Camden Coalition of Healthcare Providers Evaluating
Apr 02, 2015
Camden Coalition of Healthcare Providers
Evaluating
Camden Coalition of Healthcare Providers
Agenda:1. Why is evaluating difficult?2. Developing an evaluation framework3. The evaluation timeline
Camden Coalition of Healthcare Providers
§1 Why isevaluating
difficult?
Camden Coalition of Healthcare Providers
having data
Camden Coalition of Healthcare Providers
patients w/ complexities = complex intervention
Camden Coalition of Healthcare Providers
finding an appropriate
comparison group
~~
Camden Coalition of Healthcare Providers
regression to the mean
Camden Coalition of Healthcare Providers
regression to the mean
Camden Coalition of Healthcare Providers
12 12
3
4567
8
9
1011
it takes time!
Camden Coalition of Healthcare Providers
§2 Developingan evaluation
framework
Camden Coalition of Healthcare Providers
Camden Coalition of Healthcare Providers
Camden Coalition of Healthcare Providers
Camden Coalition of Healthcare Providers
Process Outcomes
Camden Coalition of Healthcare Providers
Customized Data System Solutions
Camden Coalition of Healthcare Providers
Camden Coalition of Healthcare Providers
Link2Care Dashboard 2 Week Ending 7/26/14
Home Visit in 72 HoursAll Enrolled Patients
PCP in 7 DaysAll Enrolled Patients
Current CMI Panel Size
Graduated
Total Enrolled
0
5
10
15
20
25
30
Days Since Last Patient Encounter
Link2Care Dashboard 2 Week Ending 7/26/14
Home Visit in 72 HoursAll Enrolled Patients
PCP in 7 DaysAll Enrolled Patients
61
Current CMI Panel Size
Graduated
Total Enrolled
Days Since Last Patient Encounter0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
Running Sum of Currently Enrolled
Removed 'Lossed to Fol-low-up'
Running Panel Size
Dashboards
Camden Coalition of Healthcare Providers
Actions Taken • Weekly scorecard to measure success and failure
• Daily operations dashboard to identify opportunities
• Weekly root cause deep dives by management team
Problem The stated goal of our intervention was to see our patients once a week in person. Too many failures were occurring and our relationship with patients was diluting
Results• Reduced median days since last engagement
• Dramatically reduced process variability & outliers
WEEK 1 WEEK 10
168 DAYS
38 DAYS
Days Since Last In-Person Engagement
Camden Coalition of Healthcare Providers
Time N Minimum Maximum Mean Standard DeviationPre 186 2 10 2.9 1.43Post 186 0 11 1.3 1.77
October-1
2
November-12
December-1
2
January-13
Febru
ary-13
March-13
April-13
May-13
June-13
July-13
August-13
September-13
October-1
3
November-13
December-1
30.00
0.50
1.00
1.50
2.00
2.50
Number of 6x month readmissionsby enrollment date
Camden Coalition of Healthcare Providers
§3 Choosingan appropriate
evaluation timeline
Camden Coalition of Healthcare Providers
Planning &Data Analysis
Piloting and Early Evaluation
More RobustEvaluation (throughRandomization)
Scaling
Program Timeline
Camden Coalition of Healthcare Providers
80%
Medical Studies
Camden Coalition of Healthcare Providers
20%
Healthcare Delivery Studies
Camden Coalition of Healthcare Providers
KeyIssues
Sample Size
Number of treatment arms
Randomizing before or after consent?
Data (what’s collected administratively?-available for the control group?
When? Timing – not too early and not too late
Time, expertise, and funding to do it right
Why? Clear, credible results on causal effects
Helpful in attracting sustainable funding and scaling a program
Camden Coalition of Healthcare Providers
Standard ofCare
Key Outcomes: reduced re-hospitalizations and ED visits in 12 month period following discharge
CCHP’s Care Management RCT
n = 138 / 800