practice for patients with T2D at elevated CV and/or kidney risk.