1311159339 IV Warning Signals and Transition Points Dan Beckett by DFigR9zB

VIEWS: 0 PAGES: 40

									           Dr Dan Beckett
Consultant Acute Physician
         NHS Forth Valley
   Warning signals
    ◦   Four hour emergency access standard
    ◦   ED LoS - time profiles
    ◦   Boarding
    ◦   (Cancelled elective activity)
    ◦   (Delayed discharges)
   Whole system overview
    ◦ NHSFV capacity and flow dashboard
   Elective vs Emergency imbalance
    ◦ Optimising patient flow by reducing its variability
   Four hour emergency access standard
    ◦ Useful as an indicator of whole system pressure
    ◦ Poor compliance indicates with ED overcrowding
      Associated with an increase in mortality both in
       patients admitted and patients discharged from the ED
    ◦ Limited usefulness as an early indicator of pressure
      to trigger escalation
   ED LoS distribution
    ◦ Can demonstrate pressure in the system that is not
      evident when just looking at compliance with the
      four hour emergency access standard
    ◦ ‘Crisis spike’
   ED time curve
    ◦ Useful for retrospective analysis
    ◦ Crisis spike correlates with poor performance
    ◦ Useful for proactive escalation?
      Dynamic monitoring of the proportion of patients
       leaving the ED after 210 minutes?
                                               SRI A&E attendances 08:00 - 12:00 hrs
                         5
No. of attendances



                         4

                         3

                         2

                         1

                     BUT STILL 97% COMPLIANT AT
                         0
                              20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 240 260 280 290 300


                     THIS STAGE                                       Time in ED

                                                 SRI A&E attendances 12:00 - 16:00 hrs
                     7
                                                                                                                  27%
                     6
No. of attendances




                     5
                     4
                     3
                     2
                     1
                     0
                             20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 240 260 280 290 300

                                                                     Time in ED
                                             SRI A&E attendances 16:00 - 20:00 hrs
                     7
                     6
No. of attendances




                     5
                     4                                          91%
                     3
                     2
                     1
                     0
                         20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 240 260 280 290 300


                                                                 Time in ED

                                               SRI A&E attendances 20:00 - 00:00 hrs
                     7
                     6
No. of attendances




                                                                86%
                     5
                     4
                     3
                     2
                     1
                     0
                         20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 240 260 280 290 300

                                                                 Time in ED
                                               SRI A&E attendances 00:00 - 04:00 hrs
                     7
                     6
No. of attendances




                                                                79%
                     5
                     4
                     3
                     2
                     1
                     0
                         20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 240 260 280 290 300


                                                                 Time in ED

                                                SRI A&E attendances 04:00 - 08:00 hrs
                     6
No. of attendances




                     5
                     4


                                                                77%
                     3
                     2
                     1
                     0
                         20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 240 260 280 290 300


                                                                 Time in ED
   Boarders
    ◦ Different models of boarding exist
      Exclusively ‘front door’
      Exclusively ‘back door’
      Mixed model
    ◦ Irrespective of model, increasing numbers of
      boarders indicates system pressure and should be
      monitored/controlled
    ◦ Boarded patients have poor outcomes
   NHSFV capacity dashboard
   Real time information
    ◦   Pressure vs Capacity
    ◦   Admissions vs Discharges
    ◦   Emergency vs Elective
    ◦   Predicted vs Observed activity
    ◦   Whole system vs Individual patient
    ◦   Warning signals across the whole system as a
        trigger to escalation
   Competition between emergency and elective
    flow ‘silos’ can directly lead to ED
    overcrowding
   Perceived conflict between the 18 week RTT
    target and the 4 hour emergency access
    standard
   Significant variation in numbers of patients
    admitted over the week
              Hospital admissions, NHSScotland, October 2010
3000
                                                             131%
2500



2000



1500



1000



500



   0
       1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
                                             Total
              Hospital admissions, NHSScotland, October 2010
3,000
                                                              131%
2,500



2,000



1,500



1,000
                                                                                54%

 500



   0
        1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
                                     Emergency       Total
               Hospital admissions, NHSScotland October 2010
3,000
                                                              131%

        BUT YOU CAN’T
2,500




        COMPARE WEEKENDS
2,000



1,500



1,000
        AND WEEKDAYS! 54%
                                                                        3288%
 500



   0
        1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
                              Elective     Emergency       Total
               Hospital admissions, NHSScotland October 2010
3,000


                                                         46%
2,500



2,000



1,500

                                                         16%
1,000




                                                                     237%
 500



   0
        1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
                              Elective     Emergency       Total
   Elective admissions display more variability
    (artificial variability) than emergency
    admissions (natural variability)
    ◦ Counter-intuitive!
   Difficult to plan staffing levels for such high
    levels of variation (largely artificial variation)
   Invariably staffed for ‘average’ levels of
    activity resulting in periods of demand >
    capacity (leading to ED overcrowding and
    poor outcomes) and capacity > demand
    (waste of resources)
  Queue



Demand                          Capacity




          Can’t pass           time
          unused capacity
          forward to next week
                               Reducing waiting times in the NHS: is lack of
                               capacity the problem?
                               Bevan et al Clinician in Management (2004)
   Need to eliminate artificial variation and
    manage natural variation
                    Eliminating artificial variation (Mon-Fri)
                          NHSScotland October 2010
2500

                                                         14%
2000



1500



1000



500

       Artificial variability eliminated Monday-Friday
   0
       1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
                              Elective    Emergency        Total
   Reduces overall variation
    ◦ Reduces ED overcrowding
    ◦ Less waste
   Reduces patient boarding
   In 2006 the IOM published a report asking
    hospitals to use operational management
    tools (queuing theory) to address patient flow
    issues that lead to ED overcrowding
   Boston Medical Centre
    ◦ Significant problems with ED overcrowding 2003
    ◦ Emergency work more predictable and less varied
      than elective work
    ◦ Reprofiled elective cases Monday-Friday
      Subsequently eliminated all block scheduling
    ◦ Split elective and emergency surgical work
    ◦ Used queuing theory to guide resources for
      emergency work
   Boston Medical Centre
    ◦ Reduced variability in demand for surgical HDU
      beds by 55%
    ◦ Reduced nursing hours – saving $130K per annum
    ◦ Reduced cancelled/delayed surgery from 334 to 3
      (99.5%) for the same time periods April-September
      2003/2004 (pre- and post-implementation)
    ◦ Reduced ED waiting time by 50% and improved ED
      throughput by 45 minutes per patient
   Now many examples of successful
    implementation
    ◦ Cincinatti Childrens Hospital
      Weekday OR waiting time reduced by 28% (despite an
       increase in case volume of 24%)
      Weekend OR waiting time decreased by 34% despite an
       increase in volume of 37%)
      Capacity boosted by equivalent of 100 bed expansion
    ◦ Great Ormond Street Hospital
   Assign responsibility for the patient flow
    problem
    ◦ Chief Operations Officer or Vice President
   Establish a multidisciplinary team
   Collect and analyze data on bottlenecks
   Eliminate or smooth artificial variation
   Manage natural variation (queuing theory)
   www.ihoptimize.org
   Managing Capacity and Demand across the
    patient journey. Clinical Medicine 2010.
    10:1 13-15
   Winter Pressures in NHS Scotland 2008-2009.
    A report for the Emergency Access Team,
    Scottish Government
   Professor Derek Bell, Imperial College
   Professor Eugene Litvak, Institute for
    Healthcare Optimisation
   Dr Claire Gordon, NHS Lothian
   Bas Gough, Scottish Government
   Guy Blackburn, NHSFV



   Thanks for listening...

								
To top