National Daily Hospital Performance Playbook Chapter 7: THE HOSPITAL WITHOUT SLACK Building Operating Resilience Before the Next Crisis

 

"Hospitals are often planned through averages and experienced through peaks."

-Chat GPT

No Beds at 7:03 A.M.

At 7:03 on a Tuesday morning, North Valley Medical Center looked full but manageable.

The overnight census was 94 percent. The emergency department held seventeen patients, including four who had been admitted and were waiting for inpatient beds. Two patients were expected to leave the intensive care unit. Eleven medical-surgical discharges were identified for the day. The operating room schedule was heavy but ordinary. Staffing was tight, as it had been for months, but every unit had a charge nurse and enough people to open its scheduled beds.

Nothing on the morning report looked like a crisis.

By 9:15, one ICU transfer was delayed because the receiving nurse had been reassigned. Three expected discharges were waiting for medication reconciliation. Two families had not arranged transportation. A patient who was medically ready to leave needed oxygen at home, but the vendor had not received the complete order. Environmental services had six rooms to turn and two employees absent. The emergency department had received two ambulances within eleven minutes.

At 10:40, the post-anesthesia care unit was holding three patients whose inpatient rooms were not ready. The surgical schedule continued because no single leader believed they had authority to slow it. At 11:20, the ED charge nurse requested help. The nursing supervisor began calling units individually. Case management was working hard on the same delayed discharges the bed office was counting as available. Pharmacy had not been told which discharge prescriptions were now urgent. Transportation knew nothing about the families who could not arrive before evening.

At 1:10, the hospital declared a capacity alert.

At 3:30, administrators assembled in a conference room. They reviewed a dashboard showing what the people closest to the work had known for hours: North Valley had run out of places to put patients.

No extraordinary disaster had occurred. There was no mass casualty event, cyberattack, or regional epidemic. The hospital had simply absorbed a series of ordinary variations without a reliable way to recognize their combined effect, assign authority, or reorganize the work.

North Valley did not run out of beds at 1:10. It began losing them at 7:03.

This distinction is the subject of this chapter.

The End of Average-Day Management

Hospitals are often planned through averages and experienced through peaks.

The annual budget assumes an average daily census, an average length of stay, an average number of procedures, and an average number of productive hours. Departmental schedules are built days or weeks in advance. Monthly reports explain what happened after the people who experienced it have moved on to the next problem.

But patients do not arrive at an average rate. Discharges do not occur evenly. A licensed bed is not an available bed when there is no nurse to staff it, no housekeeper to clean it, or no transport process to move the next patient into it. A hospital can have theoretical capacity on paper and no usable capacity at the bedside.

The Agency for Healthcare Research and Quality has emphasized that emergency department boarding originates at the hospital and health-system level and therefore requires solutions beyond the ED. Boarding is not simply an emergency-department inconvenience. It is a visible symptom of decisions and delays across inpatient nursing, medical staff, diagnostics, pharmacy, case management, environmental services, transportation, post-acute care, and executive leadership. Research summarized by AHRQ associates prolonged boarding with delayed care, medical errors, poorer outcomes, and excess mortality.[1]

The operating problem is not that leaders lack data. Hospitals have more data than at any point in their history. The problem is that much of it is retrospective, fragmented, or disconnected from authority.

A monthly metric can identify a pattern. It cannot open a bed this afternoon.

A dashboard can display a bottleneck. It cannot decide which procedure should proceed, which discharge barrier must be escalated, which staff member can safely be reassigned, or which executive must resolve a conflict between departments.

Average-day management asks, “How did we perform last month?” Resilient management asks three additional questions:

1.       What is likely to happen during the next four, eight, and twenty-four hours?

2.       What operating conditions would place patients or employees at risk?

3.       What action will we take before those conditions arrive?

The answer requires more than a command center. It requires a management system.

Slack Is Not Waste

For years, hospital leaders have been taught to view unused time, unoccupied space, and uncommitted labor as waste. Some of it is. Poor scheduling, redundant documentation, avoidable waiting, unnecessary handoffs, rework, and poorly designed processes consume resources without helping patients.

But not every unused resource is waste.

An emergency department with no ability to absorb the next ambulance is not optimally efficient. An inpatient unit with no nurse able to receive a transfer is not fully productive. A discharge process that works only when every physician, family, pharmacy, vendor, and transportation service performs exactly on time is not lean. It is brittle.

Resilience requires a small but deliberate margin for variation. This margin may be a staffed flex bed, a cross-trained employee, a protected appointment, an on-call transportation option, a physician who responds to an escalation within fifteen minutes, or a leader with authority to rebalance the hospital before each department protects its own schedule.

The objective is not to keep expensive resources idle. The objective is to make the institution responsive.

This leads to an important operating principle:

Waste is capacity that produces no value. Resilience is capacity that can be reached when conditions change.

Hospitals frequently cut both at the same time because their accounting systems see the cost but not the response capability. The immediate expense declines. Overtime, turnover, boarding, diversions, canceled procedures, delayed care, and staff injury appear later in different accounts.

The hospital without slack may look efficient on a spreadsheet. It becomes expensive the moment the day departs from plan.

The Five Reserves Every Hospital Needs

Traditional emergency preparedness focuses appropriately on disasters. Operating resilience applies the same discipline to the ordinary disruptions that threaten care every day.

A resilient hospital maintains five forms of reserve. These reserves do not have to be large, but they must be visible, usable, and governed.

1. Bed and Physical-Capacity Reserve

This is not simply the number of empty licensed beds. It is the number of appropriate spaces that can safely receive the next patient within a defined period.

Leaders should know:

·         which beds are currently usable;

·         which beds can become usable within two, four, or eight hours;

·         what prevents closed beds from opening;

·         whether PACU, observation, procedural, or transitional capacity is being used for the right patients;

·         where specialty constraints—not total beds—will stop flow first.

2. Workforce and Skill Reserve

Head count is not capability. Workforce reserve depends on the skills available at the time and place of need.

It may include cross-trained nurses, an internal float pool, flexible shift lengths, rapid credentialing, unit-based help rules, team nursing options, and managers prepared to assume operational roles. The purpose is not to make fewer people do more. It is to prevent one uneven assignment, one absence, or one surge from becoming an unsafe condition for an entire unit.

3. Time and Schedule Reserve

Hospitals lose capacity when every minute is committed before uncertainty is known. Small schedule protections can prevent large delays: an urgent imaging slot, reserved procedural time, discharge medication priority, predictable environmental-services response, or a daily window for physicians to resolve barriers.

Time reserve also means knowing which work can be advanced, deferred, redistributed, or stopped when priorities change.

4. Information and Decision Reserve

Information reserve is the ability to know what is changing before the outcome is fixed. Decision reserve is the authority to act on it.

A forecast without an owner is interesting. A threshold without a response is decoration. A resilient system links each warning to a person, a decision rule, and a time expectation.

5. Community and Post-Acute Reserve

Hospital capacity increasingly depends on services outside the hospital: skilled nursing, home health, behavioral health, dialysis, oxygen, transportation, housing, caregivers, tribal health services, primary care, and pharmacy access.

When these supports are unavailable, their work does not disappear. It migrates into the hospital and occupies its most expensive capacity.

Rural hospitals face a particularly severe version of this problem. KFF estimates that the temporary $50 billion Rural Health Transformation Fund represents only about 37 percent of projected federal Medicaid reductions in rural areas over ten years.[2] Transformation funds can help build new capability, but temporary grants cannot substitute for a working local care network or reliable operating revenue. A rural resilience plan must therefore include community capacity, transportation, workforce housing, telehealth connections, transfer agreements, and service-line preservation—not simply internal cost reduction.

The Capacity Hospitals Already Own but Cannot Reach

Before adding beds or imposing new productivity targets, leaders should look for capacity already present but trapped inside the system.

This is the modern form of “hidden gold”: paid resources, clinical capability, and patient time that already exist but cannot be converted into timely care because of operating friction.

Common examples include:

·         beds occupied by medically ready patients awaiting a solvable nonclinical need;

·         discharge orders written late because likely discharges were not identified the day before;

·         rooms waiting for cleaning because departure information reached environmental services late;

·         nurses unevenly assigned because workload intensity is not visible across units;

·         PACU capacity occupied by admitted patients while scheduled cases continue unchanged;

·         diagnostic delays caused by work arriving in unpredictable batches;

·         patients held for consultations that could occur virtually or in a different setting;

·         repeated documentation and phone calls created by incompatible communication systems;

·         employees capable of helping another area but lacking cross-training or permission;

·         appointments lost to no-shows while urgent patients wait elsewhere;

·         leaders reviewing the same constraint in separate meetings without one person owning its resolution.

Hidden capacity cannot be recovered through pressure alone. Telling people to discharge faster, clean rooms sooner, or work harder does not remove the conditions preventing them from succeeding.

The better question is: “What must be true for this patient, bed, employee, or hour to become available?”

That question converts frustration into a management problem.

The Hidden Capacity Audit

For two weeks, each major service should record four things whenever expected capacity is unavailable:

Question	What to capture
What capacity did we expect?	Bed, staff skill, room, appointment, equipment, or time block
When was it needed?	Exact time and expected duration
What prevented its use?	Specific clinical, operational, communication, or external barrier
What would have released it?	Decision, information, person, vendor, authority, or redesigned process

 

Do not begin by assigning blame. Begin by finding repeated conditions. Ten isolated explanations may reveal one common system failure.

From Dashboard to Decision Rule

Hospitals often have red, yellow, and green indicators without a shared definition of what anyone must do when the color changes.

The result is alert fatigue at the management level. Everything is visible; little is governed.

A useful warning system contains five elements:

4.       Signal: What condition are we watching?

5.       Threshold: At what point does the condition require action?

6.       Owner: Who is accountable for initiating the response?

7.       Action: What specific response is expected?

8.       Escalation: When and to whom does the issue move if it is not resolved?

The thresholds below are examples. Each hospital should test and revise them using its own patterns, resources, and patient population.

Signal	Watch threshold	Action threshold	Required response
Admitted ED boarders	Forecast exceeds normal bed availability in 4 hours	Any high-risk boarder or boarding above local limit	Hospital-wide flow review; assign beds and barriers by name
Staffed-bed availability	Less than expected admissions for next 4 hours	Zero appropriate staffed beds	Open flex plan, rebalance staff, review elective inflow
Expected discharges	Fewer than 60% have complete plans by prior afternoon	Critical barriers unresolved by 9 a.m.	Executive barrier escalation by patient and owner
PACU boarding	One inpatient hold threatens schedule	Two or more holds or clinical risk	Joint OR–inpatient decision on pacing and placement
Nursing workload	One assignment exceeds local workload rule	Two units unable to accept safe assignments	Deploy skill reserve; nursing executive review
Environmental services	Turn queue exceeds capacity for next 90 minutes	Priority rooms delayed beyond local limit	Reassign queue and support; resolve departure notification defects
Post-acute placement	Expected placement cannot be confirmed	Patient medically ready with no viable destination	Activate payer, network, transport, and executive escalation

 

Decision rules should not remove judgment. They should ensure judgment arrives before the damage.

The Hospital Resilience Huddle

The huddle is not another census meeting. It is a short decision forum organized around the next operating period.

Its purpose is to convert fragmented forecasts into coordinated action. It should occur at a time early enough to change the day, with a second brief review when conditions warrant.

Required Participants

·         nursing operations or house supervision;

·         emergency services;

·         inpatient and critical care;

·         surgery and PACU;

·         hospital medicine or medical staff leadership;

·         case management and social work;

·         environmental services and patient transport;

·         pharmacy and diagnostics when constraints are present;

·         workforce or staffing office;

·         an executive with authority to resolve cross-department conflicts.

Fifteen-Minute Agenda

Minutes 1–3: Demand

What arrivals, admissions, procedures, transfers, and discharges are expected during the next four, eight, and twenty-four hours?

Minutes 4–6: Constraints

Which beds, skills, rooms, services, or external resources will become limiting first?

Minutes 7–10: Patients at Risk

Which named patients are experiencing or approaching a delay that could cause harm, deterioration, indignity, or avoidable expense?

Minutes 11–13: Decisions

What will be opened, reassigned, accelerated, deferred, or escalated? Who owns each action, and by what time?

Minutes 14–15: Confirmation

What condition will trigger the next huddle or executive intervention?

The huddle ends with assignments, not observations.

“Case management is working on it” is not an assignment.

“Maria will confirm oxygen delivery with the vendor by 9:30; if it cannot be delivered by noon, the chief operating officer will contact the regional account executive” is an assignment.

Specificity is not micromanagement when the patient is waiting. It is stewardship.

Artificial Intelligence as Lookout, Not Captain

Artificial intelligence can strengthen each of the five reserves. It can forecast arrivals, predict admission probability, identify patients likely to be medically ready for discharge, detect staffing gaps, estimate room-turn demand, summarize barriers, and test alternative schedules.

Recent operational research suggests that proactive bed-request models using predicted admission probability and anticipated time to disposition may reduce boarding and length of stay under simulated conditions.[3] These tools are promising because they move the decision point earlier.

But prediction is not management.

An algorithm may identify that twelve patients are likely to require beds. It does not negotiate priorities between surgery and the ED. It cannot determine whether an exhausted nurse can safely take another patient. It does not know that a family’s hesitation reflects fear rather than transportation. It cannot own the moral consequences of a decision.

AI should therefore serve as lookout, not captain.

Hospitals adopting predictive tools should require:

·         transparent definitions and confidence ranges;

·         validation on the hospital’s own population and workflow;

·         monitoring for bias and performance drift;

·         a visible human override;

·         documentation of who acts on each alert;

·         evaluation against patient, workforce, and equity outcomes—not only throughput;

·         a fallback process when the technology is unavailable.

The strongest use of AI is not to replace experienced managers. It is to give them earlier sight of the conditions they are responsible for changing.

Protecting People While Recovering Capacity

Operational improvement becomes destructive when every solution is translated into greater pressure on front-line employees.

People are not slack.

A nurse’s uncommitted minute is not automatically waste. It may be the minute needed to recognize deterioration, teach a family, assist a colleague, verify a medication, or recover enough attention to avoid an error.

Resilience does not ask people to run at maximum intensity for longer periods. It creates a system that detects overload, redistributes work, removes preventable friction, and makes assistance legitimate before a unit reaches failure.

This requires leaders to track the human conditions under which capacity is produced:

·         missed meals and breaks;

·         involuntary overtime;

·         assignment imbalance;

·         frequent floating without preparation;

·         reliance on premium labor;

·         sick calls and vacancy patterns;

·         workplace violence and injury;

·         turnover and intent to leave;

·         time spent locating information, supplies, equipment, or decision-makers;

·         employee reports that escalation did not produce help.

These are not soft measures. They are early indicators of operating fragility.

The leader’s responsibility is not merely to demand an outcome. It is to make the outcome reasonably achievable.

That principle was true when daily workload tools were maintained on paper. It remains true in an era of predictive command centers. Every employee can help manage outcomes only when expectations, information, authority, and support are shared honestly.

The Resilience Scorecard

The scorecard should balance prediction, response, patient outcomes, workforce conditions, and financial consequences. No single metric should be optimized at the expense of the others.

Dimension	Core measure	Management question
Demand accuracy	Forecast-to-actual census and admission variance	Are we seeing the day early enough to prepare?
Usable capacity	Appropriate staffed beds available by time horizon	How much real—not theoretical—capacity can we reach?
Access	ED boarding hours and longest boarding time	Who is waiting, for how long, and at what risk?
Discharge reliability	Percent of likely discharges with complete plan by prior afternoon; discharge time	Are we resolving barriers before the patient is ready to leave?
Procedural flow	PACU boarding hours; same-day cancellations	Are inpatient constraints disrupting scheduled care?
Workforce	Workload-rule breaches, overtime, premium hours, missed breaks	Are we producing capacity by improving the system or exhausting people?
Quality	Falls, medication delays, deterioration, infections, and safety reports during escalation	Does capacity pressure change the reliability of care?
Community flow	Avoidable days by external barrier	Which community constraints are consuming hospital capacity?
Response reliability	Time from warning threshold to assigned action	Do our alerts create timely decisions?
Financial resilience	Contribution loss from cancellations, excess days, premium labor, and diversion	What is fragility costing us?

 

Measures should be reviewed by shift, day, week, and month according to their purpose. The closer a measure is to patient harm, the shorter the review interval should be.

Post results where the people doing the work can see them. Invite correction. If staff know the data are wrong but leaders continue presenting them, the scorecard will become theater.

A 90-Day Resilience Build

Hospitals do not need to wait for a new building, enterprise platform, or consulting engagement to begin. A practical resilience system can be built and tested in ninety days.

Days 1–15: See the Real System

·         Establish an executive sponsor and operational lead.

·         Map the flow from expected arrival through discharge and follow-up.

·         Complete the Hidden Capacity Audit in the ED, inpatient units, surgery/PACU, case management, environmental services, and staffing office.

·         Identify the five most frequent capacity failures and the people affected.

·         Define current escalation practices, including where requests for help stop.

·         Select a small balanced set of baseline measures.

Deliverable: A one-page description of the hospital’s recurring capacity failure pattern—not a list of departmental complaints.

Days 16–30: Define Reserves and Decision Rules

·         Inventory the five reserves and identify gaps.

·         Establish watch and action thresholds using local data.

·         Assign one owner and escalation route to every threshold.

·         Create safe workload and flex-capacity rules with front-line participation.

·         Identify two or three community partners for recurring discharge barriers.

·         Test whether the required data can be produced reliably and early enough to matter.

Deliverable: A working threshold-and-response table understood by the people expected to use it.

Days 31–60: Pilot the Daily Discipline

·         Conduct the resilience huddle on selected days or services.

·         Use named assignments with owners and deadlines.

·         Hold a ten-minute end-of-day review: What did we predict, what happened, and what should change tomorrow?

·         Record decisions that were delayed because authority was unclear.

·         Track employee workload and safety effects alongside throughput.

·         Revise rules weekly rather than defending the first design.

Deliverable: A living management process that produces visible actions and learns from misses.

Days 61–75: Extend Across Boundaries

·         Connect the ED, inpatient, OR/PACU, diagnostics, and discharge processes.

·         Add pharmacy, transport, environmental services, behavioral health, and post-acute partners where they constrain flow.

·         Build cross-training and backup plans for the most fragile functions.

·         Test one high-demand and one workforce-shortage scenario.

·         Require executive participation when conflicts cross departmental authority.

Deliverable: A hospital-wide response capable of acting across silos.

Days 76–90: Institutionalize and Report

·         Finalize the first resilience scorecard.

·         Quantify recovered capacity, avoided delays, workforce effects, and financial impact.

·         Publish the decision rules and escalation expectations.

·         Recognize teams that surfaced problems early and tested solutions honestly.

·         Assign a leader to maintain, audit, and improve the system.

·         Present the next ninety-day plan to the board and front-line staff.

Deliverable: A management system owned by the hospital—not a temporary project owned by its designers.

Return to North Valley

Three months after its Tuesday capacity crisis, North Valley was still a busy hospital.

It still experienced sick calls. Families still encountered transportation problems. Post-acute facilities still declined patients. The ED still received arrivals in clusters, and the operating rooms still ran late.

The difference was that ordinary variation no longer remained invisible until it became an emergency.

At 7:03 on another Tuesday morning, the forecast showed that expected admissions would exceed staffed-bed availability by noon. Six likely discharges had unresolved barriers. Two required medication changes, one needed oxygen, one needed a family decision, and two were awaiting skilled-nursing confirmation.

The hospital did not declare a crisis.

Pharmacy assigned the two medication reconciliations for completion by 8:30. The oxygen order was corrected while the vendor still had a morning delivery route. A physician called the family before rounds. The payer escalation team contacted the two skilled-nursing facilities. Environmental services adjusted its first assignments based on expected departures. Nursing moved a cross-trained employee before either unit became unsafe. The OR and inpatient leaders agreed on the condition that would trigger a schedule adjustment rather than waiting for PACU to fill.

Not every plan succeeded. One discharge was delayed. A second patient remained in the ED longer than desired. But the hospital maintained control of the day because people saw the same risks, acted under shared rules, and escalated before options disappeared.

North Valley had not created a surplus of beds. It had created a reserve of attention, information, skill, time, and authority.

That is operating resilience.

The resilient hospital is not the hospital that never becomes crowded, short-staffed, or uncertain. It is the hospital that notices earlier, speaks honestly, reorganizes intelligently, and acts together before strain becomes harm.

Executive Action Guide

At your next leadership meeting, ask:

9.       At what time yesterday did we first know today would be difficult?

10.   What did we do at that moment?

11.   Which capacity did we technically possess but could not use?

12.   Which employee or department absorbed the consequences?

13.   What threshold, authority, or reserve would have changed the result?

If the answers are unclear, begin there.

Sources and Notes

14.   Agency for Healthcare Research and Quality, “AHRQ Report Identifies Strategies to Reduce Emergency Department Boarding,” March 25, 2025; and AHRQ Summit to Address Emergency Department Boarding: Summary Report. https://www.ahrq.gov/news/newsletters/e-newsletter/951.html and https://www.ahrq.gov/sites/default/files/wysiwyg/topics/ed-boarding-summit-report.pdf

15.   KFF, “A Closer Look at the $50 Billion Rural Health Fund in the New Reconciliation Law,” August 4, 2025; and “Medicaid: What to Watch in 2026,” January 23, 2026. https://www.kff.org/medicaid/a-closer-look-at-the-50-billion-rural-health-fund-in-the-new-reconciliation-law/ and https://www.kff.org/medicaid/medicaid-what-to-watch-in-2026/

16.   Qian Cheng, Nilay Tanik Argon, Aniruddhan Ganesaraman, and Serhan Ziya, “Proactive Inpatient Bed Requests for Emergency Department Admissions,” working paper, July 2026. https://arxiv.org/abs/2607.15432. This emerging research is cited as a promising simulation result, not as settled clinical evidence.

17.   Centers for Medicare & Medicaid Services, “Calendar Year 2026 Hospital Outpatient Prospective Payment System and Ambulatory Surgical Center Final Rule,” November 21, 2025. https://www.cms.gov/newsroom/fact-sheets/calendar-year-2026-hospital-outpatient-prospective-payment-system-opps-ambulatory-surgical-center

18.   KFF, “10 Things to Know About Rural Hospitals,” April 16, 2025. https://www.kff.org/health-costs/10-things-to-know-about-rural-hospitals/