cd

Document Sample
cd Powered By Docstoc
					Christy Dempsey, RN MBA CNOR
              August 26, 2009
   Understand how other areas of the hospital
    directly impact the flow of patients in the ED
   Demonstrate how queuing analysis and
    simulation modeling can be employed inside
    and outside the ED to improve flow and
    increase capacity without building
    infrastructure or hiring more staff
   Learn how other organizations have used this
    information and methodology for significant
    and sustainable results
   "Hospital chief executive officers should
    adopt enterprise-wide operations
    management and related strategies to
    improve the quality and efficiency of
    emergency care.‖
   ―By smoothing the inherent peaks and valleys
    in patient flow, and eliminating the artificial
    variabilities that unnecessarily impair patient
    flow, hospitals can improve patient safety and
    quality while simultaneously reducing
    hospital waste and cost.‖
   ED and PACU boarding/overcrowding
   Staff shortages - nursing and physician
   The “any bed available” phenomenon
   Quality concerns related to nurse:patient ratios, medical
    errors, and adverse events
   Frustration due to unpredictable schedules and inability
    to care for patients the way physicians want to care for
    them
   Increasing workloads and decreasing reimbursement
           ED
  OR


       Direct
       Admits     Which do we have
                  the most control over??




Hospital Census
   NO
    ◦ These are usually sick patients
    ◦ Sent from physician office
    ◦ May be scheduled through Cath Lab or other
      procedural area – higher risk patients
    ◦ Random arrivals
                 NO

                        Elective Vs. Emergent Daily Admissions

                  70
                  60
                                                              # of Emergent
# of Admissions




                  50                                          Admissions
                  40                                          Average Emergent
                  30
                  20                                          # of Elective
                                                              Admissions
                  10
                   0
                   Monday   Tuesday    Wednesday   Thursday
                                      Day
   ―ED overcrowding is caused by a complex set
    of conditions that occur across hospital units
    and across the entire health care system.
    Inability to move admitted patients from the
    ED to the appropriate inpatient unit stands
    out as a major driver of ED overcrowding.‖

Emergency Department Utilization and
  Capacity
July 2009
   YES!!
    ◦ Variability in the elective surgery schedule is the
      culprit
    ◦ Totally schedulable
    ◦ Totally within our control
    ◦ Peaks and valleys in the elective schedule result in
      peaks and valleys in inpatient census
   Boarding
    ◦ ED
    ◦ PACU
    ◦ OR
   Inappropriate patient placement
    ◦ Any bed available
   Increased length of stay
   Increased risk of morbidity/mortality
   Increased risk of adverse events
   Physicians
    ◦ Frustration due to unpredictable schedules
    ◦ Rounding in multiple locations
    ◦ Long waits to do cases – elective and non-elective
    ◦ Frequent phone calls from nurses unaccustomed to
      care for their patients
    ◦ Longer lengths of stay result in increased risk of
      complications, infection, adverse events
    ◦ Inability to grow, practice and revenue implications
   Unpredictable schedules
    ◦ OT
    ◦ Low workload days
    ◦ Staffing unfamiliar cases
   Equipment competition
   Recruitment and retention issues
   Training issue for downstream nursing units
   Overcrowding
   Boarding
   Diversions
   Safety
   Quality
   Liability
   Burnout
   Recruitment/Retention
   Lower overall utilization despite overcrowding
   Loss of contracted payors
   Liability
   Reduced reimbursement – medical errors,
    never events, boarding
   Capital constraints
   Duplication of human and material resources
    during peaks
   Wasted human and material resources during
    valleys
   Recognize flow is an organizational issue
      ED is at the mercy of the inpatient census
   Manage uncontrollable variability
      ED admissions
   Reduce/eliminate controllable variability
      Smooth elective hospital admissions
   Assure transparent and credible data
   Involve physicians
   Make progress
    The Cause?




Controllable (Artificial) Variability
        Types of Variability


 Uncontrollable (Natural)
   – Random but often predictable
   – Manageable but cannot be eliminated
   – Example: emergent/urgent ED volume

 Controllable (Artificial)
   – Non-random
   – Caused by management practices such as
     scheduling, staffing practices
   – Example: elective surgery schedule
 Combine the hard science of rigorous data
  collection and analysis with the soft science
  of change management and operations
  expertise
 Collaboration between physicians and
  hospital leadership
 Culture must change—if you always do what
  you’ve always done, you’ll always get what
  you’ve always gotten!
                                                                  Count




                                              10
                                                   20
                                                        30
                                                             40
                                                                          50
                                                                               60
                                                                                    70
                                                                                         80
                                                                                              90




                                          0
                           Tue,1/2/07

                           Thu,1/4/07

                          Mon,1/8/07

                         Wed,1/10/07

                          Fri, 1/12/07

                         Tue,1/16/07

                         Thu,1/18/07

                         Mon,1/22/07

                         Wed,1/24/07

                          Fri, 1/26/07

                         Tue,1/30/07

                           Thu,2/1/07




Add-On
                          Mon,2/5/07

                          Wed,2/7/07

                            Fri, 2/9/07




Add-On Mean
                         Tue,2/13/07




                 Da te
                         Thu,2/15/07

                         Mon,2/19/07




Scheduled
                                                                                                        WellStar Kennestone Hospital




                         Wed,2/21/07
                                                                                                   Non-Holiday Weekdays, 1/2/2007-3/30/2007




                          Fri, 2/23/07
                                                                                                   Add-On and Scheduled OR Cases By Date




                         Tue,2/27/07

                           Thu,3/1/07
Scheduled Mean




                          Mon,3/5/07

                          Wed,3/7/07

                            Fri, 3/9/07

                         Tue,3/13/07

                         Thu,3/15/07
                                                                                                                                              Real-Life Variability in the OR




                         Mon,3/19/07

                         Wed,3/21/07

                          Fri, 3/23/07

                         Tue,3/27/07

                         Thu,3/29/07
Inappropriate Patient Placement


 Destination Units for Post-op Patients from PACU:
            Orthopedic Inpatients Only
    Pre-Project: WellStar Kennestone Hospital


                    Missing
                      6%



                Other
                 9%
           7W
           8%


           7S
          11%
                              7N (Ortho)
                                 66%
         Three Typical “Fixes”




 Build and staff to peak demand in EDs, ORs and
  in downstream units; tolerate overspending on
  staff and material expenses, underutilization
  during non-peak times
 Staff below the peaks; tolerate ED diversions,
  nursing overloading and medical errors
 Staff for averages and try to flex up or down to
  manage unpredictable demand; tolerate the
  same negative effects
The Real Solution




Smooth artificial variability and provide
   resources to meet patient-driven
(vs. schedule-driven) peaks in demand




             3-step process
                  Step 1


Step 1
Separate Scheduled from Unscheduled
OR Flow

Step 1 implementation
• Collect and analyze data on emergent/urgent (add-on)
  cases, including arrival patterns and urgency
• Apply queuing theory to determine capacity needed to
  accommodate add-on cases within clinically acceptable
  wait times
• Adjust plan based on physician and hospital input
• Allocate resources to meet the separate demands of
  scheduled and unscheduled volumes
                      Step 2


Step 1                                     Step 2
Separate Scheduled from Unscheduled    Smooth flow of
OR Flow                               scheduled patients


Step 2 implementation
• Evaluate daily case variation in scheduled cases by surgical
  service as well as by destination units
• Work in collaboration with surgical practices to redesign
  the OR schedule to smooth daily case volume based on
  destination unit
• Smoothing should take into account clinic schedules,
  surgeons’ teaching and other responsibilities, hospital case
  mix, and size of destination units
                     Step 3


                                                              Step 3
Step 1                                     Step 2
                                                           Determine Bed
Separate Scheduled from Unscheduled    Smooth flow of
                                                                 and
OR Flow                               scheduled patients
                                                           Staffing needs



   Step 3 implementation
   • Apply simulation models to determine the number of beds and
     staff needed to achieve a desired level of service
   • Maximize throughput by streamlining the discharge process and
     addressing length of stay issues
   • Implement process improvement in downstream units including
     admission and discharge processes, ED specific flow
     improvements, hospitalist and medicine specific flow
     improvements
Queuing Theory
   Mathematical tool used to determine
    capacity needed to handle random
    arrivals with constrained resources
   Used in industry since the early 1900’s
   Relevance to improving patient flow
    newly recognized
   Can be applied to any procedural area
    with a mix of elective and add-on cases
   Arrivals are random – ED volume and
    urgent/emergent OR cases
   Average service time - ED visit lengths or
    urgent/emergent surgical case duration
    + room turnover time - can be calculated
   Number of servers (ED treatment rooms/
    physicians, OR, cath labs) is limited
   Optimum number of treatment or operating
    rooms for add-on (urgent/emergent) cases

   Optimum number of ED physicians

   Average wait time by triage or urgency class

   Percent of time an ED, ED physician, or OR
    will be available immediately for an
    emergency patient

   Utilization rates of ORs and ED rooms or
    physicians
   Patient arrivals

   Triage level of patient arrivals


   Average visit length – door to door
   Inputs
    ◦   Arrival rates by hour
    ◦   Acuity of arrivals
    ◦   Average service rate
    ◦   Room turnover between patients
    ◦   Staffed shifts
    ◦   Desired waiting times
   Outputs
    ◦ Waiting time for each acuity
    ◦ Utilization rates of rooms
    ◦ Outputs by shift
           # Classes       5       5       5       5       5       5
          Start Class      A       A       A       A       A       A
           End Class       E       E       E       E       E       E
          # Servers       40      41      42      43      44      45
             Day Type    WD      WD      WD      WD      WD      WD
          Start Time       7       7       7       7       7       7
             End Time     15      15      15      15      15      15
                 TOT       5       5       5       5       5       5



Results
Service Rate            0.335   0.335   0.335   0.335   0.335   0.335
Arr Rate 1               0.14    0.14    0.14    0.14    0.14    0.14
Arr Rate 2               1.89    1.89    1.89    1.89    1.89    1.89
Arr Rate 3               6.17    6.17    6.17    6.17    6.17    6.17
Arr Rate 4               3.19    3.19    3.19    3.19    3.19    3.19
Arr Rate 5               0.47    0.47    0.47    0.47    0.47    0.47
Wait 1(Immediate)         4.5     4.4     4.3     4.2     4.1      4
Wait 2 (Emergent)         5.3     5.2      5      4.9     4.7     4.6
Wait 3 (Urgent)          13.5    12.7    11.9    11.2    10.6     10
Wait 4 (Less Urgent)     76.3     63     53.2    45.8     40     35.4
Wait 5 (Non-urgent)     256.7   185.4    141    111.4    90.7    75.5
% Avail                  0.33     0.4    0.46    0.51    0.57    0.62
Util %                  88.46    86.3   84.24   82.28   80.41   78.63
           # Classes        5        5       5       5       5       5
          Start Class       A        A       A       A       A       A
           End Class        E        E       E       E       E       E
          # Servers        40       41      42      43      44      45
             Day Type      WD       WD     WD      WD      WD      WD
          Start Time        3        3       3       3       3       3
             End Time      11       11      11      11      11      11
                 TOT        5        5       5       5       5       5



Results
Service Rate             0.335    0.335   0.335   0.335   0.335   0.335
Arr Rate 1                0.17     0.17    0.17    0.17    0.17    0.17
Arr Rate 2                2.62     2.62    2.62    2.62    2.62    2.62
Arr Rate 3                6.11     6.11    6.11    6.11    6.11    6.11
Arr Rate 4                3.95     3.95    3.95    3.95    3.95    3.95
Arr Rate 5                0.39     0.39    0.39    0.39    0.39    0.39
Wait 1                     4.5      4.4     4.3     4.2     4.1      4
Wait 2                     5.7      5.5     5.4     5.2     5.1     4.9
Wait 3                    16.8     15.5    14.4    13.5    12.6    11.9
Wait 4                   319.8    190.5   132.6   100.1    79.5    65.3
Wait 5                  8590.7   1833.9   819.2   470.4   308.1    219
% Avail                   0.03     0.09    0.15    0.21    0.26    0.32
Util %                   98.75    96.34   94.05   91.86   89.77   87.78
             # Classes      5       5       5       5       5
          Start Class       A       A       A       A       A
             End Class      E       E       E       E       E
          # Servers        18      19      20      21      22
             Day Type     WD      WD      WD      WD      WD
          Start Time       11      11      11      11      11
             End Time       7       7       7       7       7
                  TOT       5       5       5       5       5


Results
Service Rate             0.335   0.335   0.335   0.335   0.335
Arr Rate 1                0.09    0.09    0.09    0.09    0.09
Arr Rate 2                1.26    1.26    1.26    1.26    1.26
Arr Rate 3                2.28    2.28    2.28    2.28    2.28
Arr Rate 4                 0.8     0.8     0.8     0.8     0.8
Arr Rate 5                0.05    0.05    0.05    0.05    0.05
Wait 1                     9.9     9.3     8.8     8.4      8
Wait 2                    12.7    11.8    11.1    10.4     9.8
Wait 3                    31.5    27.2    23.8    21.2     19
Wait 4                     92     70.3     56     46.1    38.9
Wait 5                   142.4   101.9    77.5    61.4    50.3
% Avail                   2.04    2.34    2.61    2.86    3.08
Util %                   74.25   70.34   66.83   63.64   60.75
   Arrival patterns change
    ◦ New hospital or closure of an ED increases volumes
    ◦ Flu season


   Treatment times change
    ◦ Additional physician or nursing staff
    ◦ Reduction of boarding allows for a reduction in
      average treatment time
      # Urgency Classes (ESI groups) Included         5          5          5          5          5          5

                                  Start Class         1          1          1          1          1          1

                                    End Class         5          5          5          5          5          5

                        # Treatment Rooms            40         41         42         43         44         45

                                    Day Type        WD         WD         WD         WD         WD         WD

                                   Start Time       11p        11p        11p        11p        11p        11p

                                    End Time         7a         7a         7a         7a         7a         7a

                                 Service Rate   150 mins   150 mins   150 mins   150 mins   150 mins   150 mins

                                         TOT          5          5          5          5          5          5

                                   Arr Rate 1       0.17       0.17       0.17       0.17       0.17       0.17

                                   Arr Rate 2       2.62       2.62       2.62       2.62       2.62       2.62

                                   Arr Rate 3       6.11       6.11       6.11       6.11       6.11       6.11

                                   Arr Rate 4       3.95       3.95       3.95       3.95       3.95       3.95

                                   Arr Rate 5       0.39       0.39       0.39       0.39       0.39       0.39



Results
                           Wait 1-Immediate          3.9        3.8        3.7        3.6        3.5        3.5

                            Wait 2-Emergent          4.8        4.6        4.5        4.3        4.2        4.1

                              Wait 3-Urgent        11.1       10.4        9.8        9.2        8.8        8.3

                        Wait 4-Less Urgent         53.3         45       38.7       33.8       29.8       26.7

                           Wait 5-Not Urgent      156.5      119.2        94.3       76.8        64        54.3

                                     % Avail        0.43        0.5       0.56       0.61       0.67       0.72

                                       Util %     85.51      83.42      81.44      79.54      77.73      76.01
   Trade-offs between waiting time and
    resources applied


   Hard science vs soft science balance
   Active involvement by project committee of
    physician leaders, top hospital management

   Timely review of questionable urgency/acuity
    classifications

   Performance monitoring
   Average wait time by triage/urgency class
   Compliance with maximum wait time by
    triage/urgency class
   Treatment room/physician utilization
   Availability of a room when a level one
    (emergency) case arrives
   Boarding days/times
   Appropriate patient placement in downstream units
   Frequency of Re-evaluation
    ◦ Quarterly under normal circumstances
    ◦ Immediately if major issues
   Triggers for Change
    ◦ Non-elective volume increases or decreases
    ◦ New services or surgeons with non-elective cases
    ◦ Expansion or contraction of ED or OR capacity
   Trade-offs related to Changes
    ◦ Staff availability
    ◦ Resource constraints
                                  Demonstrated Results:
                                  Physician Satisfaction


Ease of Admitting                             Physician Satisfaction Increase
Patients Ease of admitting                                                        7.1
                 patients


                  Access to
                                                            3.9
               transcriptions*


              Sched inpatient
                                                          3.6
               tests/therapy

             Sched outpatient
                                                      3.4
              tests/therapy*


                   Pharmacy*                          3.3


       Info re hos changes infl
                                                    3.0
             pers pract*

                                  0.0   1.0   2.0     3.0       4.0   5.0   6.0    7.0   8.0   9.0   10.0


             The most noticeable shifts (scores which changed >= +/- 3.0
             points) tended to involve patient flow issues.
                  Reduction of Variability
Highest volume day (69 cases) is 1.6 times the lowest volume day
(42 cases) vs. Substantial variability in elective surgery cases
before: highest volume day (82 cases) is 2.15 times the lowest
volume day (38 cases)
                                                      BEFORE




                                                      AFTER




                                                       Wellstar-Kennestone
   http://www.rwjf.org/pr/product.jsp?id=45929&c
   http://www.acep.org/uploadedFiles/ACEP/Membersh
    ip/Sections_of_Membership/intnatl/news/2008Boardi
    ngReport.pdf
   http://www.referenceforbusiness.com/encyclopedia/
    Pro-Res/Queuing-Theory.html
   http://www.hhnmag.com/hhnmag_app/jsp/articledis
    play.jsp?dcrpath=HHNMAG/Article/data07JUL2008/0
    80715HHN_Online_Eitel&domain=HHNMAG
   Litvak E, Long MC, Cooper A, McManus M. Emergency
    department diversion: Causes and solutions.
    Academic Emergency Medicine. 2001;8(11):1108-
    1110.
Christy Dempsey, RN MBA CNOR
SVP of Clinical Operations
Press Ganey Associates, Inc
417-877-7666
cdempsey@pressganey.com

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:7
posted:10/24/2011
language:English
pages:48