Grand Rounds Archive

Grand Rounds April 17, 2026: Keeping a Human in the Loop: Scientific Publishing and AI (Roy Perlis, MD, MSc)

Posted on by Sophia Ramirez

Speaker

Roy Perlis, MD, MSc
Editor in Chief, JAMA+ AI
Director, MGH Center for Quantitative Health
Vice Chair for Research, Mass General-Brigham Psychiatry
Professor of Psychiatry, Harvard Medical School

Keywords

Artificial Intelligence; Publishing; Medical Publishing; Peer Review; Trust

Key Points

  • Though medical publishing was in flux before artificial intelligence (AI) became ubiquitous, the contemporary state of the industry is defined by AI-driven trends: an increased volume of submissions, an increased volume of low-quality submissions, and increased complexity of submissions. Academics are concurrently less willing to participate in the peer-review process – paradoxically making peer review into an area ripe for AI intervention.
  • Scientific communities are divided about the appropriate role of AI in manuscript preparation and review. In mid-2023, JAMA Network journals began requiring authors and peer reviewers to answer questions about their use of AI to create or assist with creation or editing of submitted manuscripts, or with preparation of reviews. While author disclosure of AI use is low, those figures likely represent an underestimate.
  • That peer review improves manuscript quality is a fringe benefit; its real value is instilling trust in the contents. The age of AI poses twin dangers: the simulation and politicization of expertise. AI may be capable of handling rote tasks like checking protocols, human subjects approval, and reporting checklists, but it comes with a potential cost in terms of the public’s already-eroded faith in scientific expertise.

Discussion Themes

The most commonly disclosed use of AI in manuscripts is for language and grammar cleanup, particularly by non-native English speakers. On the whole, AI has improved the clarity of international submissions.

Perlis predicted that administrative editorial tasks will be automated but human editors will remain essential for scientific taste: the ability to curate impactful stories and select which of hundreds of AI-generated hypotheses are worth pursuing.

Researchers were encouraged to master the use of AI agents as local analytic partners or writing assistants to automate aspects of their workflow without losing human oversight.

Grand Rounds April 10, 2026: Impact of Behavioral Science-Based Electronic Health Record Tools on Deprescribing for Older Adults (Julie Lauffenburger, PharmD, PhD)

Posted on by terren.green@duke.edu

Speaker

Julie Lauffenburger, PharmD, PhD
Associate Professor of Medicine
Brigham and Women’s Hospital and Harvard Medical School

Keywords

Adaptive trial design; Behavioral science; Deprescribing; electronic health record; EHR; Inappropriate prescribing; NUDGE-EHR; Overprescribing.

Key Points

  • Older adults are often overprescribed medications or prescribed potentially inappropriate medications like benzodiazepines, non-benzodiazepine sedative hypnotics, or strongly anticholinergic medications with long-term use associated with a 30% increased risk of hospitalizations and falls.
  • Medication management or optimization in older adults is often difficult due to a tendency to maintain the status quo, time constraints, patient preference, or diffusion of responsibility, and existing interventions for medication management are highly resource intensive.
  • Behavioral science techniques employed in the NUDGE-EHR and NUDGE-HER-2 trials may enhance the effectiveness of electronic health record (EHR) tools to alert clinicians to inappropriate medications during patient visits.
  • NUDGE-EHR was a 16 arm two-stage adaptive pragmatic trial among 216 primary care providers and older adult patients conducted from October 2020 to August 2022 examining 14 promising EHR tools using 9 different behavioral principles with deprescribing as a primary outcome.
  • The 2 most promising tools were included in the second 3 parallel arm pragmatic trial, NUDGE-EHR-2, in a different health system from November 2022 to March 2024. EHR tools used pop-up windows to suggest deprescribing. The study provided a set of helpful options to providers including a tapering algorithm, instructions for patients, orders for alternative medications, and referrals to behavioral health providers to make this process faster and easier.
  • Deprescribing increased by 6.5% to 10.4% over usual care. Active discontinuation by primary care providers appeared to drive the results.

Discussion Themes

The adaptive trial design of the first NUDGE-EHR study helped inform the more traditional confirmation trial NUDGE-EHR-2.

The way EHR tools are used varies widely from provider to provider. Tools may be adapted over time so the tool works best for the individual provider.

 

Read more about the NUDGE-EHR study.

 

Grand Rounds April 3, 2026: AI for Diabetes Prevention (Nestoras Mathioudakis, MD, MHS)

Posted on by Sophia Ramirez

Speaker

Nestoras Mathioudakis, MD, MHS
Associate Professor of Medicine
Johns Hopkins University School of Medicine

Keywords

Artificial Intelligence; Diabetes; Prevention; Coaching; Automation

Key Points

  • Of the nearly 100 million U.S. adults with prediabetes, approximately 70% will progress to type 2 diabetes in their lifetime. The Diabetes Prevention Program (DPP), a gold-standard program focused on lifestyle interventions, has demonstrated a 58% reduction in diabetes incidence. However, an effort to implement the program nationally fails to reach 99% of eligible individuals.
  • The research team sought to investigate whether a fully automated, AI-based DPP could effectively replicate the outcomes of the human coach-based DPP and potentially bridge this access gap. This was the first trial comparing a fully automated versus human DPP. It adds to a limited evidence base evaluating AI interventions against established standards in medicine.
  • The research team found that the AI-driven DPP delivered without human intervention was non-inferior to the traditional human coach-based DPP. Participants in the AI-driven DPP arm had comparable health outcomes and adequate engagement – though they were less likely to express a preference for their intervention than those in the human coach-based arm. The study team concluded that diabetes prevention remains an implementation challenge, not an efficacy problem.

Discussion Themes

While absolute weight loss was modest when compared to the effect of new medications like GLP-1s, Dr. Dakis argued that lifestyle interventions remain more cost-effective and that future automation efforts may bridge the effectiveness gap.

In the future, large language models could bolster the trust and human connection lacking in fully automated digital interventions.

Grand Rounds March 27, 2026: Text Messaging and Video Stories to Support Hypertension Self-Management in Black Veterans: A Randomized Clinical Trial (Sarah L. Cutrona, MD, MPH)

Posted on by Sophia Ramirez

Speaker

Sarah L. Cutrona, MD, MPH
Acting Director, Center for Health Optimization and Implementation Research (CHOIR)
VA Bedford Healthcare System
Professor, Division of Health Informatics and Implementation Science
Department of Population and Quantitative Health Sciences
UMass Chan Medical School

Keywords

Blood Pressure; Hypertension; Self-Management; Video; Texting; Black Americans; Veterans

Key Points

  • Black Americans experience disproportionate morbidity and mortality due to hypertension (HTN). Simultaneously, self-management is made more difficult by differential rates of diagnosis and treatment titration; decreased access to, trust in, and engagement with the healthcare system; and cost, access, and environmental barriers. This study sought to improve HTN control and self-management among about 600 Black Veterans by supplementing a preexisting video story intervention with longitudinal texting support.
  • Those in the intervention arm watched 5 video stories featuring other Black Veterans, chose their preferred storyteller, and received educational and interactive messages aligned with that storyteller. Those in the control arm received only interactive messages. The researchers hypothesized that peer stories would promote participants’ emotional engagement with the messages via a parasocial relationship with the storyteller, enhancing self-efficacy, influencing health behaviors and ultimately improving HTN control.
  • While participants in both arms saw marginal improvements in systolic and diastolic blood pressure (BP), there was no significant difference in the BP change between the arms. The intervention saw high treatment fidelity (92%) and sustained engagement, with over 55% of participants responding to texts through the end of the 6-month period. The researchers concluded that the interactive text messages, which were present in both arms, were an effective way to maintain engagement in a multi-month study and may serve as a useful strategy for future longitudinal interventions supporting Black Veterans.

Discussion Themes

The control group was intentionally kept active – i.e., received educational texts – to isolate the specific impact of storytelling rather than just the impact of receiving text messages.

The intervention was shaped by direct feedback from veteran consultants, who provided guidance on tone, word choice, and structure to ensure the messages felt authentic.

Attendees noted that negative studies are vital for informing future research and that the intervention provided valuable moments of outreach to a population facing socioeconomic vulnerabilities and isolation.

Grand Rounds March 20, 2026: FLUID: A Cross-Over Trial of Hospital Wide Ringer’s Lactate vs Normal Saline (Lauralyn McIntyre, MD, FRCPC, MSc; Monica Taljaard, PhD)

Posted on by Sophia Ramirez

Speakers

Lauralyn McIntyre, MD, FRCPC, MSc
Senior Scientist, Acute Care Research Program
Ottawa Hospital Research Institute
Professor, University of Ottawa

Monica Taljaard, PhD
Senior Scientist
Methodological and Implementation Research
Ottawa Hospital Research Institute
Full Professor, School of Epidemiology and Public Health
University of Ottawa

Keywords

Fluids; Crystalloid fluids; Mortality; Hospital Readmission; Health Systems; Hospitals; Ringer’s Lactate; Saline

Key Points

  • Crystalloid fluids are among the most common interventions administered to patients; they touch nearly every person admitted to the hospital. However, the evidence base to inform their use is limited. Few trials were done until this past decade, and recent trials tend to focus on the critically ill. Though those studies found very small clinical outcome differences between fluids, even a 0.5% reduction in death and hospital readmission translates to 2,500 lives and $5 million saved in Ottawa alone.
  • The study team sought to address this study question at the hospital/health system level by comparing the effects of two usual care fluids, normal saline and Ringer’s Lactate, on the death or readmission of all hospitalized patients. After a pilot trial indicated feasibility, they launched FLUID: a cluster-randomized, cross-over trial across four Ontario hospitals.
  • The FLUID trial found a 0.5% reduction in death in the patients that received Ringer’s Lactate. This is a small but clinically meaningful reduction, with major implications for mortality at the hospital and health care system level. The findings are limited by the study’s early termination due to the COVID-19 pandemic.

Discussion Themes

While there are theoretical benefits to multiple crossovers, including the mitigation of period effects, the logistical burden of restocking an entire hospital’s fluid supply more than once was insurmountable.

The study team used a composite endpoint that gave hospital readmission the same weight as death. McIntyre attributed this decision to the significance of readmission for their patient partners and to its status as an indicator of resource use and subsequent mortality risk.

Trials disrupted by crises such as a pandemic or supply chain issues may be salvaged by Bayesian sequential testing, which allows researchers to assess data continuously without alpha penalties.

FLUID exemplifies the importance of asking simple, high-impact research questions. It was also incredibly cost-effective, costing less than $400,000 (in Canadian dollars) to conduct.

Grand Rounds March 13, 2026: Behavioral Economics and Medication Adherence for Hypertension: A Randomized Clinical Trial (John A. Dodson, MD, MPH)

Posted on by Sophia Ramirez

Speaker

John A. Dodson, MD, MPH
Associate Professor of Medicine and Population Health
Director, Geriatric Cardiology Program
Director, Cardiovascular Digital Health Laboratory
NYU Langone Health
NYU Grossman School of Medicine

Keywords

Behavioral Economics; Cardiovascular Disease; Hypertension; Medication Adherence

Key Points

  • Roughly half of patients with cardiovascular disease (CVD) are nonadherent with their medications. This occurs across conditions and for multiple reasons, including cost, side effects, lack of symptoms, and inconvenience. BETTER-BP, a phase 2, multisite trial, sought to test the effect of a lottery on antihypertensive adherence. The lottery intervention was based in behavioral economics, a field that represents a novel and potentially scalable approach to improving medication adherence.
  • In a safety-net population, the lottery doubled adequate antihypertensive medication adherence from baseline to 6 months. This did not translate to a significant reduction in office-measured systolic blood pressure (BP) and increased adherence was not sustained after the lottery was removed. Other strategies will likely be required for long-term behavior change.
  • BETTER-BP had pragmatic components and non-pragmatic components. It was pragmatic in that it utilized minimal exclusion criteria; took place in a real-world setting; used existing medications; paired study visits with regular ambulatory visits; and used the electronic health record to ascertain some measures. It was not pragmatic in that it utilized traditional informed consent; the intervention and monitoring strategies were not usual practice; and the primary outcome was measured via an in-person BP assessment.

Discussion Themes

Dr. Dodson noted that many trials of behavioral economics in CVD medication adherence have not had durable effects on either adherence or clinical outcomes. Even with trials that are positive, no intervention has been a “home run.”

The lottery payouts (ranging from $5 to $50) were based on an estimated daily value intended to influence behavior without being coercive or exceeding study budgets.

Recruitment for the trial was a challenge; the team made approximately 9,000 phone calls to reach their final sample of 400 participants.

Behavioral incentives might be better suited for time-limited interventions, such as smoking cessation, rather than the lifelong management required for conditions like hypertension.

Grand Rounds March 6, 2026: Bringing Primary Care Clinical Trials Research Into the 21st Century: Lessons Learned and Developments From Large-Scale European Adaptive Platform Trials of Therapeutics for Acute Respiratory Infections (Christopher C. Butler, BA, MBChB, DCH, CCH, MD, FRCGP)

Posted on by Sophia Ramirez

Speaker

Christopher C. Butler, BA, MBChB, DCH, CCH, MD, FRCGP
Associate Head for Research
Professor of Primary Care
Professorial Fellow, Trinity College
Clinical Director, Primary Care Clinical Trials Unit
University of Oxford

Keywords

Primary Care; COVID-19; Respiratory Medicine; Adaptive Platform Trials; Large-Scale Trials; Pandemic; General Practitioners

Key Points

  • Effective primary care (PC) interventions have the potential to reach a significant portion of the community and, concurrently, have a significant impact. PC is also a good vehicle for early, self-initiated interventions, which can limit illness and enhance the sustainability of health care. However, traditional clinical trials face several challenges in PC, including general practitioners’ workload crisis, opportunistic recruitment, and geographical constraints.
  • The research team sought to address these challenges by conducting adaptive platform trials (APTs) that utilized sophisticated digital methods in a PC setting. PRINCIPLE and PANORAMIC evaluated multiple COVID-19 treatments simultaneously using innovative methods like remote eligibility checks, e-consent, and direct-to-participant drug delivery. The use of APTs demonstrated that large-scale, rigorous research can be successfully conducted outside of hospitals and provide real-world evidence.
  • Next, the research team will conduct the ECRAID-Prime trial: An international community-based APT evaluating early treatments for acute respiratory infections to prevent hospitalizations and reduce societal illness duration. They are currently in the recruitment phase, having shifted from traditional site-based recruitment to a more decentralized approach.

Discussion Themes

The National Health Service (NHS) and publicly funded clinical trials units are mission-critical because they prioritize health outcomes over profit and provide a “warm base” for rapid research.

Dr. Butler noted that busy general practitioners participated because they bought into the values and the importance of the research questions, especially when administrative burdens were minimized.

Grand Rounds February 27, 2026: Optimal Transfusion Strategy in High Cardiac Risk Patients After Major Surgery: Results of the TOP Trial (Panos Kougias, MD, MSc; Sherene Sharath, PhD, MPH)

Posted on by Sophia Ramirez

Speakers

Panos Kougias, MD, MSc
Professor and Chair
Department of Surgery, SUNY Downstate Health Sciences University

Sherene Sharath, PhD, MPH
Assistant Professor
Department of Surgery, SUNY Downstate Health Sciences University

Keywords

Blood Transfusion; Surgery; Cardiology; Cardiac Outcomes

Key Points

  • Postoperative anemia is common after major operations. While the current guidelines recommend transfusion for hemoglobin (Hb) below 7 g/dL, the safety of this restrictive strategy is unclear – particularly after major operations and in high cardiac risk patients.
  • The study team sought to evaluate whether a more liberal transfusion strategy reduces the risk of mortality or major ischemic events in high cardiac risk patients undergoing major operations. They randomized eligible veterans from 16 VA medical centers to receive transfusion at either Hb<10 g/dL (the “liberal” arm) or Hb<7 g/dL (the “restrictive” arm).
  • They found that the rates of mortality and major ischemic events were similar between the groups. While the rates of most secondary outcomes were also similar, a composite outcome of heart failure and arrhythmias was more frequent in the restrictive arm.
  • The study team noted that a one-size-fits-all approach may not be applicable for transfusion strategies. Potential cardiac-outcome related benefits should be explored in future trials.

Discussion Themes

In the weighting of composite outcomes, the analysis treated all events (e.g., death vs. minor complications) as having equal clinical severity. This may have reduced the trial’s statistical power.

The actual event rate (10%) was much lower than the anticipated 30%. The study team attributed this to improved perioperative care and better cardiovascular outcomes over the last decade compared to the historical data used for initial calculations.

Communicating nuanced findings may require some extra legwork; clinicians often struggle to interpret results that are not clear-cut “wins” or “losses.”

Grand Rounds February 20, 2026: Applying Mind-Body Skills for Pain Using a Whole Health Telehealth Intervention (RAMP): Early Progress and Lessons Learned (Diana Burgess, PhD; Roni Evans, DC, MS, PhD; Katie Hadlandsmyth, PhD)

Posted on by Sophia Ramirez

Speakers

Diana Burgess, PhD
Professor of Medicine,
University of Minnesota
Director, VA Advanced Fellowship Program in Health Systems Research, Center for Care Delivery and Outcomes Research (CCDOR),
Minneapolis Veterans Affairs Healthcare System
Director, QUERI Complementary and Integrative Health Evaluation Center (CIHEC),
Veterans Affairs Healthcare System

Roni Evans, DC, MS, PhD
Research Professor
Director, Integrative Health & Wellbeing Research Program
Earl E. Bakken Center for Spirituality & Healing,
University of Minnesota

Katie Hadlandsmyth, PhD
Associate Professor
College of Nursing
University of Iowa

Keywords

Chronic Pain; Whole-Health; Biopsychosocial; Veterans Administration (VA); Rural Populations

Key Points

  • Rural Veterans exist at the intersection of 2 populations that are disproportionately affected by chronic pain. Compared to urban Veterans, rural Veterans are less likely to receive comprehensive and specialty pain care; are more likely to be prescribed opioid medication; and utilize pain self-management strategies at lower rates. The Veterans Administration (VA) serves 2.7 million rural veterans.
  • Seeking to improve pain management and reduce opioid use among rural patients, VA researchers developed the Rural Veterans Applying Mind Body Skills for Pain (RAMP) intervention. RAMP addresses pain as a biopsychosocial condition, providing rural VA patients with the opportunities and resources to enhance their capabilities and motivations to engage in helpful pain self-management behaviors.
  • A pilot study found that the intervention met milestones for enrollment, satisfaction, and fidelity. The researchers concluded that a full-scale randomized trial of a complementary and integrative health telehealth program for rural VA patients with chronic pain is feasible and can meet pain self-management needs. Their intervention and study processes have been refined to increase engagement and data collection.
  • In March 2026, the team will begin enrollment for a Type II randomized hybrid-effectiveness implementation trial.

Discussion Themes

The team emphasized the need for “resilient interventions” that can withstand external disruptions, such as natural disasters and VA workforce restructuring.

Relationship-building with high-level and local stakeholders was essential for navigating the VA’s complex and dynamic organizational structure. The researchers detailed their strategy for managing stakeholder panels, which included roughly 21 core members (i.e., patients and community advisors) and 10 internal VA stakeholders.

Grand Rounds February 13, 2026: The Making of the COMPARE-Pediatric IBD Study (Michael D. Kappelman, MD, MPH)

Posted on by Sophia Ramirez

Speaker

Michael D. Kappelman, MD, MPH
Professor, Pediatric Gastroenterology
University of North Carolina at Chapel Hill

Keywords

PCORnet; PCORI; Inflammatory Bowel Disease; Pediatrics; Common Data Model; Study Design

Key Points

  • Inflammatory Bowel Disease (IBD) is a chronic gastrointestinal condition affecting roughly 100,000 youth in the United States. It has a profound impact on nutrition, growth, physical, and psychosocial development. Anti-TNF biologics are the only FDA-approved advanced therapies for children, and approximately 30% of patients experience treatment failure within 2 years. There’s an urgent need for comparative effectiveness research that can guide treatment decisions when anti-TNF fails.
  • COMPARE-Pediatrics IBD, a PCORnet® study, includes 2 parallel multi-center, prospective cohort studies and retrospective cohort studies. The former, developed with multi-stakeholder input, will compare the effectiveness of emerging therapies in children with IBD; the latter will characterize the safety of these treatments and explore the heterogeneity of treatment effects across subgroups.
  • The study is utilizing PCORnet’s® infrastructure, including Prep-to-Research Queries and the PCORnet® Common Data Model (CDM), to inform the study design; identify administrative efficiencies; support recruitment; ease site burden; assess representativeness of the study population; and otherwise bolster their research.

Discussion Themes

Planning a PCORnet® study is a lot of work (and takes time). Start the process early and know that benefits may be on the back-end.

The study team opted not to conduct a randomized pragmatic trial because they anticipated that desperate families would be reluctant “roll the dice” with randomization and because insurance coverage for expensive off-label medications often dictates which therapy a patient can receive.

While the CDM is effective for structured data (like labs and diagnoses), Dr. Kappelman noted it cannot yet capture nuanced interpretations, such as specific MRI findings, which require more advanced AI or manual review.