Using large language model to guide patients to create efficient and comprehensive clinical care message.

OBJECTIVE: This study aims to investigate the feasibility of using Large Language Models (LLMs) to engage with patients at the time they are drafting a question to their healthcare providers, and generate pertinent follow-up questions that the patient can answer before sending their message, with the goal of ensuring that their healthcare provider receives all the information they need to safely and accurately answer the patient's question, eliminating back-and-forth messaging, and the associated delays and frustrations. METHODS: We collected a dataset of patient messages sent between January 1, 2022 to March 7, 2023 at Vanderbilt University Medical Center. Two internal medicine physicians identified 7 common scenarios. We used 3 LLMs to generate follow-up questions: (1) Comprehensive LLM Artificial Intelligence Responder (CLAIR): a locally fine-tuned LLM, (2) GPT4 with a simple prompt, and (3) GPT4 with a complex prompt. Five physicians rated them with the actual follow-ups written by healthcare providers on clarity, completeness, conciseness, and utility. RESULTS: For five scenarios, our CLAIR model had the best performance. The GPT4 model received higher scores for utility and completeness but lower scores for clarity and conciseness. CLAIR generated follow-up questions with similar clarity and conciseness as the actual follow-ups written by healthcare providers, with higher utility than healthcare providers and GPT4, and lower completeness than GPT4, but better than healthcare providers. CONCLUSION: LLMs can generate follow-up patient messages designed to clarify a medical question that compares favorably to those generated by healthcare providers.

Vanderbilt Electronic Health Record Voice Assistant Supports Clinicians.

BACKGROUND: Electronic health records (EHRs) present navigation challenges due to time-consuming searches across segmented data. Voice assistants can improve clinical workflows by allowing natural language queries and contextually aware navigation of the EHR. OBJECTIVES: To develop a voice-mediated EHR assistant and interview providers to inform its future refinement. METHODS: The Vanderbilt EHR Voice Assistant (VEVA) was developed as a responsive web application and designed to accept voice inputs and execute the appropriate EHR commands. Fourteen providers from Vanderbilt Medical Center were recruited to participate in interactions with VEVA and to share their experience with the technology. The purpose was to evaluate VEVA's overall usability, gather qualitative feedback, and detail suggestions for enhancing its performance. RESULTS: VEVA's mean system usability scale score was 81 based on the 14 providers' evaluations, which was above the standard 50th percentile score of 68. For all five summaries evaluated (overview summary, A1C results, blood pressure, weight, and health maintenance), most providers offered a positive review of VEVA. Several providers suggested modifications to make the technology more useful in their practice, ranging from summarizing current medications to changing VEVA's speech rate. Eight of the providers (64%) reported they would be willing to use VEVA in its current form. CONCLUSION: Our EHR voice assistant technology was deemed usable by most providers. With further improvements, voice assistant tools such as VEVA have the potential to improve workflows and serve as a useful adjunct tool in health care.

Before-Visit Questionnaire: A Tool to Augment Communication and Decrease Provider Documentation Burden in Pediatric Diabetes.

OBJECTIVE: To develop and evaluate an electronic tool that collects interval history and incorporates it into a provider summary note. METHODS: A parent-facing online before-visit questionnaire (BVQ) collected information from parents and caregivers of pediatric diabetes patients prior to a clinic encounter. This information was related to interval history and perceived self-management barriers. The BVQ generated a summary note that providers could paste in their own documentation. Parents also completed postvisit experience questionnaires. We assessed the BVQs perceived usefulness to parents and providers and compared provider documentation content and length pre- and post-BVQ rollout. We interviewed providers regarding their experiences with the system-generated note. RESULTS: Seventy-three parents of diabetic children were recruited and completed the BVQ. A total of 79% of parents stated that the BVQ helped with visit preparation and 80% said it improved perceived quality of visits. All 16 participating providers reviewed BVQs prior to patient encounters and 100% considered the summary beneficial. Most providers (81%) desired summaries less than 1 week old. A total of 69% of providers preferred the prose version of the summary; however, 75% also viewed the bulleted version as preferable for provider review. Analysis of provider notes revealed that BVQs increased provider documentation of patients' adherence and barriers. We observed a 50% reduction in typing by providers to document interval histories. Providers not using summaries typed an average of 137 words (standard deviation [SD]: 74) to document interval history compared with 68 words [SD 47] typed with BVQ use. DISCUSSION: Providers and parents of children with diabetes appreciated the use of previsit, parent-completed BVQs that automatically produced provider documentation. Despite the BVQ redistributing work from providers to parents, its use was acceptable to both groups. CONCLUSION: Parent-completed questionnaires on the patient's behalf that generate provider documentation encourage communication between parents and providers regarding disease management and reduce provider workload.

Electronic Health Record Interactions through Voice: A Review.

BACKGROUND: Usability problems in the electronic health record (EHR) lead to workflow inefficiencies when navigating charts and entering or retrieving data using standard keyboard and mouse interfaces. Voice input technology has been used to overcome some of the challenges associated with conventional interfaces and continues to evolve as a promising way to interact with the EHR. OBJECTIVE: This article reviews the literature and evidence on voice input technology used to facilitate work in the EHR. It also reviews the benefits and challenges of implementation and use of voice technologies, and discusses emerging opportunities with voice assistant technology. METHODS: We performed a systematic review of the literature to identify articles that discuss the use of voice technology to facilitate health care work. We searched MEDLINE and the Google search engine to identify relevant articles. We evaluated articles that discussed the strengths and limitations of voice technology to facilitate health care work. Consumer articles from leading technology publications addressing emerging use of voice assistants were reviewed to ascertain functionalities in existing consumer applications. RESULTS: Using a MEDLINE search, we identified 683 articles that were reviewed for inclusion eligibility. The references of included articles were also reviewed. Sixty-one papers that discussed the use of voice tools in health care were included, of which 32 detailed the use of voice technologies in production environments. Articles were organized into three domains: Voice for (1) documentation, (2) commands, and (3) interactive response and navigation for patients. Of 31 articles that discussed usability attributes of consumer voice assistant technology, 12 were included in the review. CONCLUSION: We highlight the successes and challenges of voice input technologies in health care and discuss opportunities to incorporate emerging voice assistant technologies used in the consumer domain.

Breadcrumbs: Assessing the Feasibility of Automating Provider Documentation Using Electronic Health Record Activity.

This study assessed the feasibility of automating the generation of the outpatient encounter summary. We reviewed screen tracking video and log-file metadata from electronic health record (EHR) interactions based on two simulated encounters. We mapped the sequence of metadata to key concepts in the video to assess the precision with which the log files aligned with user activity and to generate the Breadcrumbs encounter summary (BES). The BES captured all interactions documented in clinical notes with the exception of the physical exam. The videos addressed all Evaluation and Management (E/M) requirements, while the log files did not contain the physical exam. The BES was as comprehensive as the gold standard visit summary. The BES offers a promising method for the collection and compilation of necessary elements of outpatient clinical documentation. The combination of log files and video could provide evidence of EHR activity satisfying documentation requirements.

Use of Commonly Available Technologies for Diabetes Information and Self-Management Among Adolescents With Type 1 Diabetes and Their Parents: A Web-Based Survey Study.

BACKGROUND: For individuals with Type 1 diabetes (T1D), following a complicated daily medical regimen is critical to maintaining optimal health. Adolescents in particular struggle with regimen adherence. Commonly available technologies (eg, diabetes websites, apps) can provide diabetes-related support, yet little is known about how many adolescents with T1D use them, why they are used, or relationships between use and self-management. OBJECTIVE: This study examined adolescent and parent use of 5 commonly available technologies for diabetes, including proportions who use each technology, frequency of use, and number of different technologies used for diabetes. Analyses also investigated the reasons adolescents reported for using or not using technologies for diabetes, and factors correlated with adolescents' technology use. Finally, this study examined relationships between the type and number of technologies adolescents use for diabetes and their self-management and glycemic control. METHODS: Adolescents (12-17 years) and their parents (N=174 pairs), recruited from a pediatric diabetes clinic (n=134) and the Children with Diabetes community website (n=40), participated in this Web-based survey study. Glycosylated hemoglobin (A1C) values were obtained from medical records for pediatric clinic patients. Adolescents reported their use of 5 commonly available technologies for diabetes (ie, social networking, diabetes websites, mobile diabetes apps, text messaging, and glucometer/insulin pump software), reasons for use, and self-management behavior (Self-Care Inventory-Revised, SCI-R). RESULTS: Most adolescents and parents used at least one of the 5 technologies for diabetes. Among adolescents, the most commonly used technology for diabetes was text messaging (53%), and the least commonly used was diabetes websites (25%). Most adolescents who used diabetes apps, text messaging, or pump/glucometer software did so more frequently (≥2 times per week), compared to social networking and website use (≤1 time per week). The demographic, clinical, and parent-technology use factors related to adolescents' technology use varied by technology. Adolescents who used social networking, websites, or pump/glucometer software for diabetes had better self-management behavior (SCI-R scores: beta=.18, P=.02; beta=.15, P=.046; beta=.15, P=.04, respectively), as did those who used several technologies for diabetes (beta=.23, P=.003). However, use of diabetes websites was related to poorer glycemic control (A1C: beta=.18, P=.01). CONCLUSIONS: Adolescents with T1D may be drawn to different technologies for different purposes, as individual technologies likely offer differing forms of support for diabetes self-management (eg, tracking blood glucose or aiding problem solving). Findings suggest that technologies that are especially useful for adolescents' diabetes problem solving may be particularly beneficial for their self-management. Additional research should examine relationships between the nature of technology use and adolescents' T1D self-management over time.

Technology Use for Diabetes Problem Solving in Adolescents with Type 1 Diabetes: Relationship to Glycemic Control.

BACKGROUND: This study examines technology use for problem solving in diabetes and its relationship to hemoglobin A1C (A1C). SUBJECTS AND METHODS: A sample of 112 adolescents with type 1 diabetes completed measures assessing use of technologies for diabetes problem solving, including mobile applications, social technologies, and glucose software. Hierarchical regression was performed to identify the contribution of a new nine-item Technology Use for Problem Solving in Type 1 Diabetes (TUPS) scale to A1C, considering known clinical contributors to A1C. RESULTS: Mean age for the sample was 14.5 (SD 1.7) years, mean A1C was 8.9% (SD 1.8%), 50% were female, and diabetes duration was 5.5 (SD 3.5) years. Cronbach's α reliability for TUPS was 0.78. In regression analyses, variables significantly associated with A1C were the socioeconomic status (ß = -0.26, P < 0.01), Diabetes Adolescent Problem Solving Questionnaire (ß = -0.26, P = 0.01), and TUPS (ß = 0.26, P = 0.01). Aside from the Diabetes Self-Care Inventory--Revised, each block added significantly to the model R(2). The final model R(2) was 0.22 for modeling A1C (P < 0.001). CONCLUSIONS: Results indicate a counterintuitive relationship between higher use of technologies for problem solving and higher A1C. Adolescents with poorer glycemic control may use technology in a reactive, as opposed to preventive, manner. Better understanding of the nature of technology use for self-management over time is needed to guide the development of technology-mediated problem solving tools for youth with type 1 diabetes.

Mitigating micro-and macro-vascular complications of diabetes beginning in adolescence.

Diabetes is a chronic disorder, which manifests when insulin levels or resistance to insulin action becomes insufficient to control systemic glucose levels. Although the number of available agents to manage diabetes continues to expand rapidly, the maintenance of euglycemia by individuals with diabetes remains a substantial challenge. Unfortunately, many patients with type 1 and type 2 diabetes will ultimately experience diabetes complications. These complications result from the toxic effects of chronic hyperglycemia combined with other metabolic derangements that afflict persons with diabetes. This review will present a comprehensive look at the complications of diabetes, the risk factors for their progression, the mechanistic basis for their development, and the clinical approach to screening for, preventing, and treating these sequelae. In addition, since diabetes is commonly diagnosed in childhood, we will provide a special focus on the care of the adolescent patient.