Welcome to the
New England QIN-QIO
Using Data to Improve Care
Transitions and Reduce
Readmissions
Thank you for joining. Our presentation will begin shortly.
If you haven’t already, please dial in to the audio line:
888-895-6448 Passcode: 519-6001
Using Data to Improve Care
Transitions and Reduce
Readmissions
Rebekah Gardner, MD
Blake Morphis Healthcentric Advisors, New England QIN-QIO
May 22, 2017 This material was prepared by the New England QIN-QIO, the Medicare Quality Innovation Network-Quality Improvement Organization for New England, under
contract with the Centers for Medicare & Medicaid Services (CMS), an agency of the U.S. Department of Health and Human Services. The contents presented
do not necessarily reflect CMS policy. CMSMA_C3-1_050117_0993
Sandra Parker Vice President & General Counsel
3
Sandra Parker is Vice President & General Counsel for the Maine Hospital Association. She is also responsible for policy analysis and development for key health policy issues, such as quality of care and patient safety. Sandra is the state lead for the Centers for Medicare & Medicaid Services’ Hospital Improvement Innovation Network contract with the American Hospital Association’s Health Research & Educational Trust.
.
QIN-QIO Hospital Reports
4
Enhancements
• Who
• What
• When
• Where
5
Rebekah Gardner, MD
Senior Medical Scientist, New England QIN-QIO
Attending Physician, Rhode Island Hospital
Associate Professor of Medicine, Brown University
Blake Morphis
Manager of Analytic Services
New England QIN-QIO
Healthcentric Advisors
Maureen Leary
Senior Program Coordinator
New England QIN-QIO
Healthcentric Advisors
Today’s speakers
Inform your quality
improvement efforts
6
Perform root cause
analysis
Identify readmission
drivers
Select and implement
intervention
Track process and outcome
data
Determine effectiveness
The purpose of these reports
• Illuminate opportunities
at your hospital
• Identify readmission
patterns specific to
your community
• Guide intervention
selection and focus
7
Build the full picture
8
1. Analyze available data
2. Capture input from frontline
staff, patients, caregivers,
3. Gather community
information and input
Build the full picture
9
1. Analyze available data
2. Capture input from frontline
staff, patients, caregivers,
3. Gather community
information and input
10
Key differences
• Initially developed to support compliance
programs, prevent improper payments
• Runs on the fiscal year, not calendar year
• Different data lag, likely longer
• Separate reports for psych and critical
access hospitals – annual
Hospital Improvement
Innovation Network
11
• MHA/HRET HIIN readmission measures include
– All-cause unplanned 30-day readmissions (all payors)
– All-cause unplanned 30-day readmissions (Medicare FFS only)
• For MHA/HRET HIIN hospitals, measures are
calculated using the inpatient discharge dataset you
upload each month
CMS Goal: By September 27, 2018, each participating
HIIN hospital reduces unavoidable readmissions by at
least 12 percent
New England QIN-QIO
Hospital Reports
12
Hospital report overview
13
• Trends over time
• Readmissions, ED and obs
• Index and readmission diagnoses
• Days to readmission
• Demographics
• Disposition
Good to know
14
• Medicare fee-for-service population
• Readmissions anywhere
• Includes psych
• Not risk adjusted
• Data from claims
• Available quarterly
QIN-QIO Hospital Report
Section 1:
All Cause 30 Day
Readmissions
15
How to use these data
• Understand big picture
• Compare to state data
• Trend over time
• Track overall progress and
impact of interventions
20
QIN-QIO Hospital Report
Section 1:
Number of
Days Until
Readmission
How to use these data
• Identify when most readmissions
coming back in
• Consider potential drivers for
findings
• Target interventions to most
vulnerable time period
26
QIO-QIO Hospital Report
Section 2:
Condition-Specific
Readmissions
27
Readmission rates for
specific diagnosis categories
Acute MI Heart failure
COPD
Diabetes Pneumonia Sepsis
Based on AHRQ Clinical Classifications Software
30-day readmission measure
Index admission diagnosis
– Based on principal diagnosis at discharge
– Readmissions apply to index admission
28
30-day readmission measure
Index admission diagnosis
– Based on principal diagnosis at discharge
– Readmissions apply to index admission
Diagnosis: Acute MI
29
30-day readmission measure
Index admission diagnosis
– Based on principal diagnosis at discharge
– Readmissions apply to index admission
Diagnosis: Acute MI 30 Days
30
30-day readmission measure
Index admission diagnosis
– Based on principal diagnosis at discharge
– Readmissions apply to index admission
Diagnosis: Acute MI 30 Days Diagnosis: Pneumonia
31
30-day readmission measure
Index admission diagnosis
– Based on principal diagnosis at discharge
– Readmissions apply to index admission
Diagnosis: Acute MI 30 Days Diagnosis: Pneumonia
Readmission counts under Acute MI
32
How to use these data
• Understand more traditional
readmission drivers
• Predict HRRP penalties
• Evaluate effects of diagnosis-
directed interventions
• Identify clear outlier diagnoses
38
QIO-QIO Hospital Report
Section 3:
Top 10 Index &
Readmission
Diagnosis
Categories
Two tables
39
• Top 10 Index Admit Diagnosis Categories
– What diagnoses your patients left the hospital
with on the initial admission
• Top 10 Readmission Diagnosis Categories
– What diagnoses your patients came in with
when readmitted
Index diagnoses
40
Index diagnoses
41
Index diagnoses
42
Index diagnoses
43
Readmission diagnoses
44
Readmission diagnoses How to use these data
• Look for drivers—may not be
HRRP associated diagnoses
• Highlight behavioral health and
substance use contribution
• Notice if top 10 not a significant
proportion of all readmissions
45
46
QIO-QIO Hospital Report
Section 4:
Top 10 Index
Admission
Categories with
Associated
Readmission
Categories
Linking index diagnoses to
readmission diagnoses
47
• Takes the top 10 index admission
diagnosis categories
• Pairs them up with their associated
readmission diagnosis categories
• All based on principal diagnosis
Index linked with readmission
48
Index linked with readmission
49
Index linked with readmission
50
Index linked with readmission How to use these data
• Dive deeper into a diagnosis
• Recognize a pattern, pull charts,
interview patients
• Get curious—lots of pneumonia
coming back with UTIs? Check
out catheter use
51
52
QIO-QIO Hospital Report
Section 5:
Demographics,
Co-morbidities,
Readmitted to
Demographics
53
Demographics
54
Demographics
55
Demographics
56
Co-morbidities
57
Where readmissions occurred
58
Where readmissions occurred
59
Where readmissions occurred
60
Where readmissions occurred
61
Where readmissions occurred
62
Where Readmission Occurred How to use these data
• Identify highest risk age groups
• Look for disparities among
different populations—why?
• Notice co-morbidities associated
with high readmission rates
• Pinpoint where patients
readmitted—partnership needed?
63
64
QIO-QIO Hospital Report
Section 6:
Emergency
Department
Visits
Emergency department visits
65
• Total number of ED visits
• Percent of patients who visit the ED
within 30 days of inpatient admission
• Top 10 diagnosis categories
Emergency department visits
66
• Total number of ED visits
• Percent of patients who visit the ED
within 30 days of inpatient admission
• Top 10 diagnosis categories
ED visits that result in
admission or observation
stay not included here
Total number of ED visits
67
ED visits within 30 days of
inpatient admission
68
Top ED visit diagnoses
69
Top ED visit diagnoses How to use these data
• Broaden framework of unplanned
utilization
• Consider how the ED fits into
your transitions program
• Identify common ED diagnoses
and alignment with other efforts
70
71
QIO-QIO Hospital Report
Section 7:
Observation
Stays
Observation stays
72
• Total number of observation stays
• Percent of patients with observation
within 30 days of inpatient admission
• Top 10 diagnosis categories
Observation stays
73
• Total number of observation stays
• Percent of patients with observation
within 30 days of inpatient admission
• Top 10 diagnosis categories
Observation stays that
result in admission are
not included here
74
Top observation diagnoses
75
Top observation diagnoses How to use these data
• Complete the utilization picture
and account for trends more fully
• Analyze similarly to traditional
readmissions
• Adjust approach for ED-based
obs unit vs intermingled on floor
76
77
QIO-QIO Hospital Report
Section 8:
Readmissions
From Post-
Acute Care
Post-acute care readmissions
78
• Readmissions stratified by status
code/discharge disposition on claim
• Available settings
– Home
– Home health
– Skilled nursing facility
– Other
Readmissions from SNF
79
Readmissions from SNF
80
Readmissions from SNF How to use these data
• Compare rates among settings
• Explore high rates from a
particular setting
• Engage community partners to
improve discharge transitions
81
In summary…
82
• Consider overall readmission goals
• Analyze available data
• Provide leadership with full picture
• Identify focus areas from your data
• Track, trend, revise, revisit
In summary…
83
• Consider overall readmission goals
• Analyze available data
• Provide leadership with full picture
• Identify focus areas from your data
• Track, trend, revise, revisit
In summary…
84
• Consider overall readmission goals
• Analyze available data
• Provide leadership with full picture
• Identify focus areas from your data
• Track, trend, revise, revisit
In summary…
85
• Consider overall readmission goals
• Analyze available data
• Provide leadership with full picture
• Identify focus areas from your data
In summary…
86
• Consider overall readmission goals
• Analyze available data
• Provide leadership with full picture
• Identify focus areas from your data
• Track, trend, revise, revisit
Next steps
87
Once you know
what to target . . .
. . . the QIN-QIO
and the HIIN can
help identify
interventions and
feasible
measurement
strategies
Perform root cause
analysis
Identify readmission
drivers
Select and implement
intervention
Track process and
outcome data
Determine effectiveness
We want to hear from you!
88
https://www.surveymonkey.com/r/MEC305
2217
Contact us…
Maureen Leary
Rebekah Gardner, MD
Blake Morphis
Sandra Parker
89