The Science of Patient Safety: Longitudinal Studies in an Environment of Change. Wagar EA, Hilborne LH, Yasin B, Tamashiro L, and Bruckner DA. UCLA Healthcare and Department of Pathology

1
The Science of Patient Safety Longitudinal
Studies in an Environment of Change. Wagar
EA, Hilborne LH, Yasin B, Tamashiro L, and
Bruckner DA. UCLA Healthcare and Department of
Pathology Laboratory Medicine, David Geffen
School of Medicine at UCLA
2
Patient Identification Safety Initiative

• November, 2002, reviewed all types of specimen
  errors and created categories
• Consulted with nursing, physicians, laboratory
  professionals
• Began collecting continuous data

3
Specimen Error Information Categories

• Clotted specimen
• Container leaking
• Duplicate order
• Hemolyzed specimen
• Improperly collected
• Improperly handled
• Mislabeled specimen

• Quantity not sufficient
• Requisition mismatch
• Specimen not suitable for test
• Tube overfilled
• Tube underfilled
• Unlabeled specimen

4
Methods

• Baseline data collect 11-02 through 3-03
• Critical patient identification categories
  targeted
• Three patient safety initiatives implemented at
  4, 10, and 14 months
• Statistical analyses by paired studentst-test
  and linear trend analysis

5
Three Critical Patient Identification Errors

• Specimen/requisition mismatch
• Unlabeled specimens
• Mislabeled specimens (wrong blood in tube)

6
Three Patient Safety Initiatives

• Phlebotomy service reorganization and education
  4 months
• Electronic event reporting system 10 months
• Automated processing system 14 months

7
Total Errors by Category(4-03 through 2-05)
8
Patient Identification Errors
9
Patient Identification Errors ICUs
10
Patient Identification Errors

• Critical identification errors were 12.0 of all
  specimen errors
• Over 4.29 million specimens and 2.31 million
  phlebotomy requests
• Critical identification errors are lt0.1 of all
  procedures or all specimens
• Patient identification errors occurred frequently
  in ICUs

11
Longitudinal Data Patient Identification Errors
12
Longitudinal Data Patient Identification Errors
Error Specimen Category P-Value
Mislabeled Specimens 0.014
Requisition Mismatches 0.001
Unlabeled Specimens 0.002
13
Other Things that Happed Along the Way.

• Outside consultant, November 2002
• JCAHO, April, 2004
• Departure of the outside consultant, June, 2004
• New CEO appointment, July, 2004
• No significant changes in trends over the period
  March, 2003, through February, 2005

14
Conclusions

• Critical patient identification errors can be
  decreased in an environment of change Leadership
  commitment!
• Expensive IT solutions are helpful but not
  essential as change factors
• Awareness is a key factor for change
• Changes were sustainable (April, 2003, to
  February, 2005)

15
Patient Safety Paradigm for Change
SUSTAINABILITY
16
THANK YOU
THE UCLA PATIENT SAFETY TEAM