1. 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. 2 Agenda Framework for PSI analysis within the hospital – Making Sense To Clinicians Case Studies Conclusions and Recommendations

3. 3 Northwestern Memorial HealthCare Feinberg and Galter Pavilions May 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. 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. 5 Quality and Patient Safety Program Eliminate avoidable adverse events Deliver evidence-based care Enable the best possible outcomes

6. 6 Eliminate Avoidable Severe Adverse Events Avoidable Severe Adverse Events (G,H,I) Total Incidents Reported Severe Harm # of Severe Harm Events # of Incidents Reported

7. 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. 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. 9 Patient Safety Indicators

10. 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. 11 Administrative Data for Quality Metrics AdvantagesDisadvantages Convenient and inexpensiveIncomplete Standardized rulesDepends 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. 12 NMH Patient Safety Indicators

13. 13 Framework for PSI Use

14. 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. 15 Case Studies CODING Foreign Body Retained Infection Due to Medical Care DEFINITION Post-op Bleed CLINICAL IMPROVEMENT Pneumothorax Post-op PE / DVT

16. 16 Framework on a Small Sample (2007) AHRQ PSICodingDefinition Potential Clinical Issue Pneumothorax5 (12%)0 (%)38 (88%) Post-op Bleed3 (8%)10 (26%)26 (67%) Post-op PE / DVT12 (30%)0 (0%)28 (70%)

17. 17 Clinical Case Studies Iatrogenic Pneumothorax Post-Operative DVT/PE

18. 18 AHRQ Validation Study: Summary of PPVs Preliminary estimates (2007) PSI%PPV Accidental puncture or laceration90% Iatrogenic pneumothorax75% Postoperative DVT/PE72% Postoperative sepsis42% Selected infections due to medical care61%

19. 19 AHRQ Validation Study: Iatrogenic Pneumothorax and Outcomes (N=154)* Patient Outcomes% Treated with chest tube44.8 Discharge delay11.7 Readmitted within 30 days of discharge (generally for reasons unrelated to pneumothorax according to the abstractor) 9.1 Moved to a higher level of care7.8 Tension pneumothorax None or Unable to Determine 6.5 29.9 *Check all that apply

20. 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. 21 Procedure Resulting in PTX Type 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% 9% 15% 21% 012345678 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. 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. 23 Interventions to Reduce Complications

24. 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 Embolism Frequency of DVT/PE; 2007-2008(Q1-Q3) 17.4 17.3 14.3 13.9 13.7 11.9 11.5 10.6 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 Loyola UCLA Stanford Brigham Hopkins Mayo Rush UCSF Mass Gen Frequency (rate per 1000 discharges*)

25. 25 New VTE Prophylaxis Protocol – Electronic Medical Record Screenshot

26. 26 Hospital DVT/PE Rates Source: EPSI Coded Diagnosis Data Excludes patients with DVT/PE Present on Admission Bleeding Data represents patients that had a bleeding complication due to an anticoagulant Protocol Implemented

27. 27 Definition Case Study Post-Operative Hemorrhage / Hematoma

28. 28 Observed and Expected Post-Op Bleed Rates with and without Transplant - Calendar 2008 StratificationNumeratorDenominator Observed Rate/1000 Expected Rate/1000O/E RatioPercentile All eligible cases (includes Transplant) 62121585.102.861.78 Between the bottom 25 th and 10 th Percentile Liver/kidney/pancreas transplant2236061.114.8412.62Bottom 10 th Percentile Liver transplant11104105.775.4719.34Bottom 10 th Percentile MS-DRG 5: Liver transplant w MCC or intestinal transplant 766106.065.9817.73Bottom 10 th Percentile MS-DRG 6: Liver transplant w/o MCC438105.264.5823.01Bottom 10 th Percentile Kidney/pancreas transplant1125642.974.599.37Bottom 10 th Percentile MS-DRG 8: Simultaneous pancreas/kidney transplant 515333.333.6591.32Bottom 10 th Percentile MS-DRG 10: Pancreas transplant11662.503.2519.20Bottom 10 th Percentile MS-DRG 652: Kidney transplant522522.224.744.69Bottom 10 th Percentile All other MS-DRGs (Excludes above Transplant MS-DRGS) 40117983.392.801.21 Just Below Top 25 th Percentile

29. 29 – In organizations that performed more then 300 Transplants 60% of the Organizations were in the worst 3 rd 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 Post-Op Bleed Rates with and without Transplant - Calendar 2008

30. 30 Conclusions / Next Steps

31. 31 Transparency, Accountability

32. 32 Conclusions: The Framework Works Coding Definition Clinical Opportunity Results: – Improved quality – Reduced harm – Reduced cost – Improved learning

33. 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