Elmer Victor Bernstam

Recent advances in artificial intelligence (AI) have fuelled a boom phase and promise to revolutionise many domains, including the traditional activities of academic medical centres (AMCs): biomedical research, education and healthcare services. Progress in AI is driven by three factors: (1) faster and cheaper processors (compute), (2) increasingly efficient training algorithms and neural network architectures and (3) an increase in electronically readable data. Outside of healthcare, industry investment dwarfs AI investment by governments. The largest computer clusters and datasets are now managed by industry. Similarly, most PhDs specialising in AI research are employed by industry. Therefore, it is not surprising that most of the largest and best-performing models are now developed by industry. Despite the industry lead in the general domain, we argue that AMCs have a key advantage in healthcare; they steward high-value biomedical data, a key driver of AI progress. Data are a strategic asset for AMCs. However, to capitalise on that asset, AMCs must invest in data ecosystems that improve data quality and utility, including the processes that produce data as well as data management. AMCs must also implement policies and procedures that lower the barriers to data use and reuse. With burgeoning interest in AI, AMCs should take on the role of careful, yet efficient, stewards of healthcare data to maximise its potential, while acting in the public’s best interests.

Background: Rural US populations bear a high burden of cardiovascular disease, with hypertension contributing to the rural–urban gap. We examined hypertension prevalence and control across the rural–urban continuum among patients in a large healthcare system in Northeast Texas. Methods: We analyzed the electronic health record data from 363,539 patients with at least one outpatient encounter between September 2021 and April 2025. We defined hypertension using ICD-10-CM codes, antihypertensive prescriptions, or a blood pressure ≥ 130/80 mm Hg. We classified rurality using the Rural–Urban Commuting Area categories. Multivariable logistic regression models were adjusted for age, sex, race, ethnicity, insurance, and level of engagement with the healthcare system. Results: Compared with metropolitan areas, micropolitan residents had lower odds of hypertension (OR = 0.85). We found uncontrolled hypertension also to be more common in metropolitan areas, with a lower prevalence observed across non-metropolitan categories, particularly among isolated rural residents (OR = 0.82). Black patients exhibited higher rates of uncontrolled hypertension across all rural–urban categories. Continuity of care was associated with hypertension prevalence but not control. Conclusions: Hypertension prevalence and control vary across the rural–urban continuum in Northeast Texas, with consistent differences observed between metropolitan and some non-metropolitan residents. Persistent racial differences across all geographic settings support the need for targeted prevention and management strategies.

Objectives: We report on findings from a meeting convened by the American College of Medical Informatics (ACMI) to characterize aspects of the patient experience that could be improved using informatics. Materials and Methods: The American College of Medical Informatics fellows were invited to share their experiences as patients and suggest informatics approaches that may improve the patient experience. Results: We identified 4 themes: (1) getting the right care, (2) data sharing and data interoperability, (3) guiding low-cost evaluations, and (4) predictive analytics. Discussion: Despite widespread adoption of health IT, patient experiences remain far from optimal. Conclusion: The American College of Medical Informatics fellows identified informatics approaches, applications, and research areas that have the potential to improve patient experiences with health care systems.

Objective: We present a set of definitions and a conceptual model to support primary and consulting physician schedule integration into the electronic health record (EHR) in an inpatient setting and show how utilization of this functionality supports patient-centered communication. Materials and Methods: Our institution transitioned to the Epic EHR and implemented modules to connect primary and consulting provider schedules from external scheduling systems to secure messaging within an inpatient EHR context. We evaluated legacy functionality, met with provider groups to map their shifts to hospital teams, built a crosswalk tool to extract, transform, and load data from the scheduling systems to the EHR, evaluated the utilization, and assessed issues related to the implementation. We used our experience from the project to develop a set of definitions and a conceptual model for provider scheduling. Results: We met with over 100 groups to map over 2000 shifts to nearly 700 teams across 15 facilities in our health system. Utilization was high with an average of 6500 on-call provider searches per day in the 30 days following implementation. The conceptual model for inpatient provider scheduling defines 11 terms. Discussion: Our definitions and conceptual model sufficiently represent the inpatient provider scheduling domain as evidenced by high initial utilization and few reported defects. The standardized terminology for provider scheduling aids integration of scheduling data into the EHR. Conclusion: Our successful integration of real-time scheduling data within the EHR guided development of a provider scheduling conceptual model. Standardized provider scheduling terminology promotes interoperability of scheduling systems.

Background: Delirium is often underdiagnosed in clinical practice and is not routinely coded for billing. While manual chart review can identify delirium, it is labor-intensive and impractical for large-scale studies. Natural language processing (NLP) can analyze unstructured text in electronic health records (EHRs) to extract meaningful clinical information. Methods: To support national integration of NLP for EHR-based delirium identification across different institutions, we launched the Delirium Interest Group within the national Evolve to Next-Gen Accrual to Clinical Trials (ENACT) NLP Working Group. This paper outlines our initial efforts to standardize, evaluate, and translate an NLP-based delirium detection model into the i2b2/ENACT platform. Results: Multisite contextual inquiry identified several key challenges, including variations in local screening practices (eg, tools used, documentation frequency, and quality control), the need for harmonized definitions in the context of EHRs, and the complexity of modeling temporal logic. Multisite NLP evaluation revealed variable performance degradation driven by differences in delirium screening practices, clinical documentation patterns and semantics, and note syntactic structures. Conclusion: Our work represents an important first step toward enabling scalable and standardized NLP-based delirium detection across institutions. By engaging diverse institutions through the ENACT NLP Working Group, we identified shared challenges and site-specific variations that impact model implementation and performance. Our collaborative approach enabled the development of a more robust framework for delirium identification across heterogeneous EHR systems. Future efforts will build on this foundation to enhance the validity, usability, and translational impact of delirium detection.

Background: Effective blood pressure (BP) management is essential for several disease states, including secondary stroke prevention. Remote BP monitoring used in combination with an electronic data capture (EDC) system offers potential for improved BP control for patients. While such systems currently exist, there are many patient-, provider-, and system-specific challenges these systems currently do not address. In addition, such systems are rarely described transparently and comprehensively in the literature with enough detail to be reproduced by other clinicians. Objective: This study aims to describe the development and design of a web-based EDC system tailored for remote BP monitoring and management to overcome common patient-, provider-, and system-specific challenges. Methods: The EDC system was developed for use in the Video-based Intervention to Reduce Treatment and Outcome Disparities in Adults living with Stroke (VIRTUAL) and transient ischemic attack clinical trial, a randomized study involving adult stroke survivors receiving care for secondary stroke prevention. The EDC system supports remote monitoring using a BP monitor, which was specifically and methodically chosen for its suitability to the population. The EDC system integrates the data from the remote monitors into a web-based application, facilitating real-time monitoring for health care team intervention. The system includes a BP dashboard displaying time-series data, visual alerts, and email notifications for out-of-range readings. Proxy measures assessed system effectiveness in addressing patient-, provider-, and system-specific needs. Results: Seventeen BP monitors were evaluated, and the BlipcareBlip BP800 was selected based on its cellular connectivity, ergonomic design, and application programming interface compatibility. The EDC system successfully linked patient data to the dashboard, enabling real-time visualization of BP trends and provider alerts. A total of 97% (184/190) of participants in the intervention arm engaged in BP monitoring, averaging 1.2 readings per day. Clinicians actively used the system, with mean monthly logins ranging from 19 to 44 per provider. Alerts were triggered for 69% ( of participants, primarily for systolic BP abnormalities. Data transmission delays were minimal, with a median time of 1.4 minutes from data collection to dashboard entry. Outlier alerts were efficiently processed, with an average notification time of 17.9 hours. The system demonstrated high functionality in addressing patient-, provider-, and system-specific needs, supporting effective BP monitoring and intervention delivery. Conclusions: The VIRTUAL EDC system facilitates remote monitoring of BP in stroke survivors, offering a model for similar systems in clinical trials. Clinicians can analyze, track, and interpret monitored data in near real-time, remotely, which may enhance BP management approaches, particularly in communities with limited access to care. This approach has the potential to improve hypertension management for patients and clinicians when standard data transparency is insufficient, standard monitor features are incompatible with the user, or when integrating a commercial dashboard may threaten electronic health record security.

Objectives Duplicate patient records can increase costs and medical errors. We assessed the association between demographic factors, comorbidities, health care usage, and duplicate electronic health records. Methods We analyzed the association between duplicate patient records and multiple demographic variables (race, Hispanic ethnicity, sex, and age) as well as the Charlson Comorbidity Index (CCI), number of diagnoses, and number of health care encounters. The study population included 3,018,413 patients seen at a large urban academic medical center with at least one recorded diagnosis. Duplication of patient medical records was determined by using a previously validated enterprise Master Person Index. Results Unknown or missing demographic data, Black race when compared with White race (odds ratio [OR]: 1.35, p < 0.001), Hispanic compared with non-Hispanic ethnicity (OR: 1.48, p < 0.001), older age (OR: 1.01, p < 0.001), and “Other” sex compared with female sex (OR: 4.71, p < 0.001) were associated with higher odds of having a duplicate record. Comorbidities (CCI, OR: 1.10, p < 0.001) and more encounters with the health care system (OR: 1.01, p < 0.001) were also associated with higher odds of having a duplicate record. In contrast, male sex compared with female sex was associated with lower odds of having a duplicate record (OR: 0.88, p < 0.001). Conclusion The odds of duplications in medical records were higher in Black, Hispanic, older, nonmale patients with more health care encounters, more comorbidities, and unknown demographic data. Understanding the epidemiology of duplicate records can help guide prevention and mitigation efforts for high-risk populations. Duplicate records can contribute to disparities in health care outcomes for minority populations.