Making A Title Slide

1
Active Surveillance using Electronic Triggers to
Detect Adverse Events in Hospitalized Patients
Presented to Chicago Patient Safety
Foundation March 11, 2005 By Marilyn Szekendi,
RN Cindy Barnard, MBA, Anne Bobb, RPh, John
Clarke, MD, Kristine Gleason, RPh, Marla Husch,
RPh, Mary Lankford, RN, Gary Noskin, MD, Denise
Rooney, RN, Carol Sullivan, RN
2
Project Objectives

• Method to identify actual and potential adverse
  events
• Identification of hospital systems that
  contribute to adverse events
• Ongoing assessment

3
Hypotheses

• Active surveillance can be an effective
  methodology for identifying adverse events.
• Screening with electronic triggers will improve
  the efficiency of chart review.
• Active surveillance may reduce the incidence of,
  and/or lessen the duration or severity of,
  adverse events.

4
Design and Methodology
5
Study Design

• Harvard Medical Practice Study
• Institute for Healthcare Improvement (IHI)
  Trigger Tool
• Epidemiologic Surveillance

6
Chart Selection
7
Phase 1 Triggers

• Glucose
• Creatinine
• INR
• PTT
• Digoxin level
• Dilantin level
• Amakacin level
• Gentamicin level
• Tobramycin level
• Vancomycin level
• Positive Blood Culture

• C. Difficile
• Fresh Frozen Plasma
• Dalteparin
• Enoxaparin
• Warfarin
• Sodium Polystyrene Sulfate
• Phytonadione
• Flumazenil
• Naloxone
• Protamine Sulfate

8
Chart Selection
9
Phase 3 Triggers

• Length of Stay
• All patients with a LOS of 7 or more days
• Transfer to a Higher Level of Care
• All patients transferred from a general care or
  stepdown unit to an ICU
• Readmission within 3 Days
• All patients readmitted within 3 days of
  discharge from NMH

10
Sample Selection

• Data collection from June 2003 through present
• Daily trigger data received electronically from
  Lab and Pharmacy 10 of all triggers reviewed
• Cases assigned for review Monday through Friday,
  but covering triggered events which occurred on
  all days of the week

11
Data Collection and Analysis

• Trained investigator teams of one RN and one
  pharmacist reviewed each chart
• The entire chart was reviewed and assessed for
  adverse events
• Events rated for harm/error, preventability, and
  severity, and classified by type of event
• Actual adverse events underwent independent MD
  review.

12
Event Types

• Arrest/resuscitation
• Clinical management failure
• Delay in assessment/diagnosis/treatment
• Diagnostic testing error
• Documentation
• Fall related
• Improper positioning

• IV related
• Medication related
• Nosocomial infection
• Patient noncompliance
• Premature discharge or transfer
• Procedure related
• Surgery related
• Transfusion related

13
Results
14
Demographic Data
15
Results
16
Severity of Preventable Adverse Events
17
Actual Adverse Events by Type of Event
18
Yield of Triggers

• Decision on whether to retain triggers based on
  their yield
• Triggers with low yield or thought not to be
  clinically significant were not retained for use
  in current trigger list

19
Conclusions
20
Conclusions

• Active surveillance is an effective method for
  detecting adverse events in hospitalized patients
• Screening with electronic triggers improves the
  efficiency of chart review
• Interventions during active surveillance can
  prevent or reduce the duration or severity of
  adverse events

21
Future Direction

• Retain triggers with highest yield to improve
  efficiency of ongoing operational implementation
  of active surveillance, using triggers known to
  be best at signaling events
• Further refinement of data collection tools and
  analysis to improve efficiency of surveillance
• Look for ways to improve detection of non-drug
  related events (i.e., clinical management,
  surgery related)

22
Future Direction

• Develop method to identify events in specialty
  areas (i.e., OR, ED, LD, and PARR)
• Leverage IT capabilities to improve access to
  triggers
• Findings, trends utilized for quality and patient
  safety improvement initiatives

23
Questions