Challenges & Opportunity For

Quantitative Analysis in Hospitals

March 4th, 2014

1
Quantitative Analysis in Hospitals
• Simulation
– Monte-Carlo
– Discrete Event Simulation
– Real-time
• Queuing Theory
– M/M/s
– Reneging
– Blocking
• Linear Programming
• Approximate Dynamic Programming

2
 Quantitative Analysis in Hospitals
• LITTLE’s Law (from Queuing Theory) for capacity
management
– Av # Beds = Av Admission Rate * ALoS
– Little’s Law holds true whatever the variability of
Admission Rate and LoS
– Little’s Law holds true for sub-systems within
system (i.e for QCH, OR/Surgery, OR, Surgery,
Medicine, ED, ED/Medicine,
ED/OR/Surgery/Medicine,…)
– Little’s Law doesn’t say anything about # people
waiting in queues, nor about wait times
– Note: if Little’s Law is used without a real
understanding of flow management, the results can 3
 Application Areas in Hospitals
• Patient Flow Management, across continuum of
care
• Capacity Planning and Control
– Beds
– Staffing
– Physician shift scheduling
• Master Surgery Scheduling, for elective
procedures
• Wait Time reduction
• …

4
 Main Patient Flows in QCH
DI

Lab ICU

OT, PT, RT
Emergency
ED Medicine

ALC
OR Surgery
Electives

Geriatrics
Transitio
n Home
Program
Mental
Health
Rehabilitation
Maternal &
New Born

Ambulatory
Care
 Main Patient Flows in QCH
DI

Lab ICU

Emergency OT, PT, RT

ED Medicine
200 patients per day 12 patients per day
2 OUT-patients per day 5 patients per day ALC
OR Surgery
Electives
4 IN-Patients per day
15 Day-Surgery 8 IN_AM per day
per week-day (11 IN_AM per week-day) Geriatrics
Transitio
n Home
Program
Mental
Health
Rehabilitation
Maternal &
New Born

Ambulatory
Care
 Patient Flows in OR/Surgery at QCH
• From ED
– Current classification: 1A, 1B, 1C, 1D
– OUT-patients go to Day-Surgery ward and wait for
operation
– 1A and some 1B IN-patients got directly to OR and then
are admitted to Surgery
– Other IN-patients are admitted to Surgery, and wait for
operation
– Some IN-patients are admitted to Surgery and then
discharged without going to OR
• Electives
– Day-Surgery patients are scheduled, go to OR directly,
and then to Day-Surgery ward
– IN-AM patients are scheduled, go to OR directly, and
then are admitted to Surgery
 Variability Management Methodology

Phase III – Determination of

Phase I - Separation of Phase II - Smoothing
Inpatient Bed Capacity and
Patient Flows of Patient Flows
Staffing Needs

Reduces
Improves OR, ED Competition for
and Inpatient Unit Beds, Stress, Right Sizes Units
Access and Overload and for Optimal
Efficiency by Waste by Matching of Patient
Separating managing natural Flow, Census, and
Different Types of variability and Capacity
Patient Flows eliminating artificial
variability in Flows
 Variability Management
Patient Census - 1
70

60

50

Patient Census - Unsmoothed

Patient Census

40
Average Patient Census
Bed Capacity Strategy (1990s)
30

20

10

0
Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day Day Day Day Day
10 11 12 13 14

9
 Variability Management
Patient Census - 2
70

60

50

Patient Census - Unsmoothed

Patient Census

40
Average Patient Census
Bed Capacity Strategy (now)
30

20

10

0
Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day Day Day Day Day
10 11 12 13 14

10
 Variability Management
Patient Census - Smoothed
70

60

50

Patient Census - Unsmoothed

Patient Census

40 Patient Census - Smoothed

Average Patient Census
Bed Capacity
30

20

10

0
Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day Day Day Day Day
10 11 12 13 14

11
NHS England Study

12
NHS England Study

13
Variability in Utilization

14
 Variability and Wait Times

Average Fixed Ave # Ave wait Average Average Ave # Ave wait
Arrival Service people time per Arrival Service people time per
Rate Rate waiting person Rate Rate waiting person
0.5 1.0 0.3 0.5 0.5 1.0 0.5 1.0
0.6 1.0 0.5 0.8 0.6 1.0 0.9 1.5
0.7 1.0 0.8 1.2 0.7 1.0 1.6 2.3
0.8 1.0 1.6 2.0 0.8 1.0 3.2 4.0
0.9 1.0 4.1 4.5 0.9 1.0 8.1 9.0

15
 Types of Variability in Hospitals
• Natural Variability
– Flow Variability: Patient Arrivals in ED (random)
– Clinical Variability: Patient Acuity
– Professional Variability: “Slow” vs “Fast” Provider
• Artificial Variability
– Batching
– Sub-optimal Scheduling
– Some Provider practices
– Inefficient Hospital Policies
• Most of the time, counter-intuitively, Natural
Variability is not as bad as Artificial Variability, and
is more predictable
• Solution is to manage Natural Variability and
16
remove Artificial Variability
 Tools for Variability Management
• Manage Natural Variability
– Queuing Theory
– Simulation
• Remove Artificial Variability  Scheduling
– Linear Programming
– Approximate Dynamic Programming
– Simulation

17
 Challenges
• Hospital units are working in silos  unclear
view of end-to-end optimization in care delivery
• Heavy cognitive load on Physicians and Nursing
who are already overloaded with work, constantly
interrupted in their tasks, multi-tasking, etc
• Operations Research tools are not much known in
Healthcare, even though they have been used
pervasively in other industries for decades
• …

18
Major Opportunity in Canadian Hospitals
• Use Quantitative Analysis together with LEAN
– LEAN is becoming a major thrust because of
necessity for process improvements
– Burning Platform from MOHLTC (similar to
“ObamaCare” in US hospitals)
– Patient & Family Centric care

19
 LEAN is…
• A set of Tools & Techniques
– Waste Removal
– Continuous Improvements,
PDSA
– Value Stream Mapping, Voice of
Customer
– Kaizen events
– Visual Management, Spaghetti
diagram,…

20
 LEAN is also…

• A Strategy Deployment System

– Alignment with Strategic Plan
– Framework for Change
Management: Culture,
Behaviors, Engagement
– Value Stream Management,
with PULL from Customer
– Guidelines for Leadership
engagement at all levels
– Concept of Breakthrough
system projects
– Framework for Sustainability
21
Top 10 reasons why LEAN Transformations fail

22
 LEAN + Quantitative Analysis @ QCH
• Expand LEAN with the addition of Quantitative
Analysis tools
– Linear Programming
– Queuing Theory
– Discrete Event Simulation
• Patient Flow Transformation project across ED,
Internal Medicine, OR/Surgery
– Streamline the patient flows from ED to Internal
Medicine and OR/Surgery
– Optimize the bed capacity in Internal Medicine
– Optimize the scheduling of Electives in the OR in
order to right-size the bed capacity in Surgery
23
Thank You!

24