Organizers and Facilitators: Lucy Shi, Vasundhara Singh, Sonia Dalal, Gregory Leslie, Jack Cunningham, Amit Pahwa, Jack Badawy

Background

In March 2026, we conducted a first of its kind use of HOMERuN's focus group infrastructure for the qualitative part of a mixed-method needs assessment to help guide us in the creation of a curriculum for caring for patients with cirrhosis in the hospital. This is an initial summary of the discussion that took place during that meeting.

Discussions highlighted significant variation in the knowledge, processes, and systems supporting inpatient cirrhosis care. Participants described a shared sense that cirrhosis management is high-stakes given high mortality rates of patients with decompensated cirrhosis, nuanced, and often under-supported, particularly outside tertiary hepatology/transplant centers. The themes that emerged point to clear opportunities for targeted education paired with workflow redesign.

Where Knowledge Gaps are Most Pronounced

Clinicians identified recurring uncertainty across the spectrum of cirrhosis management. Diagnostic distinctions — such as hepatorenal syndrome versus pre-renal AKI, appropriate use of albumin challenges, and recognition of acute alcoholic hepatitis — remain challenging and often delay definitive management. Similarly, core day-to-day questions persist: Who truly needs SBP prophylaxis? How aggressively should coagulopathies be corrected before procedures?

Hepatic encephalopathy was described as a particularly difficult area. Many teams struggle with early recognition, overuse of ammonia levels to inappropriately guide treatment, uncertainty about next steps when lactulose fails, and identifying precipitating triggers.

Participants emphasized the importance of asking a fundamental but often overlooked question: why did this patient decompensate now? It is important to get to the cause of the decompensation when possible.

Additional gaps included diuretic strategy (PO vs IV, hepatic vs cardiac volume overload, avoiding AKI), management of hyponatremia, portal vein thrombosis, thrombocytopenia, nutrition and sarcopenia, and understanding the indications — and complications — of TIPS.

Repeatedly, clinicians noted uncertainty about when to involve hepatology, how to integrate palliative care earlier, and how to navigate transplant eligibility, referral processes, and insurance barriers.

"The translation from evidence generation/guideline creation to then actual clinical practice at the front lines is a gap for generalists and is an opportunity to stay updated and teach it."

System-level Barriers Undermining Best Care

Even when knowledge exists, system barriers frequently prevent timely, guideline-concordant care. Delays in paracentesis were the most common frustration, driven by comfort level, unclear ownership, staffing limitations (especially on weekends and holidays), and resource constraints such as inadequate kits or albumin shortages. Participants questioned whether upskilling APPs and clarifying responsibility could close this gap.

Other barriers included inconsistent order sets, lack of cirrhosis-specific quality indicators, and uncertainty about who is accountable for meeting them. Clinicians faced backlogs for transfer to referral centers and unclear criteria for who should be transferred. Structural inequities were also front and center. Patients with cirrhosis are disproportionately affected by socioeconomic disadvantage, substance use disorders, and limited social support, which can make discharge planning, follow-up, and transplant pathways especially challenging. Variable access to hepatology expertise — particularly in community settings — was seen as a major driver of inconsistent care and missed opportunities to discuss prognosis.

"Just the fact that a group of experienced hospitalists like us couldn't necessarily name what the cirrhosis quality indicators are is telling."

How Clinicians Are Closing the Gaps

Participants rely on a patchwork of resources to stay current: UpToDate, Epic-based tools and order sets, consultant notes, conferences, podcasts, books, and increasingly AI-enabled clinical decision support. Many emphasized that some of the most meaningful learning occurs informally — through difficult cases, conversations with hepatology colleagues, and real-time bedside decision-making.

"One thing that has been really helpful is that the Hep fellows have these nice, short templates of assessment and plans that they go through with things that even the metrics don't capture, like did we screen for Budd-Chiari Syndrome. In addition to order sets — bringing attention to all the factors in a templated fashion in our notes has been a good facilitator (to knowledge acquisition)."

What a Successful Curriculum Should Achieve

There was strong alignment around what success would look like. A well-designed cirrhosis curriculum should measurably improve confidence and efficiency, increase appropriate transplant referrals, and ultimately improve patient outcomes. Participants emphasized pairing education with system support, such as standardized order sets, note templates, and process metrics (e.g., paracentesis within 24 hours).

Equally important was feasibility: education should be high-yield, practical, and easy to access, embedded within existing teaching structures and the EMR rather than added as another burden. Short, focused content reinforced by workflow-based decision support was viewed as the highest-value approach.

"Education is necessary but not sufficient. You need to convince people that something is the right thing to do but then we need the nudges to do them in real time."

Conclusion:



Improving cirrhosis care will require more than knowledge alone. As Hospitalists, the participants emphasized the need for integrated education, clear processes, and system-level support to close persistent gaps, reduce variation, and better meet the needs of a complex and vulnerable patient population reflecting a robust quality improvement mindset.

New Publication

Investigators from the UPSIDE Study have recently published a secondary data analysis to determine whether patients with sepsis have a different risk of diagnostic error (DE) and/or underlying diagnostic process faults compared with other patients who die or are admitted to an ICU. DE rates were similar between those with sepsis and those without (22.3% vs. 23.9%, p = .734) and adjusted analysis showed no association between sepsis and DE (aRR: 1.06, 95% confidence interval [CI]: 0.91–1.23). Sepsis was less strongly associated with DE if gaps in history-taking (interaction-term p = .02) or consultation problems (interaction-term p = .005) were present. No other DEER interactions were significant. Our results suggest that diagnostic issues in sepsis are tied to broader problems with clinical care rather than sepsis-specific factors.

Prasad PA, Hubbard C, Lee T, et al. The association between sepsis and diagnostic errors: A secondary analysis of the Utility of Predictive Systems for Diagnostic Error study. J Hosp Med. 2026;1-7.

Our next HOMERuN meeting will be on August 7, 2026.

Image Attributions: Icon images from https://www.flaticon.com and https://www.vecteezy.com.