Practical Exercises

1
Practical Exercises
2
Alzheimers Disease Intervention Program

• Intended to provide evidence-based interventions
  to a population of patients with Alzheimers
  disease
• Care to be delivered by a multidisciplinary team
  working with the clients PCP
• The PCPs are incentivized with a P4P program
• Intended outcomes include an array of clinical
  variables as well as client and family
  satisfaction
• Program, although initially grant-funded, must be
  self-sustaining

3
Suggestions

• Start small
• One clinical unit
• One district
• One providers caseload
• Perfect the process on small scale and spread
  later
• Multiple iterative PDSA Cycles (Rapid-cycle CQI )

4
Evaluate the Key Processes Using the SIPOC Diagram

• Intake process
• Intervention process
• Evaluation of program viability

5
The SIPOC Diagram
Suppliers
Customers
Process

1
1
Step 1
Step 2
Step 3
Inputs
Outputs
Step 4
Step 5
2
2
3
3
6
Intake Process
Outputs -Evaluated patients ready for
intervention -Engaged Physicians
Inputs -Clients -Solicitation letters -Evaluation
tools -Time availability -Staff
availability -Details of P4P program
7
Intervention Process
Outputs -Clinical improvements -Satisfied client
and family -Satisfied stakeholders (staff and
PCPs)
Inputs -Evaluated client -Treatment plan
8
Measuring the Intake Process
Outcome Measures -Number of clients or percentage
of patients with Alzheimers disease enrolled in
health plan who are evaluated per month -Number
of physicians or percentage of closed panel PCPs
participating in P4P per month
Process Measures -Percentage of solicited
patients agreeing to have an evaluation -Average
time between acceptance and evaluation in days,
measured monthly - Percentage of time that
physician agrees with treatment plan measured
monthly -Average time between evaluation and team
visit
9
Measuring the Intervention Process
Process Measures -Number or percentage of clients
withdrawing from program per month before
evaluation -Percent of participating clients
receiving anticholinergics -Percentage of
participating clients receiving all
evidence-based intervention measures
or -Percentage of participating clients receiving
a particular individual measure (one or
all) -Percentage of participants whose cases are
reviewed by team per month
Outcome Measures -Functional performance
score -Perceived health status score -Days
between ED visits evaluated monthly -Days between
hospital readmissions evaluated monthly -Client
and family satisfaction with program
10
Presenting You the Data

• Intake Process
• Outcome Measures
• Percentage of Potential Clients Enrolled
• Percentage of Physicians Enrolled in P4P
• Process Measures
• Percentage of Clients Agreeing to Participate
• Days Between agreement and Participation
• Days Between Agreement and First team Visit
• Intervention Process
• Outcome Measures
• Functional Performance Score
• Days Between ED Visits
• Client and Family Satisfaction
• Process Measures
• Percentage of Patients Withdrawing
• Percentage of Participants Whose Cases are
  Reviewed Monthly

11
Q1 Intake Outcomes
Month Value Median
January 2 0.125
February 5 0.125
March 10 0.125
12
Q1 Intake Outcomes
Month Value Median
January 2 0.075
February 5 0.075
March 7 0.075
13
Q1 Intervention Outcomes
Month Value Median
January 25 0.28
February 28 0.28
March 32 0.28
14
Q1 Intervention Outcomes
Month Value Median
January 23.0 23
February 20.0 23
March 25.0 23
15
Q1 Intervention Outcomes
Month Value Median
January 98 0.825
February 90 0.825
March 90 0.825
16
Q1-Q3 Intake Outcomes
Month Value Median
January 2 0.225
February 5 0.225
March 10 0.225
April 15 0.225
May 20 0.225
June 20 0.225
July 25 0.225
August 30 0.225
September 30 0.225
Test 1-Length of Run 9 data points and 9 useful
observations longest run6 data points requires
7 or more for special cause Test 2-Trend for 9
observations 5 consecutive rising data points
requirement for special cause 6 Test 3-Requires
at least 15 data points
17
Q1-Q3 Intake Outcomes
Month Value Median
January 2 0.1
February 5 0.1
March 7 0.1
April 8 0.1
May 10 0.1
June 10 0.1
July 15 0.1
August 18 0.1
September 18 0.1
Test 1-Length of Run 7 useful observations
Longest run4 requirement for is 7 for special
cause Test 2-Trend 9 observations 5
consecutive rising data points requirement for
special cause 6 Test 3-Number of runs Too few
data points to use
18
Q1-Q3 Intake Process
Month Value Median
January 10 22
February 12 22
March 11 22
April 23 22
May 22 22
June 25 22
July 24 22
August 22 22
September 23 22
Test 1-Length of Run 7 useful observations
Longest run3 requirement for is 7 for special
cause Test 2-Trend 9 observations 3
consecutive falling data points requirement for
special cause 6 Test 3-Number of runs Too few
data points to use
19
Q1-Q3 Intake Process
Month Value Median
January 10 13
February 12 13
March 15 13
April 13 13
May 10 13
June 13 13
July 14 13
August 15 13
September 14 13
Test 1-Length of Run 7 useful observations
Longest run3 requirement for is 7 for special
cause Test 2-Trend 9 observations 4
consecutive rising data points requirement for
special cause 6 Test 3-Number of runs Too few
data points to use
20
Q1-Q3 Intake Process
Month Value Median
January 85 0.9
February 88 0.9
March 82 0.9
April 90 0.9
May 90 0.9
June 95 0.9
July 93 0.9
August 95 0.9
September 95 0.9
Test 1-Length of Run 7 useful observations
Longest run4 requirement for is 7 for special
cause Test 2-Trend 9 observations 2
consecutive rising or falling data points
requirement for special cause 6 Test 3-Number of
runs Too few data points to use
21
Q1-Q3 Intervention Outcomes
Month Value Median
January 25 0.48
February 28 0.48
March 32 0.48
April 48 0.48
May 50 0.48
June 50 0.48
July 52 0.48
August 48 0.48
September 50 0.48
Test 1-Length of Run 7 useful observations
Longest run3 requirement for is 7 for special
cause Test 2-Trend 9 observations 5
consecutive rising data points requirement for
special cause 6 Test 3-Number of runs Too few
data points to use
22
Q1-Q3 Intervention Outcomes
Month Value Median
January 23 30
February 20 30
March 25 30
April 28 30
May 30 30
June 32 30
July 35 30
August 38 30
September 35 30
Test 1-Length of Run 8 useful observations
Longest run4 requirement for is 7 for special
cause Test 2-Trend 9 observations 7
consecutive rising data points requirement for
special cause 6 (SPECIAL CAUSE IS PRESENT) Test
3-Number of runs Too few data points to use
23
Q1-Q3 Intervention Outcomes
Month Value Median
January 98 0.8
February 90 0.8
March 90 0.8
April 85 0.8
May 80 0.8
June 80 0.8
July 75 0.8
August 70 0.8
September 70 0.8
Test 1-Length of Run 7 useful observations
Longest run4 requirement for is 7 for special
cause Test 2-Trend 9 observations 3
consecutive falling data points requirement for
special cause 6 Test 3-Number of runs Too few
data points to use
24
Q1-Q3 Intervention Process
Month Value Median
January 1 0.09
February 5 0.09
March 6 0.09
April 8 0.09
May 9 0.09
June 10 0.09
July 9 0.09
August 12 0.09
September 10 0.09
Test 1-Length of Run 7 useful observations
Longest run4 requirement for is 7 for special
cause Test 2-Trend 9 observations 6
consecutive rising data points requirement for
special cause 6 (SPECIAL CAUSE IS PRESENT) Test
3-Number of runs Too few data points to use
25
Q1-Q3 Intervention Process
Month Value Median
January 100 0.88
February 100 0.88
March 92 0.88
April 95 0.88
May 87 0.88
June 85 0.88
July 86 0.88
August 88 0.88
September 83 0.88
Test 1-Length of Run 8 useful observations
Longest run5 requirement for is 7 for special
cause Test 2-Trend 9 observations 3
consecutive falling data points requirement for
special cause 6 Test 3-Number of runs Too few
data points to use
26
Do We Have a Problem?
27
Q1-Q8 Intake Process
Month Value Median
January 10 10
February 12 10
March 15 10
April 13 10
May 10 10
June 13 10
July 14 10
August 15 10
September 14 10
October 14 10
November 10 10
December 8 10
January 7 10
February 8 10
March 10 10
April 15 10
May 9 10
June 8 10
July 8 10
August 6 10
September 5 10
October 6 10
November 5 10
December 6 9
Process Change
Test 1-Length of Run 20 useful observations
Longest run9 requirement for is 8 for special
cause (SPECIAL CAUSE IS PRESENT) Test 2-Trend
24 observations 4 consecutive rising or falling
data points requirement for special cause 7
Test 3-Number of runs 5 runs expected for 20
useful observations 6-15 (SPECIAL CAUSE IS
PRESENT)
28
Q1-Q8 Intake Process
Month Value Median
January 10 9
February 12 9
March 15 9
April 13 9
May 10 9
June 13 9
July 14 9
August 15 9
September 14 9
October 10 9
November 8 9
December 9 9
January 7 9
February 8 9
March 6 9
April 7 9
May 8 9
June 6 9
July 6 9
August 5 9
September 4 9
October 5 8
November 6 8
December 5 8
Process Change
Test 1-Length of Run 23 useful observations
Longest run12 requirement for is 8 for special
cause (SPECIAL CAUSE IS PRESENT) Test 2-Trend
24 observations 4 consecutive falling data
points requirement for special cause 7 Test
3-Number of runs 3 runs expected for 23 useful
observations 8-16 (SPECIAL CAUSE IS PRESENT)
29
Q1-Q8 Intake Outcomes
Month Value Median
January 2 0.38
February 5 0.38
March 10 0.38
April 15 0.38
May 20 0.38
June 20 0.38
July 25 0.38
August 30 0.38
September 30 0.38
October 35 0.38
November 38 0.38
December 42 0.38
January 40 0.38
February 43 0.38
March 50 0.38
April 55 0.38
May 62 0.38
June 64 0.38
July 70 0.38
August 83 0.38
September 88 0.38
October 92 0.4
November 95 0.42
December 94 0.43
Process Change
Test 1-Length of Run 23 useful observations
Longest run13 requirement for is 8 for special
cause (SPECIAL CAUSE IS PRESENT) Test 2-Trend
24 observations 11 consecutive rising data
points requirement for special cause 7 (SO
THERE IS SPECIAL CAUSE) Test 3-Number of runs 2
runs expected for 23 useful observations 8-16
(SO THERE IS SPECIAL CAUSE)
30
Q1-Q8 Intake Outcomes
Month Value Median
January 2 0.35
February 5 0.35
March 7 0.35
April 8 0.35
May 10 0.35
June 10 0.35
July 15 0.35
August 18 0.35
September 18 0.35
October 22 0.35
November 35 0.35
December 41 0.35
January 48 0.35
February 57 0.35
March 62 0.35
April 68 0.35
May 71 0.35
June 75 0.35
July 77 0.35
August 79 0.35
September 82 0.35
October 84 0.41
November 84 0.48
December 85 0.57
Process Change
Test 1-Length of Run 23 useful observations
Longest run13 requirement for is 8 for special
cause (SO THERE IS SPECIAL CAUSE) Test 2-Trend
24 observations 16 consecutive rising data
points requirement for special cause 7 (SO
THERE IS SPECIAL CAUSE) Test 3-Number of runs 2
runs expected for 23 useful observations 8-16
(SO THERE IS SPECIAL CAUSE)
31
Q1-Q8 Intervention Outcomes
Month Value Median
January 25 0.57
February 28 0.57
March 32 0.57
April 48 0.57
May 50 0.57
June 50 0.57
July 52 0.57
August 48 0.57
September 50 0.57
October 52 0.57
November 57 0.57
December 63 0.57
January 72 0.57
February 78 0.57
March 82 0.57
April 81 0.57
May 85 0.57
June 88 0.57
July 86 0.57
August 90 0.57
September 92 0.57
October 91 0.63
November 93 0.72
December 92 0.78
Process Change
Test 1-Length of Run 23 useful observations
Longest run13 requirement for is 8 for special
cause (SO THERE IS SPECIAL CAUSE) Test 2-Trend
24 observations 8 consecutive rising data
points requirement for special cause 7 (SO
THERE IS SPECIAL CAUSE) Test 3-Number of runs 2
runs expected for 23 useful observations 8-16
(SO THERE IS SPECIAL CAUSE)
32
Q1-Q8 Intervention Outcomes
Month Value Median
January 23 42
February 20 42
March 25 42
April 28 42
May 30 42
June 32 42
July 35 42
August 38 42
September 35 42
October 42 42
November 45 42
December 42 42
January 53 42
February 57 42
March 63 42
April 62 42
May 75 42
June 78 42
July 82 42
August 83 42
September 93 42
October 97 45
November 95 53
Process Change
Test 1-Length of Run 22 useful observations
Longest run11 requirement for is 8 for special
cause (SO THERE IS SPECIAL CAUSE) Test 2-Trend
24 observations 7 consecutive rising data
points requirement for special cause 7 (SO
THERE IS SPECIAL CAUSE Test 3-Number of runs 3
runs expected for 21 useful observations 7-15
(SO THERE IS SPECIAL CAUSE)
33
Q1-Q8 Intervention Outcomes
Month Value Median
January 98 0.84
February 90 0.84
March 90 0.84
April 85 0.84
May 80 0.84
June 80 0.84
July 75 0.84
August 70 0.84
September 70 0.84
October 74 0.84
November 75 0.84
December 77 0.84
January 80 0.84
February 82 0.84
March 84 0.84
April 87 0.84
May 89 0.84
June 88 0.84
July 90 0.84
August 93 0.84
September 91 0.84
October 92 0.84
November 93 0.84
December 95 0.84
Process Change
Test 1-Length of Run 23 useful observations
Longest run10 requirement for is 8 for special
cause (SO THERE IS SPECIAL CAUSE) Test 2-Trend
24 observations 9 consecutive rising data
points requirement for special cause 7 (SO
THERE IS SPECIAL CAUSE Test 3-Number of runs 3
runs expected for 21 useful observations 7-15
(SO THERE IS SPECIAL CAUSE)
34
Some Final Comments

• Since the program and the processes were new,
  there were no historical data with which to
  compare
• Once the program is running, further
  interventions should be undertaken (only) after
  the process under evaluation is in control
• In the course of the process improvement, there
  is a role for descriptive statistics to help you
  understand your data

35
Using Descriptive Statistics
0.98
0.9
0.9
0.85
0.8
0.8
0.75
0.7
0.7
Client Satisfaction
Point Column1 Rank Percent
9 0.98 1 100.00
7 0.9 2 75.00
8 0.9 2 75.00
6 0.85 4 62.50
4 0.8 5 37.50
5 0.8 5 37.50
3 0.75 7 25.00
1 0.7 8 0.00
2 0.7 8 0.00
Mean 0.82
Standard Error 0.0320156
Median 0.8
Mode 0.9
Standard Deviation 0.0960469
Sample Variance 0.009225
Kurtosis -0.86634
Skewness 0.2416068
Range 0.28
Minimum 0.7
Maximum 0.98
Sum 7.38
Count 9
       
  0.75      
0.6 0.65 0.7   0.75 0.8 0.85 0.9 0.95 1
Box and Whisker Diagram
36
Using Descriptive Statistics
r0.859776
37
Modifications Following Practice Change Fellows
Meeting
38
New Disease Management Program for CHF

• Our CHF outcomes are poor
• Risk-adjusted mortality 22
• ALOS 8 days
• 30 day readmits to our facility 10
• There are new innovations in treatment to be
  introduced to our patients
• We are going to design a program to upgrade care
  and improve outcomes

39
The SIPOC Diagram
Suppliers
Customers
Process

1
1
Step 1
Step 2
Step 3
Inputs
Outputs
Step 4
Step 5
2
2
3
3
40
The Process From 30,000 Feet
Process
Patient Referred
Patient Evaluated
Intervention Begun
Periodic Reevaluation
Modify Intervention
Supplier -Referring Physician
Inputs -Patient -Records
Outputs Clinical outcomes Improved health-related
QOL
Customers Patient and family