MDAware: Upgrading Clinical Judgment

In this episode of MDAware: Upgrading Clinical Judgment, Dr. Graham Walker talks with Dr. Howard Dinh about how to get physicians to better use clinical decision tools, and how a health system decides which tools are most worth building in the first place. A conversation about why low-value testing often has less to do with a doctor's knowledge than with time pressure, anxious patients, and the incentives they work under, and what it takes to put guidance at the point of care in a way clinicians trust.

Dr. Dinh is Chair of Cardiology for Kaiser Permanente Northern California and leads the region's clinical decision support subgroup. He spent ten years in practice before joining Kaiser, and he and Graham are former colleagues at TPMG. He traces the group's start a decade ago, when about 70% of ischemic testing across KP Northern California was going to patients who didn't need it, and explains how talking with primary care doctors, and understanding what they were dealing with, led to a stable chest pain tool that does the pretest probability math for them. From there, he covers the practical side of scaling and driving adoption across 22 medical centers: Dr. Walker and Dr. Dinh also take on whether clinical decision support only makes sense in value-based care, or whether fee-for-service practices face the same pressure. The conversation then turns to AI: why answers that shift with the prompt worry Dr. Dinh, where machine learning already supports KP's risk models, and a new joint venture that will let Kaiser test its tools against ten years of real patient outcomes. He closes with advice for physicians who want to start a decision support group at their own institution.

• Why much of the overuse of ischemic testing came from system pressures rather than physician judgment
• How to get medical centers aligned on one algorithm while leaving room for local differences
• Why a tool that takes six clicks can still save time if it documents the visit for you
• How value-based and fee-for-service incentives shape whether decision support gets used
• How KP plans to check its tools against a decade of patient outcomes so it can help set guidelines instead of only following them

MDCalc: mdcalc.com

Follow MDCalc on LinkedIn: linkedin.com/company/mdcalc

Follow MDCalc on X/Twitter: @MDCalc

In this episode of MDAware: Upgrading Clinical Judgment, Dr. Shazia Siddique talks with Dr. Sadiya Khan and Dr. Donald Lloyd-Jones about the PREVENT equations and why they're replacing the atherosclerotic cardiovascular disease (ASCVD) pooled cohort equations (PCEs) that have guided cardiovascular risk assessment for over a decade. A conversation about what changes when a risk tool expands beyond a single disease outcome, and why the same patient can get two different risk numbers depending on which calculator a clinician reaches for, and how this shapes the treatment conversation.

Dr. Khan is a preventive cardiologist at Northwestern who led the development and validation of the PREVENT equations. Dr. Lloyd-Jones directs the Framingham Heart Study and was a lead voice on risk assessment going back to the original pooled cohort equations. Together they trace the full arc: from the 1998 Framingham risk equations, built almost entirely on a European ancestry population, to the 2013 pooled cohort equations, to PREVENT itself, which now accounts for coronary heart disease, stroke, and heart failure in a single model, drawing on data from more than six million individuals.

Dr. Khan and Dr. Lloyd-Jones also walk through one of the more consequential decisions behind PREVENT: removing race as a variable in the equations, and why that decision didn't come at the cost of accuracy across different populations. From there, the conversation moves into how PREVENT is already being applied in the 2025 blood pressure guidelines, the CPR framework (calculate, personalize, reclassify) for translating a risk score into an actual treatment conversation, and what role tools like coronary artery calcium scoring play when the picture still isn't clear.

What You'll Learn

• Why coronary heart disease alone was an inadequate endpoint for risk prediction, and how leaving stroke and heart failure out of the equation left real risk on the table, particularly for women and Black Americans
• Why and how PREVENT drops race as a predictor in favor of directly measured risk factors
• What CPR (calculate, personalize, reclassify) means in practice
• How the new 2025 blood pressure guidelines arrived at the PREVENT CVD 7.5% threshold
• Why a risk score can tell you where a patient stands today, but can't calculate the benefit of a hypothetical intervention, and why that distinction matters at the bedside

Links & Resources

MDCalc: mdcalc.com

Follow MDCalc on LinkedIn: linkedin.com/company/mdcalc

Follow MDCalc on X/Twitter: @MDCalc

In this episode of MDAware: Upgrading Clinical Judgment, Dr. Joe Habboushe talks with Dr. Eric Steinberg about what happens after a PE is diagnosed — and why the field has spent decades obsessing over finding the clot without giving equal attention to what comes next. A conversation about risk stratification, disposition, and why not all PEs deserve the same treatment.

This episode picks up directly from Part 1, where Dr. Habboushe and Dr. Steinberg covered the diagnostic tools for PE. Here, the conversation shifts to management: once a patient has a confirmed PE, how do you decide whether they go home, to the floor, or to the ICU? Dr. Steinberg walks through why a PE causes damage through strain on the heart rather than through the lungs directly, and why hypoxia is a misleading early signal.

Dr. Habboushe and Dr. Steinberg also unpack why a low mortality score isn't the same as a green light for discharge, how the Hestia criteria closes that gap, and what the Bova score adds for patients who need a higher level of care, including when it's time to activate a PERT team.

What You'll Learn

• Why a PE doesn't cause hypoxia on its own, and why waiting for low oxygen levels can lead clinicians to miss how sick a patient really is
• Why the simplified PESI (sPESI) score has real blind spots when used alone, since it measures all-cause mortality rather than PE-specific risk
• How the Hestia criteria pairs with sPESI to answer a more practical question: is this specific patient safe to send home
• What the Bova score predicts, and how it helps guide decisions on escalation, intervention, and when to bring in a PERT team
• What treatment options exist beyond standard anticoagulation, including catheter-directed thrombolysis and mechanical thrombectomy, and when each is appropriate

Links & Resources

MDCalc: mdcalc.com

Follow MDCalc on LinkedIn: linkedin.com/company/mdcalc

Follow MDCalc on X/Twitter: @MDCalc

In part one of this two-part episode, MDCalc cofounder and CEO, Dr. Joe Habboushe, talks with Dr. Eric Steinberg about how to actually diagnose a pulmonary embolism — and considering scenarios in which the biggest risk in PE workup may not be the disease itself, but the testing we order to find it. A conversation about pretest probability, when a CT scan does more harm than good, and how to use clinical tools without losing the judgment they're supposed to support.

Dr. Habboushe and Dr. Steinberg have been talking through PE cases together for close to two decades, since Dr. Steinberg was a resident under Dr. Habboushe's supervision. Their conversation was prompted by new multi-society PE guidelines, released this year, which spotlight eight clinical tools out of the 36 PE-related calculators on MDCalc. In this episode, they focus on the diagnostic side: how to establish pretest probability, when the PERC rule applies and when it doesn't, and why the D-dimer test causes more confusion than almost anything else in emergency medicine.

Dr. Steinberg and Dr. Habboushe also walk through a common misconception about the YEARS algorithm, why age-adjusted and altitude-adjusted versions of these tools exist, and what it actually means to communicate a low-risk result to a patient who came in expecting a scan.

What You'll Learn

• Why scanning every patient isn't the safe default, and how incidental findings and false positives can cause more harm than the small risk of missing a PE
• Why the PERC rule is a one-directional tool: failing it doesn't raise a patient's risk, it just means the rule no longer applies
• How to choose between Wells and Geneva based on clinical experience, and why Geneva's objectivity makes it a better starting point for trainees
• Why the D-dimer's negative predictive value, not its false positive rate, is the piece most clinicians misunderstand
• Why the YEARS algorithm is often mistaken for a pregnancy-only tool, and what its flexible D-dimer threshold actually does

Links & Resources

DVT and PE Tool Collection

MDCalc: mdcalc.com

Follow MDCalc on LinkedIn: linkedin.com/company/mdcalc

Follow MDCalc on X/Twitter: @MDCalc

In this episode of MDAware: Upgrading Clinical Judgment, Dr. Graham Walker talks with Dr. Sarah Gebauer about what responsible AI in healthcare actually looks like — and what happens when the technology moves faster than our ability to govern it. A conversation about values, accountability, and why the physician-patient relationship still has to be at the center of it all.

Dr. Gebauer and Dr. Walker co-authored the MDCalc Physician's Charter in 2023, one of the earliest attempts by a group of practicing physicians to define a set of shared values for AI in healthcare. In this conversation, they revisit that document; what it got right, what it underestimated, and what a 2026 update would need to address. From the re-identification risks hiding inside "de-identified" data, to the tension between building accurate AI and protecting patient privacy, to the question of whether a sycophantic model is ever the right tool for a clinical setting, the conversation covers ground that most AI discussions in medicine still aren't touching.

Dr. Walker and Dr. Gebauer also look at where healthcare institutions are making real progress on AI governance, why rural hospitals may be further ahead than most people assume, and what it actually means to roll out a new tool responsibly — in medicine or anywhere else.

What You'll Learn

• Why HIPAA's definition of de-identified data may no longer be adequate in an era when AI can re-identify patients from surprisingly little information
• How the tension between diverse training data and patient privacy creates an ethical challenge the field hasn't fully resolved
• What questions healthcare systems should be asking before adopting any new AI tool, and why workflow fit matters as much as clinical accuracy
• Why AI sycophancy, models trained to be agreeable, could be a meaningful problem in clinical and patient-facing settings
• What a 2026 update to the Physician's Charter might need to address, including AI agents, foundation models, and tools that operate entirely outside the physician-patient relationship

Links & Resources

MDCalc: mdcalc.com

Physician's Charter: https://physicianscharter.ai/

Follow MDCalc on LinkedIn: linkedin.com/company/mdcalc

Follow MDCalc on X/Twitter: @MDCalc

In this episode of MD Aware: Upgrading Clinical Judgment, host Dr. Shazia Siddique sits down with Dr. Nadiem Mahmud, Assistant Professor of Medicine in the Division of Gastroenterology and Hepatology at the University of Pennsylvania, to explore one of the most clinically impactful new tools now available on MDCalc: the VOCAL-Penn Score. Designed to predict postoperative mortality and cirrhosis decompensation in patients with liver disease, the VOCAL-Penn Score addresses a longstanding gap that has challenged hepatologists, gastroenterologists, and surgeons for decades.

Dr. Mahmud walks through how the score was born out of real clinical frustration with existing tools like the MELD score, Child-Pugh score, and Mayo Risk Score, all of which frequently produced wildly discordant estimates that clashed with clinical intuition. The VOCAL-Penn Score takes a more comprehensive approach, incorporating surgery type as a key variable and delivering four distinct outputs: 30-day, 90-day, and 180-day postoperative mortality risk, as well as 90-day risk of cirrhosis decompensation. Now used by more than 50,000 clinicians annually across the globe and endorsed by both the American College of Gastroenterology and the European Association for the Study of Liver Disease, the tool is reshaping how clinicians approach surgical risk conversations with their patients.

Dr. Siddique and Dr. Mahmud also dig into the common misapplications of the score, what a VOCAL-Penn 2.0 might look like, and why the move to MDCalc represents a major step forward in making this tool accessible to the four out of five US clinicians who already use the platform every day.

What You'll Learn

• How the VOCAL-Penn Score improves on older cirrhosis surgical risk tools by incorporating surgery type and delivering four distinct risk outputs to support shared decision-making
• Why there is no single hard cutoff for proceeding with surgery, and how a 15% projected 90-day mortality threshold can serve as a trigger for preoperative liver transplant evaluation
• Common pitfalls when applying the score, including using outdated lab values, applying it to unsupported surgery types, and failing to calculate risk for both laparoscopic and open surgical scenarios
• What a future version of the tool may include, such as broader surgery categories, additional cardiometabolic risk factors, and more diverse patient populations

Links & Resources

MDCalc: mdcalc.com

VOCAL-Penn Score on MDCalc: mdcalc.com

VOCAL-Penn Score original site: vocalpennscore.com

University of Pennsylvania Gastroenterology and Hepatology: pennmedicine.org

American College of Gastroenterology: gi.org

European Association for the Study of Liver Disease: easl.eu

Follow MDCalc on LinkedIn: linkedin.com/company/mdcalc

Follow MDCalc on X/Twitter: @MDCalc

In the inaugural episode of MD Aware: Upgrading Clinical Judgment, co-hosts Dr. Shazia Siddique and Joe Habboushe, co-founder and CEO of MDCalc, pull back the curtain on one of the most ambitious projects in MDCalc's 20-year history: the Quality Rating System (QRS). Designed to help clinicians quickly identify the most evidence-based and clinically relevant tools at the point of care, the QRS represents a major step forward in bringing transparency and rigor to the hundreds of clinical decision-making tools available today.

Joining the conversation are two distinguished members of the QRS Advisory Board: Dr. Helen Burstin, CEO of the Council of Medical Specialty Societies, and Dr. Joseph Wright, Chief Health Equity Officer at the American Academy of Pediatrics. Together, they discuss how the QRS was developed using a Delphi consensus methodology, why professional society endorsement matters more than most clinicians realize, and how embedding fairness and bias detection directly into the rating criteria could reshape how clinical tools are built, validated, and used across diverse patient populations

From busy emergency departments to routine office visits, the QRS promises to give every clinician the confidence to reach for the right tool at the right moment. Whether you're a frontline provider, a researcher, or a guideline developer, this episode offers a compelling look at how raising the bar for clinical tools can ultimately translate into better outcomes for every patient.

What You'll Learn

• The four core domains of the QRS, including importance, scientific soundness, fairness and algorithmic bias, and usability, and why each one matters at the bedside
• How the Delphi consensus process brought together clinical experts to iteratively refine the rating criteria against real-world calculators
• Why professional society endorsement is a key signal for clinicians, and how the QRS could push societies to keep their guidance more current
• How the QRS can fuel future research by identifying gaps in evidence, surfacing potential population-level harms, and encouraging multi-institutional collaboration

Links & Resources

MDCalc: mdcalc.com

Council of Medical Specialty Societies (CMSS): cmss.org

American Academy of Pediatrics (AAP): aap.org

Follow MDCalc on LinkedIn: linkedin.com/company/mdcalc

Follow MDCalc on X/Twitter: @MDCalc