How Michigan Health and Safety Coalition Survey Data Can Be Used to Improve Hospital Quality and Safety in Michigan

1
How Michigan Health and Safety Coalition Survey
Data Can Be Used to Improve Hospital Quality and
Safety in Michigan

• Darrell A. Campbell, Jr. M.D.
• Henry King Ransom Professor of Surgery
• Chief of Clinical Affairs
• University of Michigan Hospitals and Health
  Centers
• AkkeNeel Talsma, PhD, RN
• Clinical and Research Consultant to MHSC

2
2005 JCAHO National Patient Safety Goals

• Improve the accuracy of patient identification.
• Improve the effectiveness of communication among
  caregivers.
• Improve the safety of using medications.
• Improve the safety of using infusion pumps.
• Reduce the risk of health care-associated
  infections.
• Accurately and completely reconcile medications
  across the continuum of care.
• Reduce the risk of patient harm resulting from
  falls.

3
The IHI 100,000 Lives Campaign

• Deploy Rapid Response Teams
• Deliver Reliable, Evidence-Based Care for Acute
  Myocardial Infarction
• Prevent Adverse Drug Events (ADEs)
• Prevent Central Line Infections
• Prevent Surgical Site Infections
• Prevent Ventilator-Associated Pneumonia

4
MHSC - April 2005

• I. Recent Survey Results
• II. Has the Leapfrog Group had an impact on the
  health care market?
• III. Quality improvement through regional
  collaboration
• IV. Why this matters

5
Recent Survey Results
6
2004 Hospital Survey

• MHSC Hospital Survey conducted for three
  consecutive years
• Content of 2003 and 2004 is nearly identical
  offering opportunities to review trends
• Hospital submissions have decreased
• 2002 Survey n 106
• 2003 Survey n 102
• 2004 Survey n 93

7
2004 Hospital Survey Response

• 93 of 133 eligible hospitals (69.9) participated
  
• Response rate Survey 2003 102 / 136 (75)
• Five new hospitals participated in 2004
• Clinton Memorial Hospital
• Marquette General Health System
• Spectrum Health United Memorial United Campus
• North Ottawa Community Hospital
• Bronson Vicksburg Hospital

8
Summary Findings 2004 Procedure Volumes

• Hospitals that reported performing a procedure
  electively, most often met the recommended
  minimum volume for
• Percutaneous Coronary Intervention 82.1 (86 in
  2003)
• Open Heart surgery 70 (77 in 2003) and
• Carotid Endarterectomy 71.4 (65 in 2003 )
• Elective procedures with the recommended volumes
  least often met are
• Abdominal Aortic Aneurysm 32.1 (51 in 2003)
• Esophagectomy 37.5 (19 in 2003)

9
Survey Responses by Region
2004 2003 Hospitals/
Hospital Hospitals/ Hospital Region
Region Responses () Region
Responses ()

• Southeast 39 33 (84.6) 41
  41 (100)
• Southwest 14 8 (57.1) 14
  7 (50.0)
• West Central 24 15 (62.5) 24
  14 (58.3)
• Mid Michigan 10 7 (70.0) 10
  9 (90.0)
• East Central 20 13 (65.0) 20
  13 (65.0) North Central
  12 8 (66.7) 12 10 (83.3)
• Upper Peninsula 15 9 (60.0) 15
  8 (53.3)
• Total 134 93 (69.4) Total 136
  102 (75.0)

10
MSSC 2004 SurveyScoring Methodology

• Volume not scored
• Non-volume scored
• appropriateness 50
• risk adjustment 25
• participation in database 25

11
Non-volume Activities

• have appropriateness criteria?
• does hospital require staff to use
  appropriateness criteria?
• hospital have a risk-adjustment system?
• collect risk-adjusted data?
• submit data to database?
• would you be willing to submit data?

12
MHSC 2004 Survey Scoring Methodology (contd)
13
Open Heart SurgeryVolume and Recommended
Activities Met
14
Percutaneous Coronary Intervention (PCI)Volume
and Recommended Activities Met
15
Abdominal Aortic Aneurysm (AAA) Volume
and Recommended Activities Met
16
Carotid Endarterectommy (CEA) Volume
and Recommended Activities Met
17
Esophagectomy Volume and Recommended Activities
Met
18
Low Birthweight Infants Volume and
Recommended Activities Met
19
ICU Physician Staffing Volume and
Recommended Activities Met
20
Overall Improvement Comparison

• Comparison of percentage of hospitals that meet
  both volume threshold and scored at least 80 on
  the activities aspect of the guidelines
• 2003 46.6 (115 / 247)
• 2004 64.8 (105 / 162)
• Notes
• Fewer hospitals reported to the survey in 2004
  than 2003
• Fewer smaller hospitals responded in 2004, these
  are typically the hospitals that have difficulty
  meeting criteria

21
Does your hospital have a risk-adjustment system?

• Type n yes
• Open heart 30 90
• PCI 29 96
• AAA 56 72
• Carotid 63 66
• Esophagectomy 32 75
• Low birth weight 20 90

22
Does your hospital collect risk-adjusted data?

• Type n yes
• Open heart 27 96
• PCI 27 89
• AAA 41 92
• Carotid 45 91
• Esophagectomy 25 92
• Low birth weight 18 100

23
Does your hospital submit data to a state-wide or
national database?

• Type n yes
• Open heart 30 80
• PCI 28 92
• AAA 56 27
• Carotid 63 27
• Esophagectomy 32 6
• Low birth weight 16 65

24
If you do not participate, would you be willing
to?

• Type n yes
• Open heart 5 40
• PCI 2 50
• AAA 38 87
• Carotid 45 89
• Esophagectomy 28 79
• Low birth weight 7 57

25
Problems with the survey

• fewer hospitals responding
• how is the data used?
• doesnt appear to have stimulated
  hospital-hospital debate
• no evident change in results
• no teeth

26
(No Transcript)
27
Nationally Has Leapfrog had an effect?

• Two main elements of Leapfrog
• Hospital Quality and Safety Survey
• Accountability and Incentive Programs

28
Hospital Quality and Safety Survey

• Not audited CEO confirms data
• Average 60 hospital participation
• 64 employers communicate results to enrollees

29
Accountability and Incentive Programs

• Designed to affect consumer selection of
  hospitals
• or
• Pay providers for performance

30
CMS

• Full market basket price guaranteed to providers
  who agree to report process measures

31
Empire Blue Cross and Blue Shield

• IBM, Pepsi Co., Verizon, Xerox
• Quarterly payments to hospitals that have fully
  implemented CPOE and ICU physician staffing

32
Boeing in Seattle -

• Employees who choose hospitals meeting Leapfrog
  standards have sizeable co-pay waived.

33
Progress to Date -

• 2001 2003
• Internists staffing ICU 12 24
• CPOE 2 5

34
Challenges to Leapfrog

• Too few hospitals
• No change in participating hospitals decisions
• Majority of consumers dont use the data
• Expectations for rapid change
• Inadequate rewards (x 200k/yr)
• Reluctance of purchasers

35
How can the system be used to stimulate
improvement activities?

• Pay for performance?
• OR

36
Pay for Participation in a state-wide or
national benchmarking activity

• collaborative data sharing
• a more detailed analysis of results
• focus on regional problem areas
• identification of best practices

37
Quality Improvement Through Regional Collaboration
38
NSQIP National Surgical Quality Improvement
Project

• prospective
• standardized definitions
• trained nurse reviewers
• interrater reliability
• defined endpoints
• 30 day mortality
• 30 day morbidity
• RISK ADJUSTMENT

39
Changes in Hospital Ranks After Risk Adjustment
for 30-day Mortality
Rank by unadjustedmortality rate ()(1
lowest rate)
Rank by risk-adjustedmortality rate (O/E
ratio)(1lowest ratio)
40
NSQIP Annual Report FY 2000 Mortality O/E
Ratios for All Operations
3
2
1
0
41
Is the NSQIP Applicable to the Private Sector?

• YES

42
Internet
VA intranet
VA CCS Hines
DHCP/VISTA
123 VA Medical Centers
UM
Emory
UK
Private Sector Initiative
NSQIP
43
(No Transcript)
44
PATIENT SAFETY IN SURGERY STUDY
Dept of Veterans Affairs
NSQIP
Am College of Surgeons
Exec Committee
QCMetrics
Data Center
14 non-VA Centers
128 VA Surgical Centers
45
14 Participating Centers
Cornell Columbia Florida Virginia

• MGH
• UCSF
• U Mich
• St. Louis U
• Maryland

• Utah
• Brigham
• Wash U
• Emory
• U Kentucky

46
(No Transcript)
47
(No Transcript)
48
Overall
Post Operative Occurrences
All cases 30 Day Documented Followup

• UM

• NSQIP

Number of cases 1401
11,559 Rate of documented 30-day f/u
99.57 96.24
n Cases with 0
postoperative 1212 86.51 82.28
occurrences
49
Respiratory Occurrences
UM (1401) NSQIP (11,599)
N
Pneumonia 121 0.86 1.94 Unplanned
intubation 14 1 1.77 Pulmonary embolism 4 0.29 0.2
4 Ventilator gt48 hrs 26 1.86 3.14 Other
(respiratory) 6 0.43 1.24
50
Urinary Tract Occurrences
UM (1401) NSQIP (11,599)
N
Acute renal failure 12 0.86 0.68 Urinary tract
infection 37 2.64 2.98 Other (urinary
tract) 0 0 0.34
51
Other Surgical Occurrences
UM (1401) NSQIP (11,599)
N
Bleeding/transfusions 4 0.29 1.03 Graft/prosthetic
/flap failure 0 0 0.16 DVT/thrombophlebitis 11 0.7
9 0.63 Sepsis 15 1.07 2.07 Septic
shock 6 0.43 1.14 Other occurrence 18 1.28 4.33
52
Wound Occurrences
UM (1401) NSQIP (11,599)
N
Superficial incisional SSI 45 3.21 3.64 Deep
incisional SSI 14 1 1.31 Organ space
SSI 20 1.43 1.64 Wound disruption 31 2.21 1.59
53
20
18
16
14
12
Morbidity Rate ()
10
8
6
4
2
0
Phase 1
Phase 2
1996
1997
1998
1999
2000
Reduction of morbidity (45) in VA hospitals
since the introduction of the NSQIP
54
Northern New England Cardiovascular Disease
Study Group
The Northern New England Cardiovascular Disease
Study Group exists to develop and exchange
information concerning the treatment of
cardiovascular disease. It is a regional,
voluntary, multi-disciplinary group of
clinicians, hospital administrators, and health
care research personnel who seek to improve
continuously the quality, safety, effectiveness,
and cost of medical interventions in
cardiovascular disease.
Eastern Maine Medical Center
Fletcher Allen Health Care
Dartmouth- Hitchcock Medical Center
Central Maine Medical Center
Maine Medical Center
Concord Hospital
Catholic Medical Center
Portsmouth Regional Hospital
Beth Israel Deaconess Medical Center
55
Fatal Low Output Heart Failure after Isolated CABG
1.60
p trend lt0.001
1.40
1.20
1.00
Fatal LOF
0.80
1.37
1.21
0.60
0.40
0.74
0.20
0.00
1996-1998
1999
2000-2002
56
BCBS of Michigan Cardiovascular Consortium (BMC2)

• A regional collaboration in Michigan involving
  PCI
• Moscucci et al

57
Regional quality collaboration with the Payer

• An idea whose time has come

58
Benchmarking Why a Consortium is Needed

• Low n for complications insufficient data in
  any given hospital
• Large, robust, risk adjusted database can be used
  to understand process-outcome link
• Comparative performance across hospitals re
  processes and outcomes used to find opportunities
  for improvement

59
BMC2
Northern Michigan, Petoskey
Participating Hospitals Since Inception of BMC2
Mclaren, Flint
Spectrum, Grand Rapids
St. Joseph Mercy, Pontiac
Harper, Detroit
St. Joseph Mercy, Ann Arbor
Henry Ford, Detroit
University of Michigan, Ann Arbor
60
(No Transcript)
61
(No Transcript)
62
Cost saving/10,000

• Transfusion 1,750,000
• Unplanned CABG 1,120,000
• Dialysis 2,185,000
• MI 100,000
• Total 5,155,000

63
Marquette, Marquette
Northern Michigan, Petoskey
St. Marys, Saginaw
Spectrum East, Grand Rapids
Mclaren, Flint
St. Joseph Mercy, Pontiac
Spectrum, Grand Rapids
Genesys, Grand Blanc
Ingham, Lansing
Providence, Southfield
Sparrow, Lansing
St. Johns, Detroit
Borgess, Kalamazoo
Harper, Detroit
St. Joseph Mercy, Ann Arbor
Henry Ford,Detroit
Oakwood, Dearborn
University of Michigan, Ann Arbor
64
A Regional Collaborative in Michigan Using the
NSQIP as a Framework

• University of Michigan (M-Score)
• 15 Selected Hospitals in Michigan
• BCBS of Michigan/BCN

65
Quality Collaborative

• BCBSM/BCN supports costs of data collection and
  ACS licensing
• NSQIP data is owned by the hospital, but
• Hospitals agree to share data (anonymously) to
  improve quality

66
Selection criteria for hospitals

• Identification of a surgical champion
• Support from hospital administration
• Geographic representation
• Academic and community hospitals
• Sufficient surgical volume
• Willingness to share data
• Motivation to improve

67
Advantages of the Regional Approach

• Better trouble shooting
• Enhanced audit
• Collegial problem solving approach
• Improved relations with payers
• Identification of region specific problems
• Enhanced public perception of quality efforts
• Peer pressure to improve
• Flexibility to include processes

68
Why all of this matters
69
(No Transcript)
70
A business case for quality

• Complications are expensive avoid them and costs
  will be dramatically reduced

71
To bring the cost factor in

• We linked cost accounting software (TSI) to the
  patients evaluated in NSQIP

72
Hospital costs and surgical complications A
report from the private-sector NSQIP
Justin B. Dimick, MD, Steven L. Chen, MD, Paul
A. Taheri, MD, MBA, William G. Henderson, PhD,
Shukri F. Khuri, MD, and Darrell A. Campbell,
Jr, MD From the University of Michigan Medical
Center, the University of Colorado, and Harvard
Medical School
73
Infectious
Cardiovascular
Respiratory
Thromboembolic

74
A value paradigm based on the NSQIP

• The Business Case for Quality

75
Price is what you pay. Value is what you
get. -Warren Buffett
76
Quality
Value
Cost
77
Quality (NSQIP)
Value
Cost (TSI)
78
The Risk-Adjusted Value Grid
Low Quality High Cost
High Quality High Cost
Cost
Quality
Low Quality Low Cost
High Quality Low Cost
79
The Impact of Surgical Quality Measurement via
the NSQIP

• Better patient care
• Massive cost savings

80
National Surgical Quality Improvement Program
FY92-00 MORBIDITY FOR ALL SURGERY
18
16

14
30-Day Morbidity ()
12
10
8
Phase 1
Phase 2
FY 96
FY 97
FY 98
FY 99
FY 00