Development of secure infrastructure for advancing generative artificial intelligence research in healthcare at an academic medical center.

Background: Generative AI, particularly large language models (LLMs), holds great potential for improving patient care and operational efficiency in healthcare. However, the use of LLMs is complicated by regulatory concerns around data security and patient privacy. This study aimed to develop and evaluate a secure infrastructure that allows researchers to safely leverage LLMs in healthcare while ensuring HIPAA compliance and promoting equitable AI.

Development of Secure Infrastructure for Advancing Generative AI Research in Healthcare at an Academic Medical Center.

The increasing interest in leveraging generative AI models in healthcare necessitates secure infrastructure at academic medical centers. Without an all-encompassing secure system, researchers may create their own insecure microprocesses, risking the exposure of protected health information (PHI) to the public internet or its inadvertent incorporation into AI model training. To address these challenges, our institution implemented a secure pathway to the Azure OpenAI Service using our own private OpenAI instance which we fully control to facilitate high-throughput, secure LLM queries.

A Randomized Controlled Trial of Brief Intervention by Interactive Voice Response.

Aims: To determine the effect of an Interactive Voice Response (IVR) brief intervention (BI) to reduce alcohol consumption among adults seeking primary care.

Methods: Patients (N = 1855) with unhealthy drinking were recruited from eight academic internal medicine and family medicine clinics and randomized to IVR-BI (n = 938) versus No IVR-BI control (n = 917). Daily alcohol consumption was assessed at baseline, 3- and 6-months using the Timeline Followback.

A Randomized Controlled Trial of IVR-Based Alcohol Brief Intervention to Promote Patient-Provider Communication in Primary Care.

Background: Brief interventions for unhealthy drinking in primary care settings are efficacious, but underutilized. Efforts to improve rates of brief intervention though provider education and office systems redesign have had limited impact. Our novel brief intervention uses interactive voice response (IVR) to provide information and advice directly to unhealthy drinkers before a physician office visit, with the goals of stimulating in-office dialogue about drinking and decreasing unhealthy drinking.

Feasibility of automated pre-screening for lifestyle and behavioral health risk factors in primary care.

Background: Screening of primary care patients for unhealthy behaviors and mental health issues is recommended by numerous governing bodies internationally, yet evidence suggests that provider-initiated screening is not routine practice. The objective of this study was to implement systematic pre-screening of primary care patients for common preventive health issues on a large scale.

Methods: Patients registered for non-acute visits to one of 40 primary care providers from eight clinics in an Academic Medical Center health care network in the United States from May, 2012 to May, 2014 were contacted one- to three-days prior to their visit.