Download presentation
Presentation is loading. Please wait.
Published byLuisa Boer Modified over 6 years ago
1
Bridging the Gap Empowering Caregivers With Real Time Access to Aggregate Patient Safety Data Jeffrey M. Ferranti, MD, MS Director, Computerized Patient Safety Initiatives Director, Enterprise Data Warehouse Duke University Health System
2
The Recent IOM Report Preventing Medication Errors (2006) 1
Employ error detection methodologies in all care settings Assess the medication use process through active monitoring Computerized detection of ADEs should complement voluntary reporting “Health care systems should capture information on medication safety and use this information to improve the safety of their care delivery systems.”1 The truth is that over the past 5-7 years our ability to capture electronic information in healthcare has OUTPACED our ability to effectively analyze that information. 1. Institute of Medicine. Preventing Medication Errors: Quality Chasm Series. Washington, DC, National Academy Press, 2006.
3
Utilizing Data To Facilitate Safety and Quality Efforts
IT Safety Systems The Data Warehouse (DSR) Business Intelligence Tools Six Sigma Analytics Performance Services The Balanced Score Card Patient Safety Office Technology We have all of the pieces in place to begin bridging this information gap. We are developing a link between the data from our information systems and operations Operations
4
Error Monitoring at Duke A Multi Faceted Approach
The Qualitative Approach Safety Reporting (SRS) Provides qualitative data that informs safety and quality initiatives Recently upgraded to SRS 2.0 which standardizes the reporting process Not a reliable metric. The Quantitative Approach ADE-S Computerized detection of ADEs; Excellent Metric Based on the gold standard of manual chart review Standardized scoring system, published inter-rater reliability
5
The Qualitative Approach Voluntary Safety Reporting System (SRS)
In 2004 we had 23 different ways of reporting safety events at Duke. Today there is 1.
6
Error Monitoring at Duke A Multi Faceted Approach
The Qualitative Approach Safety Reporting (SRS) Provides qualitative data that informs safety and quality initiatives Recently upgraded to SRS 2.0 to facilitate front line reporting Not a reliable metric. The Quantitative Approach ADE-S Computerized detection of ADEs; Excellent Metric Based on the gold standard of manual chart review Standardized scoring system, published inter-rater reliability
7
The Quantitative Approach Computerized ADE Surveillance
Database Lab Pharmacy Naranjo Pharmacist Review ADT Orders SI Extensibility : PERIOP / Ambulatory Event Report Generator CDR Possible ADEs to Pharmacist for Review Kappa Daily Reports
8
Bridging the Gap Duke’s Strategy for Operationalizing Safety Data
Empower caregivers with direct access to safety data (COGNOS) Predictive Analytics
11
OLAP Processing / Data Exploration
12
Bridging the Gap Duke’s Strategy for Operationalizing Safety Data
Empower caregivers with direct access to safety data (COGNOS) Predictive Analytics
13
ADE-S / SRS Trend Report for Narcotics
Transition Period / Surveillance Data Unavailable Serious Adverse Drug Events Reported to The FDA Wall Street Journal / Archives of Internal Medicine Thomas Moore
14
Y X Six Sigma Opiate Pilot Project Develop a Predicative Risk Model
63,033 Total Patients Exposed to Narcotics 322 Narcotic ADEs SI>=3 Y X Patient Age Patient LOS Patient Gender Clinical Service Morphine (Y/N) Fentanyl (Y/N) Midazalam (Y/N) Etc … Predictive Model- Used recursive partitioning of 63K narcotic exposed patients to uncover “high-risk” (statistically significant) subgroups of pts at higher risk for Opiate ADE; dataset included periop risk factors, opiates, and other interacting medications (SSRIs, benzos, sedatives, etc.);
15
Six Sigma Opiate Pilot Project Develop a Predictive Risk Model
63,033 322 Narcotic ADEs SI>=3 0.5 % (1/200) Opiate ADE=rare event (0.5%) - Number needed to treat =200 Subgroups of pts at higher risk- Number needed to treat =10 for 10% Allow us to electronically identify pts at higher risk & focus limited resources on these efforts 8.7 % 9.1 % 7.5 % LOS > 4 Days Age > 51 Fentanyl Fluoxetine LOS > 4 Days Age > 51 Fentanyl Scopolamine Thoracic Surgery Any Narcotic Sertraline Midazolam
16
9.1 % Six Sigma Opiate Project Develop a Predictive Risk Model
If we can predict who is at high risk we can target interventions more effectively Acute Pain Consults Increased Monitoring 9.1 % LOS > 4 Days Age > 51 Fentanyl Fluoxetine Limited resources-focus efforts Frequent Re-Assessment Pharmacy Consults
17
Clinical Information is an Enterprise Asset
18
Questions
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.