Between 2020 and 2025, inpatient volumes on the medicine service line at our large, multisite academic medical center (AMC) grew by approximately 83%, and average hospital occupancy exceeded 90%. To keep pace, the division of hospital medicine (DHM) nearly tripled its rounding capacity, expanding from five to 15 teams. While this increased capacity, it heightened the complexity of patient assignments and exacerbated bottlenecks in emergency department (ED) triage, leading to admission delays. Despite the increase in teams, triaging responsibilities were distributed across rounding teams, requiring teams to field 30 or more ED pages daily while rounding, discharging patients, and managing intensive care unit (ICU) transfers. The consequences were predictable: slow page responses, prolonged ED boarding, delayed transfers, mounting cognitive load, and growing friction between ED and DHM practitioners.
Solutions Overview
To address this, we implemented a dedicated triage hospitalist model: a single clinician whose sole role was to serve as the hospital’s front door. Triage hospitalist roles are well described at large AMCs, where they have shortened admission times and improved throughput.1,2,3 We aimed to adapt that model to a high-occupancy, multisite quaternary care center and design it so that hospitals of any size, not only large AMCs, could apply the same principles.
Our reasoning was simple: front-door triage is the highest-leverage decision point in hospital flow; if we could get it right, it would lead to bidirectional gains across the ED, ICU, and inpatient units. We hoped that this would lead to faster, more appropriate admissions upstream and fewer interruptions and earlier discharges downstream on inpatient units.
We piloted a hybrid triage hospitalist model with two shifts: (1) a remote morning shift (7 a.m. to 1 p.m.) centralizing pager management and overnight admission distribution and (2) an in-person afternoon shift (1 p.m. to 7 p.m.) continuing triage while staffing additional admissions once rounding teams reached capacity. The role functioned as a centralized clinical and operational decision-maker: assigning patients, adjudicating level of care, guiding right-site placement, and coordinating ICU downgrades.
The role was supported by centralized digital pager management, the development of a triage dashboard showing incoming demand at the point of entry, real-time physician availability, and standardized handoffs using the situation, background, assessment, and recommendation (SBAR) format. Core stakeholders included DHM, ED, case management, ICU, nursing, and operational leadership at affiliate sites that would receive redirected patients. The objectives were explicit: improved throughput, right-site placement, safer transfers, and fewer interruptions for rounding teams to facilitate rounds and discharges.
Implementation Overview
Mapping the Current State
A review of our AMC’s 2020-2025 records showed the medicine admissions pager often received more than 40 pages a day—at times up to 10 in a single hour—while the physicians answering those pages were also rounding and discharging. On paper, the problem was not effort or head count; it was that the most time-sensitive decision in the system had no owner. Triage was an unstaffed function that needed to have its own stakeholder.
Designing the Role
We knew the responsibilities of triaging varied depending on the time of day. For example, morning work (7 a.m. to 1 p.m.) is largely coordination that can be done remotely: receiving the overnight backlog, distributing admissions, and managing the pager. The remote nature of the work also widened the pool of clinicians who could staff it. In contrast, afternoon work (1 p.m. to 7 p.m.) is more variable and physical: teams often reach their caps, and someone must absorb additional admissions while triaging; thus, the afternoon triage role was designed to be in-person.
Standardizing Decisions and Handoffs
To keep decisions consistent across hospitalists who might do the role, we standardized the responsibilities whenever possible. SBAR handoffs structured every transfer of information. Standardized pathways for selected presentations (for example, routing a heart failure exacerbation to cardiology) and short-stay scenarios guided inpatient-versus-observation decisions in real time. Pre-transfer safety criteria guided ICU downgrades; for instance, we ensured that patients being downgraded from the ICU had been off vasopressors for at least 12 hours, extubated for at least 24 hours, and had unnecessary lines removed.
Building Trust
The triage hospitalist model only works if both sides (accepting hospitalists and admitting ED physicians) trust its judgment, so we positioned the role as a neutral advisor, not a gatekeeper. When a patient’s disposition was unclear, the triage hospitalist worked with the ED physician to create appropriate contingency plans and facilitate appropriate disposition. If the triage hospitalist felt the patient was unstable for the floor, for example, they would speak with the ED to streamline an ICU consult before formally accepting. This form of shared decision making led to patients being accepted in a way that was consistent and prioritized safety. Over time, having one credible and predictable point of contact did more for ED-DHM relations than any single policy.
Outcomes and Impact
The preliminary results below are gathered from our first year of implementation (with the comparison periods stated); they are broken down by throughput, appropriate placement or bed utilization, discharge flow, and clinician experience or interdepartmental culture.
Throughput
During the peak winter months (December through March), the implementation of the triage hospitalist led to a 31% year-over-year improvement in ED page-to-response time and a 14% improvement in ED boarding time. Faster responses and shorter boarding are important because boarding is associated with worse patient outcomes and because ED length of stay rises sharply once hospital occupancy exceeds 90%.4,5,6
Right-site Placement and Utilization
Implementation of the triage hospitalist coincided with a 39% increase in admissions to a satellite-site campus; roughly nine patients per day were redirected from the main hospital to a partner location. Operationally, this unlocks bed capacity in the highest-demand building and reserves it for more complex patients, while routing appropriate, clinically stable patients to sites that would otherwise sit under capacity.
Discharge Flow
Since implementation of the triage hospitalist, we have seen an 18% year-to-date improvement in discharge orders placed by 9 a.m., attributable in part to removing triage interruptions from rounding teams. These discharges free beds earlier in the day, when ED demand is building, allowing for more streamlined flow throughout the day when resources are ample, and transfers are more easily completed.
Physician Experience and Interdepartmental Culture
Among 98 respondent clinicians (ED, DHM, and inclusive of house staff working in each setting) who worked both before and after implementation of the triage hospitalist system, 88% reported that the admissions workflow improved (88% and 89% for ED and DHM physicians, respectively). Additionally, 75% reported fairer, more consistent admission decisions and 70% reported reduced admission-related burnout, with majorities reporting improved ED–DHM communication and professionalism.
Lessons Learned
The most transferable lesson is counterintuitive: adding rounding teams can make flow worse, not better, if the coordinating function is not redesigned at the same time. More teams—with different staffing models and schedules—multiply the assignment decisions, and distributing those decisions across busy clinicians guarantees that none of them owns the work. A single, dedicated, front-door decision maker reduced interruptions across the whole system, which is why the downstream discharge gains appeared alongside the upstream throughput gains.
Trust is at the heart of any new role. The literature shows that ED and triage clinicians frequently disagree on disposition, so we knew rigid rules would be insufficient for disposition.7 For this reason, since its inception, the triage hospitalist has worked with the ED to facilitate admissions. When disagreements arose, they were handled with mutual respect and professionalism, creating an environment where psychological safety was prioritized, and patients were admitted when both parties felt comfortable with the decision. At times, this required contingency planning and at other times it meant waiting for patients to stabilize; however, in all instances, the plan was created with shared decision making between the triage hospitalist and the ED physician, leading to better and safer patient care. That culture, more than any protocol, is what reduced admission-related burnout on both sides.
Build the role with durable infrastructure, so it is not dependent on one exceptional clinician. In our case, we have more than 120 faculty who can do these shifts, so creating standardized responsibilities leads to a framework that scales and helps streamline the role, regardless of who is doing it.
The gains here came not from head count or system resources, but rather from giving triage a single, accountable owner, something a community or smaller academic medical center can do at far lower cost. A smaller model might include one part-time triage shift covering the highest-volume hours rather than a full multi-shift structure and rely on a shared pager and SBAR handoffs before investing in a dashboard.
Future Directions
Our nearest-term priority is to grow the service in the two directions that matter most to a multi-site AMC. We are working to expand triage to additional services, including cardiology and oncology, while also increasing triage support for other affiliate sites beyond the main hospital.
Key Takeaways
- Expanding rounding capacity without redesigning the coordinating function increases assignment complexity; a dedicated front-door decision-maker simplifies flow.
- Centralizing admission triage produces bidirectional gains: faster, more appropriate admissions upstream and fewer interruptions and earlier discharges downstream.
- The triage clinician needs credibility with both ED and DHM; the role should be positioned as a neutral advisor, not a gatekeeper.
- Standardized pathways and digital tools (dashboard, centralized paging, SBAR handoffs) are most effective when paired with real-time clinician judgment.
- Design the role as durable infrastructure, with staffing, pager routing, coverage, education, and measurement planned from the outset, so it survives beyond any individual clinician.
Dr. Juthani
Dr. Derry
Dr. Baum
Dr. Lew
Dr. Chiu
Dr. Pandya
Ms. Scott
Dr. Hom
Dr. Juthani is a clinical assistant professor of medicine at Stanford Medicine and hospitalist at Stanford Health Care in Stanford, Calif. Dr. Derry is a clinical assistant professor of medicine at Stanford Medicine and a hospitalist and the associate chief quality officer for inpatient care at Stanford Health Care in Stanford, Calif. Dr. Baum is clinical assistant professor, co-director of the longitudinal POCUS curriculum, and co-director of education for the Stanford Hospital Medicine POCUS group at Stanford University and a hospitalist at Stanford Health Care, all in Stanford, Calif. Dr. Lew is a clinical assistant professor of medicine at Stanford Medicine, and an attending hospitalist and medical director at Stanford Health Care, all in Stanford, Calif. Dr. Chiu is a clinical assistant professor at Stanford Medicine and hospitalist at Stanford Hospital, all in Stanford, Calif. Dr. Pandya is a clinical assistant professor at Stanford Medicine and a hospitalist at Stanford Health Care in Stanford, Calif. Ms. Scott is the administrative division director at Stanford Medicine in Stanford, Calif. Dr. Hom is a clinical professor of medicine at Stanford Medicine and chief of division of hospital medicine at Stanford Health Care in Stanford, Calif.
References
- Velásquez ST, et al. Hospitalists as triagists: description of the triagist role across academic medical centers. J Hosp Med. 2020;12(2):87-90. https://doi.org/10.12788/jhm.3327.
- Howell E, et al. Active bed management by hospitalists and emergency department throughput. Ann Intern Med. 2008;149(11):804-11. https://doi.org/10.7326/0003-4819-149-11-200812020-00006.
- Bartlett C, et al. The hospitalist triage role for reducing admission delays: impacts on throughput, quality, interprofessional practice, and clinician experience of care. J Clin Outcome Mgmt. 2023.;30(3):71-78. https://doi.org/10.12788/jcom.0128.
- Laam LA, et al. Quantifying the impact of patient boarding on emergency department length of stay: All admitted patients are negatively affected by boarding. J Am Coll Emerg Physicians Open. 2021;2(2):e12401. https://doi.org/10.1002/emp2.12401.
- Rathlev NK, et al. Time series analysis of emergency department length of stay per 8-hour shift. West J Emerg Med. 2012;13(2):163-8. https://doi.org/10.5811/westjem.2011.7.6743.
- Forster AJ, et al. The effect of hospital occupancy on emergency department length of stay and patient disposition. Acad Emerg Med. 2003;10(2):127-33. https://doi.org/10.1111/j.1553-2712.2003.tb00029.x.
- Amick A, Bann M. Characterizing the role of the “triagist”: reasons for triage discordance and impact on disposition. J Gen Intern Med. 2021;36(7):2177-2179. https://doi.org/10.1007/s11606-020-05887-y.