1

Northwestern Memorial HealthCare

AHRQ Patient Safety Indicators: Constructive Use for Improvement

Presented to

AHRQ Annual Conference

September 15, 2009

By

Cynthia Barnard MBA MSJS CPHQ

Director, Quality Strategies

2

Agenda

• Framework for PSI analysis within the hospital– Making Sense To Clinicians

• Case Studies

• Conclusions and Recommendations

3

Northwestern Memorial HealthCare

Feinberg and Galter PavilionsMay 1, 1999

New Prentice Women’s Hospital October 20, 2007

• 873-bed Nationally Recognized Academic Medical Center

• Primary Teaching Hospital for Northwestern University since 1925

• Nationally Ranked for Quality

• New World-Class Facilities in 1999 and 2007

• Aa/AA Category Bond Rating for Over 25 Years

4

NMH Recognized for Quality and Excellence

• Magnet Certification since 2006

• 11 Specialties in 2009 U.S. News & World Report of Best Hospitals

• 2005 National Quality Health Care Award

• “Most Preferred Hospital” for 14 Years (NRC)

• Leapfrog Group’s “Top Hospitals List” twice

• Named to “100 Best Companies for Working Women” for 9 Years

• “Most Wired” for 9 years

• Among University Healthsystem Consortium Top 15 in Quality and Accountability

5

Quality and Patient Safety Program

• Eliminate avoidable adverse events

• Deliver evidence-based care

• Enable the best possible outcomes

6

Eliminate Avoidable Severe Adverse Events Avoidable Severe Adverse Events (G,H,I)

To

tal In

cide

nts R

ep

orte

d

Severe Harm

# of

Sev

ere

Har

m E

vent

s

# of Incidents Reported

0

5

10

15

20

25

30

35

40

45

50FY

04Q

2FY

04Q

3FY

04Q

4FY

05Q

1FY

05Q

2FY

05Q

3FY

05Q

4FY

05Q

4FY

06Q

1FY

06Q

2FY

06Q

3FY

06Q

4FY

07Q

1FY

07Q

2FY

07Q

3FY

07Q

4FY

08Q

1FY

08Q

2FY

08Q

3FY

08Q

4FY

09Q

1FY

09Q

2FY

09Q

3 0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

7

Agency for Healthcare Research and Quality (AHRQ)

• AHRQ Quality and Patient Safety Indicators (QIs/PSIs) are measures of health care quality that make use of readily available hospital inpatient administrative data.

• To improve the quality of healthcare, accessible and reliable indicators are needed to:– Flag potential problems or successes

– Follow trends over time

– Identify disparities across regions, communities and providers

– Address multiple dimensions of care

8

AHRQ – Quality Indicators

• Inpatient Quality Indicators, 2002– Reflect quality of care inside hospitals including inpatient mortality for medical

conditions and surgical procedures.

• Patient Safety Indicators (PSI), 2003– Reflect quality of care inside hospitals, but focus on potentially avoidable

complications and iatrogenic events

– Screen for adverse events that patients experience as a result of exposure to the health care systems

– Target events that are likely amenable to prevention by changes at the system provider level

– Includes 20 indicators

9

Patient Safety Indicators

Complications of Anesthesia 1Death in Low-Mortality DRGs 2Decubitus Ulcer 3Failure to Rescue 4Foreign Body Left During Procedure 5Iatrogenic Pneumothorax 6Selected Infections Due to Medical Care 7Postoperative Hip Fracture 8Postoperative Hemorrhage or Hematoma 9Postoperative Physiologic and Metabolic Derangements 10Postoperative Respiratory Failure 11Postoperative Pulmonary Embolism or Deep Vein Thrombosis 12Postoperative Sepsis 13Postoperative Wound Dehiscence 14Accidental Puncture or Laceration 15Transfusion Reaction 16Birth Trauma – Injury to Neonate 17Obstetric Trauma – Vaginal with Instrument 18Obstetric Trauma – Vaginal without Instrument 19Obstetric Trauma – Cesarean Delivery 20

10

Example of PSI Specification

• Iatrogenic Pneumothorax, (PSI 6)

• Provider Level Definition (only secondary diagnosis)

• Definition: Cases of iatrogenic pneumothorax per 1,000 discharges. • Numerator: Discharges with ICD-9-CM code of 512.1 in any secondary diagnosis field. • Denominator: All medical and surgical discharges age 18 years and older defined by specific

DRGs.

• Exclude cases: • with ICD-9-CM code of 512.1 in the principal diagnosis fiel • MDC 14 (pregnancy, childbirth, and puerperium) • with an ICD-9-CM diagnosis code of chest trauma or pleural effusion • with an ICD-9-CM procedure code of diaphragmatic surgery repair • with any code indicating thoracic surgery or lung or pleural biopsy or assigned to cardiac surgery DRGs

• Empirical Perf: Population Rate (2003): 0.562 per 1,000 population at risk

• Risk Adjustment: Age, sex, DRG, comorbidity categories

11

Administrative Data for Quality Metrics

Advantages Disadvantages

Convenient and inexpensive Incomplete

Standardized rules Depends on non-standardized charting, vague clinician usage, and ability to find evidence in chart

Audited (for billing purposes) Audit focus is not on clinical completeness but on DRGs

Includes diagnoses, procedures, age, gender, admission source and discharge status

Excludes important clinically influential data: DNR/palliative, clinical context, degree of severity

12

NMH Patient Safety Indicators

-350.00%

-300.00%

-250.00%

-200.00%

-150.00%

-100.00%

-50.00%

0.00%

50.00%

100.00%

150.00%

200.00%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Var

ian

ce t

o A

HR

Q E

mp

iric

Va

lue

Complications Of Anesthesia

Death In Low Mortality DRG

Decubitus Ulcer

Failure To Rescue

Foreign Body Retained

Iatrogenic Pneumothorax

Infection Due To Medical Care

Postoperative Hip Fracture

Postop Hemorrhage Or Hematoma

Postop Physio Metabol Derangmnt

Postop Respiratory Failure

Postoperative Pe Or Dvt

Postoperative Sepsis

Postoperative Wound Dehiscence

Accidental puncture/laceration

Transfusion Reaction

Birth Trauma

OB Trauma - Vaginal W Instrument

OB Trauma - Vaginal Wo Instrument

OB Trauma - C-Section

PE/DVTAPL

OB

OB

FTR

Pneumothorax

Decub

Size of bubble represents number of cases

13

Framework for PSI Use

14

Framework

• Coded accurately?

• Definition omits important clinical factors?

• Actual clinical process problem?

Similar approaches:

Houchens, Elixhauser, Romano. How Often are Potential Patient Safety Events Present on Admission? Joint Commission Journal on Quality and Patient Safety, March 2008

Henderson, et al. Clinical Validation of the AHRQ Postoperative Venous Thromboembolism Patient Safety Indicator. Joint Commission Journal on Quality and Patient Safety, July 2009

15

Case Studies

CODING

• Foreign Body Retained

• Infection Due to Medical Care

DEFINITION

• Post-op Bleed

CLINICAL IMPROVEMENT

• Pneumothorax

• Post-op PE / DVT

16

Framework on a Small Sample (2007)

AHRQ PSI Coding Definition

Potential Clinical Issue

Pneumothorax 5 (12%) 0 (%) 38 (88%)

Post-op Bleed 3 (8%) 10 (26%) 26 (67%)

Post-op PE / DVT 12 (30%) 0 (0%) 28 (70%)

17

Clinical Case Studies

Iatrogenic PneumothoraxPost-Operative DVT/PE

18

AHRQ Validation Study:Summary of PPVs

Preliminary estimates (2007)

PSI %PPV

Accidental puncture or laceration 90%

Iatrogenic pneumothorax 75%

Postoperative DVT/PE 72%

Postoperative sepsis 42%

Selected infections due to medical care 61%

19

AHRQ Validation Study:Iatrogenic Pneumothorax and Outcomes (N=154)*

Patient Outcomes %

Treated with chest tube 44.8

Discharge delay 11.7

Readmitted within 30 days of discharge (generally for reasons unrelated to pneumothorax according to the abstractor)

9.1

Moved to a higher level of care 7.8

Tension pneumothorax

None or Unable to Determine

6.5

29.9

*Check all that apply

20

NMH Assessment of Clinical Practice Iatrogenic Pneumothorax

• Question: Was the condition preventable?

• Variables Reviewed for Trends:

• Procedure resulting in pneumothorax (PTX)

– Type

– Location

– Physician/Service (no identifiable trend)

– Day of the week (no identifiable trend)

– Time of day (no identifiable trend)

• Patient factors

– Reason for admission

– Age (no identifiable trend)

– Pulmonary comorbidity (no identifiable trend)

21

Procedure Resulting in PTXType and Frequency of Procedure Resulting in PTX, N=33

Dincer HE, Lipchik RJ. The intricacies of pneumothorax: management depends on accurate classification. Postgraduate Medicine, Dec 2005.

24%

3%

3%

3%

3%

3%

3%

3%

9%

9%

15%

21%

0 1 2 3 4 5 6 7 8

Pacemaker insertion

Lung biopsy

Expected pleural laceration

Diaphragm resection

Bronchoscopy/biopsy

Biliary drain placement

Back surgery

Chest tube removal

Central line placement

Lung surgery

Thoracentesis

Insufficient documentation

22

Pneumothorax Interventions

• Focus on potentially preventable PTX in thoracentesis, pacemaker, and central line procedures

• Weekly case review by patient safety professional, MD

• Focus: Central Line and Pacemaker placement (clinical)– Refreshers, simulation training (central lines), supervision

• Focus: Correctly capturing exclusions (coding)

• Outcome:

• Rate has fallen from 1/week (3-4x expected) to 1-2/month (~expected)

23

Interventions to Reduce Complications

24

In 2007 and 2008(Q1-Q3), approximately 17.3 patients per 1000 discharges*experienced a DVT or PE complication at NMH.

Post-Operative Venous Thrombosis / PE

*excludes OB Product line Source: UHC Clinical Database

Venous Thrombosis/ Pulmonary EmbolismFrequency of DVT/PE; 2007-2008(Q1-Q3)

17.4 17.3

14.3 13.9 13.7

11.9 11.9 11.510.6 10.2 9.8

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

U of C

NMH

Loyo

la

UCLA

Stanf

ord

Brigha

m

Hopkin

s

May

oRus

h

UCSF

Mas

s Gen

Fre

qu

ency

(ra

te p

er 1

000

dis

char

ges

*)

25

New VTE Prophylaxis Protocol – Electronic Medical Record Screenshot

26

Hospital DVT/PE Rates

Source: EPSI Coded Diagnosis DataExcludes patients with DVT/PE Present on AdmissionBleeding Data represents patients that had a bleeding complication due to an anticoagulant

Protocol Implemented

NMH DVT/PE and Bleed Events (excluding OB, Peds, and Psych)

0.0

5.0

10.0

15.0

20.0

25.0

30.0

Jan-08

Feb-08

Mar-08

Apr-08

May-08

Jun-08

Jul-08

Aug-08

Sep-08

Oct-08

Nov-08

Dec-08

Jan-09

Feb-09

Mar-09

Apr-09

May-09

Jun-09

Jul-09

DVT/PE Rate per thousand Goal Bleed Rate per thousand

27

Definition Case Study

Post-Operative Hemorrhage / Hematoma

28

Observed and Expected Post-Op Bleed Rates with and without Transplant - Calendar 2008

Stratification Numerator DenominatorObserved Rate/1000

Expected Rate/1000 O/E Ratio Percentile

All eligible cases (includes Transplant)

62 12158 5.10 2.86 1.78Between the bottom

25th and 10th Percentile

Liver/kidney/pancreas transplant 22 360 61.11 4.84 12.62 Bottom 10th Percentile

Liver transplant 11 104 105.77 5.47 19.34 Bottom 10th Percentile

MS-DRG 5: Liver transplant w MCC or intestinal transplant

7 66 106.06 5.98 17.73 Bottom 10th Percentile

MS-DRG 6: Liver transplant w/o MCC 4 38 105.26 4.58 23.01 Bottom 10th Percentile

Kidney/pancreas transplant 11 256 42.97 4.59 9.37 Bottom 10th Percentile

MS-DRG 8: Simultaneous pancreas/kidney transplant

5 15 333.33 3.65 91.32 Bottom 10th Percentile

MS-DRG 10: Pancreas transplant 1 16 62.50 3.25 19.20 Bottom 10th Percentile

MS-DRG 652: Kidney transplant 5 225 22.22 4.74 4.69 Bottom 10th Percentile

All other MS-DRGs (Excludes above Transplant MS-DRGS)

40 11798 3.39 2.80 1.21Just Below Top 25th

Percentile

29

– In organizations that performed more then 300 Transplants 60% of the Organizations were in the worst 3rd for Observed Rates

– When we exclude transplant from the Post Operative Hemorrhage and Hematoma metric, all but 2 organizations saw a rate improvement ranging from 0.19 to 4.28

Observed Rate / 1000 for Post Operative Hemorrhage and Hematoma with and without Transplant

0

2

4

6

8

Organizations w ithin the UHC Database who Performed Greater then 300 Transplants in Calendar Year 2008

Ob

serv

ed

Rate

(P

ost

Op

era

tive

Hem

orr

hag

e &

Hem

ato

ma/T

ota

l # o

f

Su

rgic

al C

ases)

Observed Ratio / 1000 Observed Ratio without Transplant Cases / 1000

Observed Post-Op Bleed Rates with and without Transplant - Calendar 2008

30

Conclusions / Next Steps

31

Transparency, Accountability

32

Conclusions: The Framework Works

• Coding

• Definition

• Clinical Opportunity

• Results:– Improved quality

– Reduced harm

– Reduced cost

– Improved learning

33

Cynthia Barnard Director, Quality Strategies

Northwestern Memorial Hospital

Research Assistant Professor

Institute for Healthcare Studies

Northwestern University Feinberg School of Medicine

676 St Clair #700

Chicago IL 60611

voice 312.926.4822

fax 312.926.8734

cbarnard@nmh.org