1: A Faster Hospital in Five Days

With all the hoopla about healthcare reform, there’s one huge missing piece—healthcare is going to have to get dramatically faster, better, and cheaper to help pay for the changes. Each of the nation’s 5,700+ hospitals must find ways to cut millions of dollars in unnecessary costs over the next decade. This may sound difficult considering that half of all hospitals lose money. Most hospitals exist on less than a 5 percent margin. But Lean can help hospitals to start getting faster, better, and cheaper in just a few days. And we don’t have to look any further than Virginia Mason Medical Center to discover the power of Lean to transform healthcare into a patient-centered model that delivers better outcomes and more profit.


One of the key principles of Lean thinking is to eliminate delays that consume up to 95 percent of the total cycle time (57 minutes per hour). If you’ve ever been a patient in a hospital emergency room (ER) or nursing unit bed, you know that there are lots of delays. Over the years, healthcare has made tremendous strides in reducing cycle time in various aspects of care. Outpatient surgeries are one example: arrive in the morning, and leave in the afternoon. No bed required. But there is still lots of room for improvement.


GOAL: ACCELERATE THE PATIENT’S EXPERIENCE OF HEALTHCARE



In any given “factory,” there are two kinds of time: work time during a process when actual work is occurring and elapsed time —the total time a process takes (work time plus any time spent on handoffs, waiting, batches, backlog, and so on).


—KEN MILLER


Over the last decade, I’ve consulted with many hospitals on all kinds of projects. Perhaps the most powerful tool that can be applied immediately to start slashing cycle times, medical mistakes, and cost is Lean. And it doesn’t have to take weeks, months, or years. With the right focus and the right people in the room, it only takes a few hours to a few days to find ways to speed up any healthcare process, which, in turn, will reduce errors and boost profits.


Every hospital seems to have the same problem: patient flow. This shows up in many ways:



image  Patient dissatisfaction and physician or nurse dissatisfaction


image  Emergency departments (EDs) divert hours (ambulances diverted because of overcrowding), patient boarding in the ED, LWOBS (leaving without being seen), and four-hour turnaround times


image  Operating room (OR) delays, cancellations, and long turnaround times


image  Imaging delays, long turnaround times


image  Lab delays, long turnaround times


image  Bed management delays


image  Late discharges


image  Long patient length of stay (LOS)


image  Lost revenue






Healthcare delivery often involves complex processes that have evolved over time and that are neither patient-focused nor clinician-friendly. When systems do not work well, healthcare workers resort to creating “workarounds,” adding additional layers of “patches” and “fixes” to poorly functioning systems.


—CHRISTOPHER S. KIM, M.D.


A FASTER EMERGENCY DEPARTMENT IN FIVE DAYS


In 2009, Press Ganey found that ED turnaround times still average over four hours, basically unchanged over the last decade. In 2006, the Centers for Disease Control and Prevention (CDC) found that 40 percent of hospital EDs were overcrowded. One Harvard study found that ED wait times rose 36 percent from 1997 to 2004.


Robert Wood Johnson Hospital in Hamilton, New Jersey (RWJ), winner of the 2004 Baldridge Award, receives 50,000 patients a year. In 2004, RWJ had ED turnaround times of:



image  38 minutes for discharged patients


image  90 minutes for admitted patients


How is this possible? How did they do it? By systematically eliminating the delays between registration, triage, examination, lab, imaging, and discharge or admission/transport.


RWJ’s 15/30 Program


In 1998, RWJ offered a 15-minute door-to-nurse and a 30-minute door-to-doctor guarantee. Like Domino’s Pizza, if your nurse or doctor is late, your service is free! Patient satisfaction with the ED rose from 85 percent in 2001 to 90 percent in 2004. While payouts for this policy have been less than 1 percent of ED patients, ED visits doubled! This means that for 1 percent of the revenue, RWJ increased ED revenue by 99 percent.


And because ED visits doubled, hospital revenue increased as well. Seventy percent of hospital admissions come from the ED. Faster turnaround times enabled the hospital to grow by over 10 percent per year, requiring the addition of a new nursing wing.


Results



image  RWJ was New Jersey’s fastest-growing hospital from 1999 to 2003.


image  Mortality rates for patients with congestive heart failure (CHF) declined from 8 to 2.5 percent.


image  Infection rates for things such as ventilator-assisted pneumonia (VAP) fell from 10 per 1,000 vent-days to only 2 per 1,000 vent-days.


image  The cardiology market share rose from 20 to 30 percent.


image  The surgery market share rose from 17 to 30 percent.


image  Hospital occupancy rose from 70 to 90 percent.


image  Employee satisfaction with benefits rose from 30 to 90 percent.


image  Employee satisfaction with participation in decision making rose from 30 to 90 percent.


image  Retention of nurses rose to 98 percent.


image  Employee retention rose from 80 to 98 percent.








ED Turnaround Time


My mother called me on a Friday night saying that she had blood in her urine. My wife said, “She has a bladder infection.” Mom knew better than to go to an emergency room on a Friday night and asked me to take her the next morning. We arrived the next morning and were immediately escorted to an exam room. Two and a half hours later, after a short consult with the nurse and doctor and maybe 20 minutes of lab tests, we left with a diagnosis (bladder infection) and a prescription. That’s 2.5 hours of elapsed time with perhaps 30 minutes of actual diagnosis. The other two hours were unnecessary delay.


Studies have shown that patient satisfaction begins to decrease when ED LOS exceeds two hours. There are two populations of patients who visit the ED, so let’s separate the emergent from the nonemergent cases and look at patients who get discharged.


If it takes only a couple of minutes to see the triage nurse, a few more minutes to get registered, and a few more minutes for a doctor diagnosis, then the total time spent on any one patient is perhaps nine minutes. So why does it take most EDs over two hours to handle each patient? Sure, some patients need lab work (11 minutes) and others need radiology, but most of those tests take less than an hour. We’re still looking at 35 to 60 minutes, not two hours or more.


If we look at emergent patients, they are taken into the ED immediately without having to wait. They see the doctor immediately. Tests are done STAT. Registrations are done at the bedside. Nursing floor bed assignments take only a few minutes. Nursing reports are fast. Transport to the Intensive Care Unit (ICU), Cardiac Care, or Med/Surg floor takes only 15 to 20 minutes. These patients should fly through the ED, but they take longer than the discharged patients—two to three times longer. Sure, they have to be stabilized, but why does it take hours to get them into an assigned bed?


The answer, across the board, is delay. There is too much time between activities. The admission staff is busy, so patients have to wait. The triage nurse is busy, so patients have to wait. The ED is boarding patients who should be in a nursing unit, so patients have to wait. The ED nurse can’t reach the floor nurse to give a report, and vice versa. Neither nurse can leave to transport the patient. Beds are available but not staffed. Or patients are held in the ED until after a shift change. And so on.


Theory of Constraints in the ED


Another improvement tool is the Theory of Constraints, which focuses on optimizing bottlenecks. In every ED, the bottleneck is the triage nurse and registration. The triage nurse handles patients one at a time based on their condition. The ED can be empty, with nurses and doctors waiting, while the triage nurse processes patients one by one. One way to handle this is by “optimizing the constraint.” What if the ED nursing staff pulled patients from the waiting room to augment the triage nurse when there are empty exam rooms? What if registration was handled in the exam room using a computer on wheels? How many more patients could be handled by an ED with existing staff?


Imagine a Faster ED


Imagine an emergency room where patients walk in and something surprising happens:



  1. They use the magnetic strip on their driver’s license, insurance card, or credit card to check in and register using a kiosk. The kiosk automatically takes pictures of all these IDs and uses the data to find the patient’s medical history, validate insurance, and so on.
  2. Completing registration this way triggers a “pull” signal that brings the next nurse in the queue to collect the patient from the entry area and move him or her to an exam room.
  3. Entering the exam room and gathering the patient’s vital signs trigger a pull signal for the next ED doctor in the rotation.
  4. The doctor examines the patient with the nurse available and requests any tests or x-rays using a handheld device that kicks off the orders.

    1. The nurse draws any blood or other samples required and either (i) sends them to the lab for processing or (ii) uses point-of-care testing to get results in 11 minutes or less. (Approximately seven out of ten patients require lab work.)
    2. The nurse transports the patient to imaging, if needed. (Approximately three out of ten patients require medical imaging.)

  5. Completion of the tests triggers a pull signal to the ED doctor to collect the results, diagnose, and recommend treatment.
  6. The doctor then initiates treatment. Any “teaching” material or paperwork required is prepackaged and ready for the nurse to prepare the patient for discharge or admission.
  7. Initiating admission kicks off a pull signal for a bed in the appropriate unit. If there isn’t enough staff in that unit to handle the admission, a pull signal may request an on-call nurse to come to work.
  8. Instead of it all being done manually, as most of this is now, it’s all carefully orchestrated and technically linked to minimize all delay. Many of these activities can happen in parallel, not sequentially, as they do today.

A discharged patient is in and out in 30 minutes. An admitted patient is in a nursing unit bed in 60 minutes. Of course, there will be exceptions—a rush-hour accident may tie up one of the doctors—but most patients are discharged. Finding ways to handle them using “one-piece flow” will improve ED performance dramatically.


Simply speeding up discharge and housekeeping of nursing unit beds can alleviate boarding and overcrowding in the ED. Empowering nurses to order x-rays for possible fractures or lab tests without doctor involvement can accelerate diagnosis and treatment. Scheduling radiologists during the hours of highest trauma injuries (think rush hour and Friday/Saturday night) can accelerate ED throughput. Prepackaging common triage kits can accelerate treatment.



FASTER DOOR-TO-BALLOON (D2B OR DTB) TIME IN FIVE DAYS


The ED at UMass Memorial Healthcare reduced D2B time in cardiac catheterization patients from 180 minutes in 2004 to less than 60 minutes. To optimize D2B times, the hospital measured and optimized the four key steps: (1) door-to-electrocardiogram (ECG) completion, (2) data to diagnosis, (3) diagnosis to decision, and (4) decision to balloon. Door-to-ECG time fell to one to two minutes, which enabled the ED physician to stay in the room to diagnose, decide, and initiate a call-in to the surgical team. On-call teams were scheduled with at least one team member located within 20 minutes of the hospital. Valet parking of team cars cut five minutes off the D2B time. Electronic ECG transmission from ambulances to the ED removed additional delays, allowing patients to go directly to the cardiac catheterization lab, bypassing the ED and reducing D2B time to less than 50 minutes. These changes reduced acute myocardial infarction (AMI) mortality to 11.7 percent, significantly below the 16.6 percent national average.


Lessons learned from D2B times were applied to door-to-incision time for vascular surgery and door-to-diuretic time for CHF patients.


In 2002, 16 hospitals in Virginia improved 90-minute D2B times from 37 to 75 percent and reduced catheter complications from 3 to 1.4 percent and percutaneous coronary intervention (PCI) complications from 4.4 to 2.5 percent by 2006.


A FASTER OPERATING ROOM IN FIVE DAYS


Copenhagen University Hospital wanted to reduce the time between surgical operations. The improvement team found that too much time (60+ minutes) was spent:



image  Investigating whether the patient got the required information from the surgeon (10 minutes)


image  Unpacking individual sterile disposables (30 minutes)


image  Waiting for missing devices (five trips per surgery)


image  Waiting for the patient to regain consciousness to be transferred to recovery (20 minutes)


image  Waiting for transport to recovery (10 minutes)


With some basic analysis, the team implemented countermeasures to save 60 minutes:



image  Surgeon draws an X on patient’s wristband when the patient has been informed about the operation, allowing anesthesia to begin.


image  Prepackaged sterile disposable kits replaced individual disposables, saving two nurses and 30 minutes.


image  Standard checklists ensure that all materials are gathered before the operation starts.


image  Anesthetic depth was adjusted so that the patient wakes up when the operation is finished.


image  Hospital orderlies move patients to recovery immediately.


SSM Health Care reduced operating room turnaround times from 30 minutes down to 15.8 minutes.


FASTER MEDICAL IMAGING IN FIVE DAYS


North Shore University Hospital wanted to improve patient throughput on its CT scanners to decrease LOS and increase patient satisfaction. Average turnaround time (TAT) was 20.7 hours and varied from 8 to 34 hours. The target for improvement? Sixteen hours. Identified problem areas included



image  Manual scheduling process leading to calls from nursing units


image  Time-consuming prep and delivery of contrast medium


image  CT tech travel to requisition printer (6,480 feet per day)


image  Transporter availability and travel (432 feet per day)


After analysis of these various issues, the improvement team implemented several countermeasures:



image  The requisition printer was relocated in between the two CT scanners, saving over 6,000 feet per day of unnecessary travel.


image  A dedicated CT transporter was assigned.


image  An Excel-based schedule was maintained in imaging and was viewable by all nursing units (this reduced phone calls and cancellations owing to improper patient prep or availability).


image  Instead of a rigid schedule with no room for STAT orders, a pull system adjusted the patient transport and scan to accommodate just-in-time STAT scans.


image  Contrast preparation was reassigned to the evening shift, refrigerated, and delivered during the transporter’s morning run for inpatients.


image  One CT scanner was dedicated to complex procedures, and the second was dedicated to routine high-volume procedures to maximize patient flow.


image  Staffing was adjusted to demand.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 10, 2017 | Posted by in PHARMACY | Comments Off on 1: A Faster Hospital in Five Days

Full access? Get Clinical Tree

Get Clinical Tree app for offline access