�ݺ�ߣ

(4) Comparison of Accident Models
This table provides a summary of accident models analyses of the
case study.
Analysis of Medication Dosing Error related to CPOE System:
Comparison of ECF, HFACS, AcciMaps and STAMP Methods
(1) Introduction
• Safety is considered an “emergent system property” from complex
sociotechnical systems (e.g. Healthcare Systems) (Salmon et al, 2012).
• Patient safety can be affected from human computer interactions and
usability issues leading to medication errors (e.g. Computerized
Physician Order Entry Systems) (Horsky et al, 2005; IOM, 2012).
• Accident causation methods are utilised for analysing accidents,
contributing factors and development of effective recommendations
for prevention.
• Traditional based methods (event and time based) are limited in
analysing accidents in complex systems (Leveson, 2002).
• Systemic models (e.g. AcciMaps) incorporating “Systems thinking” are
arguably considered a dominant paradigm for accident analysis within
the academic community.
• Research gap exists between research and practice in the utilisation of
systemic models for investigating accidents (Underwood and
Waterson, 2014).
(3) Methodology
• A CPOE related medication dosing error case study (Horsky et al,
2005) was used for comparing accident causation methods (See
Appendix I).
• The outputs were further validated with safety researchers.
• A criteria based on a previous study (Underwood and Waterson,
2014) was used to compare respective outputs of the analyses.
(6) Limitation of Rasmussen’s AcciMaps
• The subjective nature of AcciMap analysis of accidents.
• The reliability of this systemic approach is low.
• This approach has yet to be fully accepted in practice within safety
critical systems (i.e. Healthcare systems).
(8) Future Work
• Based on the existing limitations, a domain specific taxonomy will
be developed across each of the levels of the AcciMap structure.
• This new taxonomy based AcciMap method will be utilised as part
of an ongoing ethnographic study within NHS Scotland and for
reliability and validity testing.
(7) Conclusions
• Common causes and contributing factors were identified from the
resulting outputs of the case study.
• Clinicians need to regard medication errors as a result of human-
computer interactions/usability issues from a sociotechnical
perspective to reduce the risks to patient’s safety.
• Systemic models provides a deeper analysis for identifying systemic
factors within complex sociotechnical systems than traditional event
based methods.
• The reliability and usability of the AcciMap approach needs further
improvement for accident analysis in the Healthcare system.
Igene Oseghale Osezua
School of Computing Science
o.igene.1@research.gla.ac.uk
(9) References
• Leveson, N. G. (2002). “System Safety Engineering: Back to the Future”. Aeronautics and
Astronautics Department. Cambridge, MA, Massachusetts Institute of Technology, pp 53-60.
• Salmon, P.M., Cornelissen, M., and Trotter, M.J. (2012). “Systems-based Accident Analysis
Methods: A Comparison of AcciMap, HFACS, and STAMP”. Journal of Safety Science, Vol. 50, Issue
4, pp 1158-1170.
• Underwood, P. and Waterson, P. (2014). “Systems thinking, the Swiss Cheese Model and Accident
Analysis: A Comparative Systemic Analysis of the Grayrigg Train Derailment using the ATSB,
AcciMap and STAMP models”. Journal of Accident Analysis and Prevention, Vol. 68, pp 75-94.
(2) Objectives of Study
• A comparison of ECF, HFACS, AcciMaps and STAMP methods based on
the analyses of a medication dosing error incident.
• Determine any existing “commonality” of causes based on different
accident causation methods applied in the study.
(5) AcciMap Output
Each output from the analyses (See Appendix II and III) represents
the modelling of the events and contributing factors that led to the
accident (Patient becoming hyperkalemic).
Model
Characteristics
Events and Causal
Factor (ECF) Analysis
Human Factors and
Classification Systems
(HFACS) Analysis
System Theoretic
Accident Model and
Processes (STAMP)
Analysis
Rasmussen’s AcciMap
Analysis
Graphical
representation
of the
Accident
1) Analysing events leading to
the accident in a
chronological and linear
way.
2) Uses different symbols for
denoting specific meanings
thus providing effective
visual communication.
1) Utilizes a taxonomy of failures
for analysing and classifying
causes.
2) Does not provide any
diagrammatic tools for visual
communication.
1) Its taxonomy provides a
platform for analysing
deviations between different
components for both lower
and higher levels.
2) Complex visual presentation
in depicting these
relationships.
1) Provides a very effective way
of visual communication
regarding the accident
(adverse event) that took
place
2) Depicts causal relationships
within the different
hierarchical levels of the
model.
Data
Requirements
1) Data required based on
different formats including
incident reports, interview
documentation, and
observation.
documentation, and
observation.
documentation, and
observation.
documentation, and
observation.
Usability/Ease
of Learning
1) Requires substantial
knowledge, understanding
and representing the
sequence of events leading
up to the failure.
2) Easy to learn and use for
analysing causal factors.
knowledge, understanding and
application of the model in
analysing and classifying the
failures into its different
subcategories.
2) Does not require too much
guidance in its analysis.
knowledge, understanding of
the concept of control theory
as well as its resultant system
taxonomy for analysing
deviations between
component interactions.
2) A considerable level of
difficulty in usage.
knowledge in understanding
but simple to utilize for
representing the causal
relationships (distal and
proximal).
2) Lack of a standard guide for
conducting the analysis.
Validity of
Analysis
1) Useful in multiple safety
critical domains.
2) Can be used with other
techniques including MORT,
Barrier and Change
Analysis.
3) External validity is required.
1) Useful model in multiple safety
critical domains.
2) Based on Reason’s theory of
accident causation which
classifies latent and active
failures in a systematic way.
1) Useful model in multiple
safety critical domains.
2) Based on System and Control
theory regarding accident
causation for analysing
system complexity.
1) Useful model in multiple
safety critical domains.
2) Based on Rasmussen’s theory
of accident causation for
analysing structure and
dynamics within the complex
sociotechnical system.
Reliability of
Analysis
1) A qualitative approach for
analysing case incidents.
2) The analysis of an accident
is open to interpretation by
different investigators and
so reliability is not high.
1) Provides qualitative and
quantitative means of analysing
both single and multiple cases.
2) Its taxonomy allows
investigators to analyse active
and latent failures and the
contributing factors allowing for
a higher reliability to be
achieved.
1) Provides a qualitative
approach for analysing
dysfunctions in the system.
Its taxonomy is not very
detailed for classifying
failures and contributory
factors.
2) Based on control theory
which allows for a
considerable level of
reliability.
1) Provides a qualitative
approach for analysing
multiple proximal, distal
causes and contributory
factors across different
levels.
2) Its reliability is low due to its
subjective nature (open
interpretation to different
analysts).
*Levels of
Sociotechnical
Analysis
1) Its theoretical approach
does not make the method
the most suitable for
analysing
incidents/accidents from
complex interactions in a
sociotechnical system.
1) Its theoretical approach allows
for the classification of multiple
contributing factors from the
lower to higher levels of failure
within the system.
2) Its current taxonomy does not
consider other higher levels of
systemic failures (e.g.
regulatory bodies, government)
allows for failures to be
analysed from the
interactions of system
components existing in both
the lower levels to the higher
levels of the sociotechnical
system.
2) Suitable for sociotechnical
analysis.
allows for a sociotechnical
analysis of failures and
contributing factors across
all hierarchical levels within
the sociotechnical system.
Government
Policy and
Budgeting
Company
Management,
Local Area
Government
Regulatory bodies,
Trade Unions and
Associations
Technical and
Operational
Management
Physical
processes
and Actor
activities
Equipment
and
Surroundings
The CPOE system had a
poor user interface design
Lack of clinical
handover training
and implementation
plan/strategy
Provider B
administered
more KCI
dosage
There was no
discussion
regarding the
patient’s KCI levels
during the clinical
changeover
Miscommunication
between the
attending nurse
and Provider A
Lack of
communication
between the
Provider A and
Provider B
Inadequate resources
for planning, hiring and
training of new and
existing staff
Organization’s
Low risk and
safety culture
Lack of proactive
approaches in analysing
health-IT related
incidents
Government
priorities on
budget allocation
Staff did not receive
adequate training in
using the CPOE
system
Lack of oversight relating
to safety and risk
management policies
Limited budget for IT
maintenance and
optimal operations
Patient becomes
Hyperkalemic
Events (Inferences) Adverse Event
The CPOE system did not
recognize instructions in
comment boxes
Under reporting of
errors relating to
the CPOE system
Lack of
communication
between the users
and IT department
Lack of feedback
between the health
organization and
Software Vendors
Lack of Incident
reporting on IT
related errors
The attending nurse had
difficulties reading in the
instructions
The CPOE screen did not
display patient's recent
result before KCI
administration
Underestimating
risks associated
with the use of
CPOE systems
The patient
received multiple
doses of KCI than
originally intended
Provider A did not
enter the correct
dosage into the
CPOE system
Poor information
flow between the
staff
The CPOE system
was not tested in a
simulated
environment before
live deployment
Events/Actions/Decisions

�ݺ�ߣ

Igene - PhD SICSA Poster Presentation

More Related Content

Igene - PhD SICSA Poster Presentation