AHRQ Toolkit: The Harborview Experience (Text Version)
AHRQ Toolkit: The Harborview Experience
Ellen F. Robinson, PT ATC
Clinical Quality Specialist
The Harborview Experience
Mission and Priority of care
A set of images showing the Harborview Medical Center, an Airlift Northwest helicopter, and a United States map outlining the geographical area that Harborview covers—Washington, Alaska, Montana, and Idaho— is shown.
July 2008: WHAT IS A PSI?
July 2009: Oh I wish I had a "toolkit"
July 2010: AHRQ Toolkit Project
July 2011: PSI Project Full Integration
Quality Improvement Initiative
1. External Reporting
Image: Screen shot of the 2010 Quality and Accountability Scorecard for Harborview Medical Center.
2. Internal Case Identification
Medical QI Committee (MQIC)
- Departmental M & M review reporting.
- Standardization of identification of potentially preventable harm events for clinical review.
Section A: Readiness for Change
- IQI/PSI Fact Sheets.
- AHRQ Specification Guidelines.
- Readiness to Change (Self Assessment):
- Medical Director —previous director of QI Dept.
- Leadership Support and directive for project.
- The Board was "on board".
- Challenges identified: information dissemination about quality and patient safety to staff at all levels of the organization.
Section B: Applying the Indicators to your hospital data:
- Utilizing UHC database to track rates for PSI.
- UHC Quarterly Summaries ~ 3 months behind.
- Individual Case review from UHC ~ 6 weeks.
- Too late to make an impact.
How do we get PSI data in "real time"?
Three months to implement software.
- Internal Source System for data points (3M).
- 3M Report output= 2 pages, multiple Rows.
- PERL Script to transform into usable input file.
- Windows Version 3.2.
- Validate Numerator and Denominator against UHC output.
- Take the data from AHRQ software and be able to track the QI process for each case.
Other Data Challenges
- Phase II AHRQ Validation Project Participation.
- Changed to Beta Version 4.0.
- Provided feedback to AHRQ technical staff on issues with new version.
- Challenges—Version discrepancies, running different versions concurrently on two machines to validate cases.
Section C: Identifying Priorities for Quality Improvement
- HMC Project Originally utilized UHC as source.
- UHC runs the SAS version software on each hospitals administrative data set.
Image: Screen shot of a detailed scorecard related to the AHRQ PSIs with red or green dots to represent good performance or improvement opportunities is shown.
HMC Highest Prioritization scores: PSI 3 PSI 7 PSI 12
Image: Screen shot of a prioritization matrix is shown.
Take it on the road!
- Presented to Surgical Council, Medical Executive Board, Critical Care Council, Hospital Board, Clinical Documentation Specialists, Coding:
- What are the PSIs?
- Why do we care?
- Current performance/UHC ranking.
- How are we going to review/expectations from teams.
- Possible opportunities for improvement:
- Clinical areas.
- Documentation —Coding.
Section D: Implementation Tools
- Team Charter and Goals.
- Gap Analysis.
- Implementation Plan.
Effective PSI improvement strategies.
Evidence-based best practices for select PSIs.
PSI Improvement Opportunities
Make Friends with your Coders
- Understanding of Metrics.
- Validation of Metrics.
Evidence-based Best Practices for Select PSIs
- Clinical Teams Reviewed.
- PSI 03: Clinical Nurse Specialists wound care.
- PSI 07: Infection Control.
- PSI 12: Anticoagulation Task force: Trauma Surgeon, Hospitalist, Pharmacy, Nursing.
- Monthly Case Review by QI .
- 10 days after end of previous month.
- Upload to internal database to track outcomes of each PSI.
- Providers report back through M&M conferences and MQIC.
HMC PSI Case Review
Image: A screen shot of a flow chart showing the process for HMC case review.
HMC Analysis and Tracking
Image: A screen shot of a sample patient tracking database page.
Section E: Monitoring Progress
Image: A screen shot of two bar graphs. The first one is titled "AHRQ Patient Safety Events Harm Event Categories". The other is AHRQ Patient Safety Events Harm Events Reviewed by Service FY 2011.
*Web based tool for Quality Metrics reporting.
HMC Outcomes—AHRQ Toolkit
PSI 12 Selected as Focus Area
- Clinical Event Search (CES) Tool:
- VTE case finding from internal diagnostic systems for vascular and radiology events.
- VTE prophylaxis data points from EMR.
- Anticoagulation task force review all events to determine quality concerns on a case.
Image: A screen shot of a sample page from a Web-based tool used to search clinical events is shown.
- For all VTE events, was standard of care met?
- Compliance with UW Medicine guidelines?
- Dose intensity?
- If guidelines do not specify, what is the standard?
- Compliance with UW Medicine guidelines?
- What are the opportunities for improvement?
- Transitions of care—OR Procedures.
- Missed/held doses for OR:
- Changed Dalteparin dosing to 2100.
- Education for Residents:
- Noted doses "held for surgery".
- Attending physicians were not aware doses held.
- Guideline directed therapy algorithm with increased web links through EMR and "clinical toolkit."
- Transitions in care—Communication:
- Clinical team pharmacists, ARNPs, Hospitalists receive daily list of all patients who did not receive chemical prophylaxis in the last 24 hours.
- Both "not ordered" and "pt refusal.
- Nursing Education:
- Doses held for "patient ambulatory".
PSI 12 Monitoring Progress
Task Force currently meeting with each clinical team individually to review guidelines and formalize each "standard of care".
Image: A stacked bar graph showing which DVT/PE events met the standard of care for prophylaxis and which had possible opportunities.
Section F: Return on Investment
- Currently under review with our Decision Support and Finance groups.
- Task force on "Efficiency".
- Utilizing the Cost Data from Prioritization Matrix as a quarterly tracking tool.
Section G: Existing QI Resources
- Reviewed by our Research Librarian.
- Incorporated into University of Washington Health Sciences LibGuides Web page:
- Healthcare Quality News.
- Pub Med Searches (preselected QI topics).
- eJournals related to quality and safety.
- Keep Current with Pub Med notifications.
- Measures—links to TJC, NQF, CMS, UHC, IHI, WSHA.
PSI as a Quality Measure
- Systematic review of all PSI events.
- Is a high rate of events a true indicator of a quality issue at a hospital?
- Are all PSI events "preventable"?
- What about "missed" clinical events or "false negatives"? How do we identify those?
PSI Review Findings
At HMC ~ 1/3rd are not "real" events.
- Standardized Case Review:
- Jan to June 2011.
- PSI 3,6,7,9,11,12,15.
- 132 Events:
- 58 occurred—no quality concerns.
- 30 occurred—possible opportunities.
- 33 events related to documentation or coding error.
- 11 events "flawed metric":
- PSI 11 flagged related to a planned two stage surgery.
- PSI 9 flag related to intra-operative bleeding.
At HMC ~ 1/3rd are not "real" events.
|PSI Cases Reviewed||N = 132|
|No QI concerns||44%|
VTE Events From CES PSI 12
- January to June 2011 (67 VTE Events).
- 42 AHRQ PSI 12.
- 25 additional VTE events (false negative):
- 16 cases not identified in administrative data.
- 9 cases no operative procedure.
Without out internal clinical event search tool these cases would be missed QI opportunities.
* There were also 10 Cases of PSI 12 that were not real clinical events.
HMC PSI Project Lessons Learned
Validate, validate, validate...........
- Presentations to clinical providers should focus on actual patient harm events.
- Coding department project lead/liaison.
- Leadership backing to encourage provider accountability and sustain project importance.
- Technical Issues to implement the AHRQ software may be challenging for hospitals.
- Version changes, input file specifications:
- Continue to provide support via Web site/help line.
- Consider asking hospitals what else they need?
- Technical roadmap.
- Webinars for front end users.
- Reporting options.
- Information regarding risk adjustment coefficients and "targets" for each PSI.
PSIs and Public Reporting
- AHRQ QI rates are increasingly being used by external sources as a measure of quality.
- Version used and data source may result in disparate rates for the same institution.
- Information regarding version used should be clearly published on reporting sites.
- Transparency regarding methods for "PSI composite" calculations.
The "Future" of PSI
- Transition from a rate based tracking tool to one that provides hospitals opportunity for real changes for patients?
- How can hospitals use the indicators to analyze "gaps" in current clinical care?
- What to do about the gaps? Ongoing development of the evidence based "best practice" documents.
- Should hospitals do enhanced chart audit of PSIs for additional data points regarding "preventablity"?
- Robin Weinick.
- Donna Farley.
- Rachel Burns.
- Lindsay Mayer.
Harborview Medical Center
- Dr. J. Richard Goss.
- Dr. Anneliese Schleyer.
- Dr. Joseph Cuschieri.
- Ken Jarman, PharmD.
- Ronald Pergamit, QI/IT.
- Derk Adams, QI/IT.
- Patty Calver QI.
Ellen F. Robinson
(206) 744 9550