FHAACS Partnership to Improve Surgical Quality in Florida

1
FHA-ACS Partnership to Improve Surgical Quality
in Florida
Clifford Y. Ko, MD, MS, MSHS FACS Director,
Division of Research and Optimal Patient
Care Director, ACS NSQIP American College of
Surgeons Professor of Surgery Robert and Kelly
Day Chair of Surgical Outcomes UCLA School of
Medicine
2
Dedicated to improving the care of the surgical
patient and to safeguarding standards of care in
an optimal and ethical practice environment
3

Data Aggregation, Sharing, and Reporting
every hospital should follow every patient it
treats, long enough to deter- mine whether or not
the treatment has been successfuland if not, why
not ...I am called eccentric for saying in
public that hospitals, if they wish to be sure of
improvement, 1. Must find out what their results
are 2. Must compare their results with those
of other hospitals 3. Must analyze their
results, to find their strong and weak
points. -Codman 1917

Ernest Codman 1869-1940
4
(No Transcript)
5
1915
6
National Surgical Quality Improvement Program
(NSQIP)

• Risk-Adjusted Outcomes (Hospital-based Targeting
  of Mortality and Complications)

7
National Surgical Quality Improvement Program
(NSQIP)

• Risk-Adjusted Outcomes
• Clinical data
• 30 day follow up

8
National Surgical Quality Improvement Program
(NSQIP)

• Risk-Adjusted Outcomes
• Clinical data
• 30 day follow up
• Trained abstractor
• Audited data

9
National Surgical Quality Improvement Program
(NSQIP)

• Risk-Adjusted Outcomes
• Clinical data
• 30 day follow up
• Trained abstractor
• Audited data
• Multi-specialty
• Best Practices (e.g. SSI, UTI)

10
Evolution of NSQIP

1985
86
87
88
89
90
91
92
93
94
95
96
97
98
99
2000
01
02
03
04
05
Pilot study at three private-sector hospitals
NVASRS
VA hospitals under scrutiny
Congress passes law 99-166
AHRQ grant 14 academic medical centers
HCFA reports unadjusted comparative mortality
rates
Inception of VA NSQIP
4 community hospitals join
ACS expansion of private-sector InitiativeACS
NSQIP
11
Evolution of NSQIP

1985
86
87
88
89
90
91
92
93
94
95
96
97
98
99
2000
01
02
03
04
05
Pilot study at three private-sector hospitals
NVASRS
VA hospitals under scrutiny
Congress passes law 99-166
AHRQ grant 14 academic medical centers
HCFA reports unadjusted comparative mortality
rates
Inception of VA NSQIP
4 community hospitals join
ACS expansion of private-sector InitiativeACS
NSQIP
12
Evolution of NSQIP

1985
86
87
88
89
90
91
92
93
94
95
96
97
98
99
2000
01
02
03
04
05
08
Pilot study at three private-sector hospitals
NVASRS
VA hospitals under scrutiny
Congress passes law 99-166
250 Hospital
AHRQ grant 14 academic medical centers
HCFA reports unadjusted comparative mortality
rates
Inception of VA NSQIP
4 community hospitals join
ACS expansion of private-sector InitiativeACS
NSQIP
13
Outcomes Fed Back to Hospitals

• All Cases (11 measures)
• 30 day Mortality, 30 Day Morbidity
• SSI, DVT/PE, Cardiac, Pneumonia, Unplanned
  reintubation, Prolonged ventilation, DVT/PE,
  Renal failure, UTI
• Multispecialty (2)
• General Surgery (10)
• Colorectal Surgery (3)
• Vascular Surgery (3)
• Colorectal LOS (1)
• ENT (2)
• Plastic Surgery (2)
• Orthopedic Surgery (2)
• Neurosurgery (2)

14
1. Risk Adjustment

15
1. Risk Adjustment

• My patients are sicker

16
Observed Outcomes vs. Expected Outcomes (OE)

• Observed Non-adjusted outcomes rate
• e.g. Colectomy SSI rate15 not accounting for
  case mix, patient comorbidities, etc.
• Expected Predicted outcomes rate
• e.g. Colectomy SSI rate should be 15 accounting
  for case mix, comorbidities, etc
• Observed Expected ratio 1
• OE lt 1 (better than expected)
• OE gt 1 (worse than expected)

17
Risk Adjustment Matters (When Measuring Outcomes)

• Risk adjustment has an important effect when
  determining true performance

18
Clinical Data Risk Adjustment Outcomes q 6M
Benchmarking
19
Clinical Data Risk Adjustment Outcomes q 6M
Benchmarking
High Outliers (Needs Improvement)
20
Clinical Data Risk Adjustment Outcomes q 6M
Benchmarking
Low Outliers (Better than Expected)
21
To date, variation exists wherever we measure
risk adjusted outcomes

22

WE CAN RISK ADJUST WITH 5-10 VARIABLES
(Hospital-specific morbidity rates for colectomy
2005-6 ACS-NSQIP)

70
70
Correlation coefficient 0.99
Correlation coefficient 0.99
60
60
50
50
40
40
Risk
-
adjusted
Risk
-
adjusted
morbidity ()
morbidity ()
(5 variables)
(5 variables)
30
30
20
20
10
10
0
0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
Risk
-
adjusted morbidity ()
Risk
-
adjusted morbidity ()
(20 variables)
(20 variables)
23
2. Data Source Clinical vs. Admin/Claims Data

24
Clinical vs. Administrative Data Clinical Data
tends to tell us more

25
3. How long do (should) we follow the patient?
30 days

26

Percent of Outcomes Occurring After Discharge
(Colectomy)

27

• Of those patients with a complication,
• 45 had only a post-discharge complication

28

• Do outcomes improve?
• Do ACS NSQIP Hospitals Get Better?

29
Results- Change in O/E
118 Institutions Present 2006-2007 Ch
ange 2006 to 2007 Complication
Mortality Mean Change in O/E -0.1137 -0.1740
P-value (mean not zero) lt0.000001 lt0.0001
Volume weighted mean -0.1126 -0.1631
Institutions Improved 82 66

(not c/w regression to the mean)
30
Results- Outlier Status and Improvement
Complication Mortality
2
1
31
Results- Outlier Status and Improvement
Complication Mortality
32
Results- Outlier Status and Improvement
Complication Mortality
33
Results- Outlier Status and Improvement
Complication Mortality
34
Results- Outlier Status and Improvement
Complication Mortality
35
Complications Avoided
36
ACS NSQIP
High Quality Surgical Care

Best Practices
Data Collection/ Analysis
NSQIP Cancer Trauma Bariatric Case Log
Feedback
37
Surgical Infection Prevention

• 1 Prophylactic antibiotic received within one
  hour prior to surgical incision
• 2 Prophylactic antibiotic selection for surgical
  patients
• 3 Prophylactic antibiotics discontinued within
  24 hours after surgery end time (48 hours for
  cardiac patients)
• 4 Cardiac surgery patients with controlled 6
  a.m. postoperative serum glucose
• 5 Surgery patients with appropriate hair removal
• 6 Colorectal surgery patients with immediate
  postoperative normothermia

38
Current Process Performance Measures (SCIP) have
little to no correlation with Outcomes

39
CMS

40
Risk Adjusted Outcomes Measures Developed for CMS

• Procedure Based Outcomes
• Lower Extremity Bypass
• Colectomy
• Ventral hernia repair
• Hysterectomy
• Cholecystectomy
• Hip/Knee Replacement
• Complications-Based
• Surgical Site Infection
• Urinary Tract Infection
• Elderly Measure (2)
• 65-79 years old
• 80 years old

41
FHA-ACS NSQIP-Based Approach

• Identify 5 cross cutting, risk-adjusted
  performance measures to use with hospitals.
• Collect (parsimonious set), analyze, and feedback
  data.
• Provide evidence and expert based best
  practices/guidelines.
• Implement Hospital Collaboratives Learning,
  Sharing, and Improvement.
• Expected Outcome Demonstrable Quality
  Improvement.

42
Demonstrable Hospital Improvement
118 Institutions Present 2006-2007 Ch
ange 2006 to 2007 Complication
Mortality Mean Change in O/E -0.1137 -0.1740
P-value (mean not zero) lt0.000001 lt0.0001
Institutions Improved 82
66
43

• Thank you