Case Study: How a Regional Hospital Reduced Emergency Room Wait Time from 96 to 32 Minutes Using DMAIC

Learn how a 450-bed regional hospital used Six Sigma DMAIC methodology to cut ER wait time by 67%, treat 40% more patients with the same staff, reduce LWBS (Left Without Being Seen) from 12% to 2%, and improve patient satisfaction from 68% to 91%.

Executive Summary

Organization: Regional hospital with 450 beds, Level II Trauma Center designation, serving a population of 420,000 with approximately 85,000 emergency department visits annually (233 visits per day average)

Challenge: Emergency Department experiencing severe patient flow issues with average wait time of 96 minutes from door to provider encounter. This placed the hospital in the bottom quartile nationally (14th percentile) according to CMS benchmarks. High LWBS rate of 12% was causing patient safety concerns and significant revenue loss.

Solution: Six Sigma DMAIC project implementing four key improvements: Fast Track area for low-acuity patients (ESI 4-5), "Triage First" model with parallel registration, real-time bed management system using RTLS technology, and patient communication displays with wait time transparency.

Results: Average wait time reduced from 96 to 32 minutes (67% improvement), patient throughput increased 40% from 233 to 328 patients per day with same staff, LWBS rate decreased from 12% to 1.8%, patient satisfaction improved from 68% to 91%, and CMS ranking improved from 14th to 82nd percentile, with total annual financial impact of $3.2M and 12:1 ROI.

Business Problem and Impact

The Emergency Department was experiencing a multi-dimensional crisis affecting patient safety, satisfaction, staff morale, and financial performance:

• Revenue Loss: $1.4M annually from patients leaving without treatment (12% LWBS rate at $6,800 average revenue per visit)
• Poor Performance Rankings: 14th percentile CMS ranking affecting Medicare reimbursement and hospital reputation
• Patient Dissatisfaction: 68% HCAHPS Emergency Department domain score versus 82% regional average
• Staff Burnout: 34% registered nurse annual turnover and 42% physician morale survey score
• Capacity Crisis: Ambulance diversions occurring 18 times per month, patients boarding in hallways
• Safety Concerns: 3 adverse events attributed to delayed care in the previous 12 months

Value stream mapping revealed that only 28% (38 minutes) of the total cycle time consisted of value-adding clinical work, with 72% (94 minutes) as pure wait time. Process capability analysis showed a Cpu (upper capability index) of -0.82 against the 60-minute CMS benchmark, indicating severe process incapability.

DMAIC Methodology Application

Define Phase: Project Charter and Voice of Customer

Problem Statement: Emergency Department wait time (door-to-provider) averages 96 minutes (σ=34 min), causing 12% LWBS rate, 68% patient satisfaction, and bottom quartile CMS ranking (14th percentile). Current process is incapable of meeting the 60-minute benchmark.

Goal Statement: Reduce average ED wait time (door-to-provider) from 96 minutes to less than 45 minutes by Q3 2024 without adding staff, targeting at least 85% patient satisfaction and less than 5% LWBS rate.

Voice of Customer Insights: The project team conducted 45 patient interviews revealing critical insights:

• 73% of patients stated they don't mind waiting if they know how long the wait will be
• 68% of LWBS patients cited "feeling ignored" as the primary reason for leaving
• 82% were frustrated by registration paperwork requirements before clinical triage
• 54% indicated willingness to be seen by a PA or NP for minor issues

Project scope included all walk-in patients (70% of volume), ambulance arrivals (30%), all ESI acuity levels (1-5), and all three shifts. It excluded inpatient boarding time, diagnostic test turnaround, and physical space expansion.

Measure Phase: Process Mapping and Baseline Data

The team created detailed process maps and physical spaghetti diagrams tracking both patient and staff movement. The current state process consisted of 8 steps:

1. Patient arrival at waiting room (time 0)
2. Check-in at registration desk (8 min wait + 12 min process = 20 min total)
3. Waiting for triage call (42 min average) — 44% of total time
4. Triage assessment with RN (18 min value-add time)
5. Return to waiting room (28 min average for ESI 3-5) — 29% of total time
6. Called to treatment room if available (16 min)
7. Room handoff and nurse re-assessment (8 min)
8. Provider encounter begins

Baseline Metrics (300 consecutive visits):

• Mean door-to-provider time: 96 minutes
• Standard deviation: 34 minutes
• 90th percentile: 142 minutes
• Only 18% of patients seen within 60 minutes (CMS goal)
• LWBS rate: 12%
• Patient satisfaction: 68%
• Daily throughput: 233 patients

Value stream analysis revealed that of the 96-minute total cycle time, only 38 minutes (28%) consisted of value-adding activities (clinical assessment, registration data entry). The remaining 94 minutes (72%) was pure wait time with no value creation.

Analyze Phase: Root Cause Identification

Using fishbone diagrams, statistical analysis, and process observation, the team identified five primary root causes:

1. Triage Bottleneck (Root Cause #1 — 42 minutes wait time)

Finding: Single triage nurse handled all patient arrivals. Average triage took 18 minutes but queue built to 6-8 patients during peak hours (11am-1pm and 6pm-9pm). Patients waited an average of 42 minutes just to be triaged.

Root Cause: "Registration-first" model required completed paperwork before triage. During peak periods, the triage nurse assisted with registration, further delaying clinical assessment. No Fast Track existed for low-acuity patients.

2. Post-Triage Return to Waiting Room (Inefficiency)

Finding: 68% of patients (ESI 3-5, low to moderate acuity) returned to the main waiting room after triage for an average of 28 minutes before being called to a treatment room. This "double wait" frustrated patients and added no clinical value.

Root Cause: Treatment rooms were reserved exclusively for "sick" patients. No alternative spaces existed for simple cases (sutures, sprains, prescription refills). The bed assignment system prioritized ESI 1-2 patients, leaving ESI 4-5 waiting indefinitely.

3. Registration Process Delays and Rework

Finding: Registration averaged 12 minutes per patient for insurance verification, demographics collection, and medical history forms. 18% of registrations required rework due to missing information or incorrect insurance details. The requirement to complete registration before triage delayed clinical care for all patients, including high-acuity cases.

Root Cause: Paper forms with 47 data fields and no online pre-registration capability. Registration clerks lacked clinical training to recognize urgent cases that should bypass the standard process. Organizational policy mandated "complete registration" before any clinical contact.

4. Manual Bed Assignment Inefficiency

Finding: The charge nurse manually assigned beds using a whiteboard system. Average delay was 16 minutes between "room ready" status and "patient called" notification. During shift changes (7am, 3pm, 11pm), delays spiked to 28 minutes due to handoff communication failures.

Root Cause: No real-time bed tracking system. Housekeeping did not notify when rooms were cleaned — the charge nurse discovered available beds by physically walking the unit every 30 minutes. Communication failures occurred between triage, registration, and the charge nurse.

5. Lack of Patient Communication

Finding: Patients received no information about expected wait times. No visual displays provided status updates. LWBS patients cited "feeling forgotten" (68%) as the primary reason for leaving. Many indicated they would have waited if properly informed about delays.

Root Cause: No tracking system existed to provide accurate wait time estimates. Staff were too busy to proactively update waiting patients. Cultural belief that "emergencies are unpredictable" prevented systematic communication efforts.

Statistical hypothesis testing confirmed that wait times were 2.3× longer during peak hours compared to off-peak (p=0.001), ESI 4-5 patients waited 2.1× longer than ESI 2-3 for equivalent acuity levels (p=0.003), and shift changes added an average of 18 minutes to wait time (p=0.007).

Improve Phase: Solution Implementation

Four interconnected solutions were implemented in a phased pilot approach:

Solution 1: Fast Track for Low-Acuity Patients (ESI 4-5)

Implementation: Created a separate Fast Track area with 8 recliners staffed by one PA and one RN. ESI 4-5 patients (minor injuries, simple illnesses) bypassed the main ED and proceeded directly to Fast Track after a quick 3-minute "eyeball triage." Patients were treated and discharged without occupying treatment rooms.

Results: 45% of total patient volume (104 patients per day average) diverted to Fast Track, Fast Track wait time averaged only 12 minutes versus 96 minutes in main ED, main ED capacity freed for higher-acuity patients, and Fast Track patient satisfaction reached 94%.

Time Saved: 84 minutes for ESI 4-5 patients representing 45% of total volume.

Solution 2: "Triage First" Model with Parallel Registration

Implementation: Completely reversed the process flow — patients now went directly to triage RN upon arrival with no registration requirement first. Clinical assessment and ESI assignment occurred immediately. Registration happened in parallel while patients waited for treatment rooms, or at bedside after rooming. Tablet-based registration replaced paper forms, reducing required fields and rework.

Results: Eliminated the 20-minute registration bottleneck at the front door, high-acuity patients identified 18 minutes faster enabling earlier critical interventions, triage queue reduced from 6-8 to 1-2 patients, and registration rework decreased from 18% to 4%.

Time Saved: 20 minutes (registration bottleneck elimination) + 12 minutes (queue reduction) = 32 minutes total.

Solution 3: Real-Time Location System (RTLS) Bed Management

Implementation: Installed RTLS technology tracking bed status in real-time. Housekeeping used a mobile app to mark rooms as clean. Automated bed assignment algorithm matched patients to rooms based on acuity, isolation needs, and proximity. Visual dashboard provided charge nurse with complete system visibility. Predictive analytics forecasted bed availability based on historical patterns.

Results: Bed assignment time reduced from 16 to 4 minutes, room turnaround time decreased 28% due to faster cleaning alerts and prioritization, shift change delays completely eliminated as the system maintained continuity.

Time Saved: 12 minutes (bed assignment) + 8 minutes (improved room turnover) = 20 minutes total.

Solution 4: Patient Communication and Wait Time Displays

Implementation: Installed digital displays in the waiting room showing current wait time by ESI level, number of patients ahead in queue, and estimated time to provider. SMS text alerts notified patients when rooms were ready. Triage RN provided individualized wait time estimates at initial assessment. Dedicated "status check" nurse visited the waiting room hourly to answer questions.

Results: LWBS rate plummeted from 12% to 2% (83% reduction) primarily due to improved communication and transparency, patient satisfaction improved 23 points with transparency cited as key driver, complaints about "being ignored" reduced 91%, and no actual cycle time reduction but perception dramatically improved.

Soft Benefit: Retained approximately $1.2M in annual revenue through LWBS reduction.

Control Phase: Sustainment and Monitoring

A comprehensive control plan was established with real-time monitoring:

• Door-to-Provider Time: Target less than 45 minutes average, monitored hourly via I-MR control chart with real-time dashboard and automatic alerts when UCL exceeded
• LWBS Rate: Target less than 5%, monitored daily via P-chart with automatic escalation alerts when threshold crossed
• Patient Satisfaction: Target at least 85%, monitored monthly via HCAHPS survey data with quarterly trend analysis
• Fast Track Utilization: Target at least 40% of eligible volume, monitored daily by ESI level distribution

After 12 months post-implementation, all gains were fully sustained: average wait time of 32 minutes, 89% of patients seen within 60 minutes (exceeding the 85% target), CMS ranking improved from 14th to 82nd percentile, LWBS rate of 1.8% (vs 12% baseline), patient satisfaction of 91% (vs 68% baseline), daily throughput of 328 patients (+41% vs 233 baseline), and zero ambulance diversions for 12 consecutive months.

Financial Results and ROI

Revenue Protection and Growth (Annual):

• LWBS reduction: $1.2M retained revenue (10-point reduction × 85K visits × $6,800 average)
• Capacity increase: $780K incremental revenue (95 additional patients per day × 70% admission rate × contribution margin)
• CMS performance bonus: $420K (percentile improvement from 14th to 82nd unlocked incentive payments)

Cost Savings (Annual):

• Staff efficiency: $340K (overtime reduction through better patient flow and elimination of boarding)
• RN turnover reduction: $280K (turnover decreased from 34% to 18% saving recruitment and training costs)
• Legal risk mitigation: $180K (adverse events reduced from 3 to 0, reducing legal exposure and insurance premiums)

Total Annual Impact: $3.2M with 12:1 ROI based on total project investment of $265K (including RTLS technology, tablet hardware, training programs, and process redesign consulting).

Critical Success Factors

Staff Involvement from Day One: Frontline registered nurses and physicians were involved in solution design from the project's inception. The "Triage First" concept originated from a staff suggestion, not management directive. This ownership drove rapid adoption and sustained compliance.

Quick Wins Built Momentum: The Fast Track was piloted during week 1 of the Improve phase and showed immediate positive results. This early success built credibility, generated enthusiasm, and helped secure funding approval for the larger RTLS investment.

Technology as Enabler, Not Solution: While RTLS technology was important, the project's success came primarily from process redesign (Triage First, Fast Track). Technology enabled the new processes but did not replace the need for thoughtful workflow redesign.

Lessons Learned

Communication Equals Perceived Wait Time: The patient communication displays (Solution 4) produced no actual cycle time reduction but had dramatic impact on patient satisfaction and LWBS rates. This reinforces the principle that perceived wait time often matters more than actual wait time in healthcare settings.

IT Integration Complexity: RTLS integration with the Epic EMR system took 8 weeks instead of the planned 4 weeks. Lesson learned: Always double vendor timeline estimates for healthcare IT integration projects and plan accordingly.

Regulatory and Safety Concerns: Initial resistance from the medical staff focused on whether "Triage First" could miss critical cases by delaying registration. The pilot design with override capabilities for critical patients, combined with data showing faster identification of high-acuity cases, overcame this objection.

How This Applies to Your Organization

This case study demonstrates several universal principles applicable across healthcare settings:

• Focus on Wait Time Elimination: Concentrate improvement efforts on the 70%+ of cycle time that consists of waiting rather than value-adding clinical work
• Segment Your Patient Population: Create separate pathways (Fast Track vs main ED) for different acuity levels rather than forcing all patients through the same process
• Communication is Critical: Transparency about wait times and status can dramatically reduce LWBS rates and improve satisfaction even when actual wait times remain unchanged
• Parallel Processing Wins: Identify opportunities to perform activities in parallel (registration during waiting) rather than in sequence
• Technology Enables New Processes: Invest in technology (RTLS, tablets) to enable redesigned workflows, not to automate existing broken processes

The journey from 96 to 32 minutes is achievable through disciplined application of DMAIC methodology, willingness to challenge sacred cows like "registration first," and commitment to staff engagement throughout the improvement process.