Hindawi Publishing Corporation

BioMed Research International

Volume 2014, Article ID 392596, 7 pages
http://dx.doi.org/10.1155/2014/392596

Research Article
Implementation of Incident Learning in the Safety and Quality
Management of Radiotherapy: The Primary Experience in a New
Established Program with Advanced Technology
Ruijie Yang, Junjie Wang, Xile Zhang, Haitao Sun, Yang Gao, Lu Liu, and Lei Lin
Department of Radiation Oncology, Peking University Third Hospital, Beijing 100191, China
Correspondence should be addressed to Junjie Wang; junjiewang47@yahoo.com
Received 28 May 2014; Accepted 14 July 2014; Published 22 July 2014
Academic Editor: An Liu
Copyright 2014 Ruijie Yang et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Objective. To explore the implementation of incident learning for quality management of radiotherapy in a new established
radiotherapy program. Materials and Methods. With reference to the consensus recommendations by American Association of
Physicist in Medicine, an incident learning system was specifically established for reporting, investigating, and learning of individual
incidents. The incidents that occurred in external beam radiotherapy from February, 2012, to February, 2014, were reported. Results.
A total of 28 near misses and 5 incidents were reported. Among them, 5 originated in imaging for planning, 25 in planning, and 1
in plan transfer, commissioning, and delivery, respectively. One near miss/incident was classified as wrong patient, 7 wrong sites,
6 wrong laterality, and 5 wrong dose. Five reported incidents were all classified as grade 1/2 of dosimetric severity, 1 as grade 0,
and the other 4 as grade 1 of medical severity. For the causes/contributory factors, negligence, policy not followed, and inadequate
training contributed to 19, 15, and 12 near misses/incidents, respectively. The average incident rate per 100 patients treated was 0.4.
Conclusion. Effective implementation of incident learning can reduce the occurrence of near misses/incidents and enhance the
culture of safety.

1. Introduction
With the rapid development of equipment and technology,
the technique of radiotherapy becomes more and more complicated. The treatment planning and delivery of radiotherapy
are a highly complex, multiple-step process, with multiple
professional groups involved. Every step of the care process
in radiation oncology requires knowledge in the management
of cancer and certain benign disease, radiobiology, medical
physics, and radiation safety that can only be achieved by
systematic and structured training. There are many steps
where incidents might occur, although major radiotherapy
incidents are rare. And it is challenging to determine the
actual error rate in radiotherapy. The quality management
of radiotherapy is paramountly important to guarantee the
appropriateness, quality, and safety of radiotherapy.
As a systematic tool and approach of quality management,
incident learning has proved its value in many industries
[1]. It has also already been successfully used for a number

of years in several radiation oncology clinics. A voluntary

international reporting system (Radiation Oncology Safety
Information System (ROSIS)) has been in online use for 9
years [2]. The International Atomic Energy Agency (IAEA) is
leading the development of a multi-institution incident learning system, called Safety in Radiation Oncology, SAFRON.
American Society for Radiation Oncology (ASTRO) has
called for a national incident reporting/learning system as
part of its six-point target safely plan to improve patient
safety in radiotherapy. The American association of physicist
in medicine (AAPM) Work Group on the Prevention of
Errors (WGPE) published the consensus recommendations
for incident learning database structures in radiation oncology in 2012 [3]. Learning radiotherapy incidents or near
misses through incident reporting systems may improve the
patient safety and clinical service quality [4]. Reporting,
acting on, and learning incidents and near misses may also
provide assessment of the success or failure of the quality
assurance program in preventing error. Sharing knowledge

2
about incidents allows better process optimization with
information about likely severity and frequency of specific
errors and helps prioritize quality management initiatives [5
8]. A number of references suggest reporting and learning
radiotherapy incidents or near misses.
While it is generally accepted that incident learning is
important in the radiation therapy setting, implementing
incident reporting and learning represents a considerable
challenge. It requires additional clinical resources and a welldesigned system for report, analysis, and response. What is
more important is the development of a more open mindset
and just culture for reporting near misses and incidents,
with an increased emphasis on incident learning to uncover
latent error pathways. Studies analyzing the effects of incident
learning on patient safety and quality in individual clinics
remain scarce, while several departments reported their
experience [913].
The purpose of this work is to explore the implementation and effectiveness of incident learning using a system
developed based on AAPM recommended database structure
in a new established program of radiation oncology with
advanced technology.

2. Materials and Methods

2.1. Radiotherapy Program Description. Our department of
radiation oncology is a comprehensive clinical and academic department with new established external beam radiotherapy program, delivering 537 treatment courses on
Trilogy (Varian, Palo Alto, CA) in 2012 and 725 treatment
courses on Trilogy and Axesse (Elekta AB, Stockholm,
Sweden) together in 2013. Trilogy and Axesse were put
into clinical use after commissioning on February 2012
and April 2013, respectively. The comprehensive record-andverify/information system of Aria (Varian, Palo Alto, CA)
and Mosaiq (Elekta AB, Stockholm, Sweden) were used
for Trilogy and Axesse, respectively. All the simulation,
planning, delivery, and information systems were newly
installed and commissioned. All the protocols and quality
management programs were established from scratch. For
simulation, Brilliance Big Bore CT (Philips Medical Systems,
Cleveland, USA) was used. Eclipse and Oncentra (Version
3.2.3; Elekta/Nucletron, Veenendaal, Netherlands) treatment
planning systems were mainly used for planning, in addition
to Pinnacle (Philips Radiation Oncology Systems, Fitchburg, WI) and Monaco systems. Most of the staff had
no previous intensity-modulated radiation therapy (IMRT)
and image-guided radiation therapy (IGRT) experience. The
techniques utilized included three-dimensional radiotherapy
(3D-CRT), IMRT, IGRT, volumetric modulated arc therapy
(VMAT), gated treatment, ultrasound guided radiotherapy,
four-dimensional (4D) cone beam computed tomography
(CT), and Hexapod (Schwabmunchen, Germany) treatment
couch with six degrees of freedom. The plans were transferred
electronically to the record and verify system by Digital
Imaging and Communications in Medicine (DICOM). All
the plans were checked by senior physicist after radiation oncologist approval, before treatment. At treatment,

BioMed Research International

the therapist rechecked the transferred data with printed plan
data and treatment chart parameters.
2.2. Development of the Reporting and Analysis System. With
reference to the consensus recommendations for incident
learning database structures in radiation oncology by American Association of Physicist in Medicine (AAPM), an incident learning system was specifically designed and developed for reporting, investigating, and learning of individual
external beam radiotherapy incidents in our department. The
definitions, process maps, severity scales, causality taxonomy,
and data elements were referred from AAPM recommendation. The definitions of terms were clarified. 91 common steps
were identified for external beam radiation therapy, 35 safety
barriers were set. A ten-level dosimetric and medical severity
scale were used to reflect the observed or estimated harm
to a patient and a radiation oncology-specific root causes
table was used to facilitate the root-cause analyses. Near-miss
incidents were assigned with the estimated harm that would
have occurred had the incident reached the patient.
2.3. Clarification of Definitions. We applied the definitions of
terms in accordance with the consensus recommendations
American Association of Physicist in Medicine (AAPM) [3].
An incident was defined as any unwanted or unexpected
change from a normal system behavior which causes or
has the potential to cause an adverse effect to persons or
equipment. A near miss was defined as any event or situation
that could have resulted in an accident, injury, or illness but
did not either by chance or through timely intervention. We
reported both incident and near miss mainly to highlight the
fact that all deviations are potentially of interest even those
that do not necessarily impact the patient. Safety barrier was
defined as any process step whose primary function is to
prevent an error or mistake from occurring or propagating
through the radiotherapy workflow.
2.4. Clinical Implementation. After pilot testing, clinical staff
used the system for reporting the incidents that occurred
in clinical practice. The incidents reported in the system
were then investigated, analyzed, and learned by a multidisciplinary team which includes the leaders of radiation
oncology, radiation physics, and technologists. The priority
of our implementation strategy was to strenuously promote
staff acceptance and participation through emphasis of senior
management participation and expectations with respect
to the establishment of a just environment. It was also to
promote openness about reporting incident, emphasis on
learning and support, and an obligation to act, where all
staff members felt comfortable disclosing incidents while
maintaining professional accountability. The established multidisciplinary team directs the reporting, investigation process and determines the corrective actions and follow-up, as
required to ensuring prompt response and feedback.
The reporting process starts with the initial reporter who
submits the initial description of the event. The event is then
forwarded to supervisors of the involved areas by the team
leader (the chief of the department). In our implementation,

BioMed Research International

3
Table 1: Summary of reported incidents and near misses.

First year
Second year
Total

Incident
3
2
5

Near miss
19
9
28

Patients treated
537
725
1262

even though there may be several areas involved in the

event, each event is processed by only one supervisor who
performs the initial analysis of the event and determines
the best approach for corrective action. Which supervisor
is selected to address an event depends on the main area
of involvement and on which supervisor believes to be in
the best position to address the issue. Each supervisor can
review other supervisors comments and event disposition.
This transparency and active involvement of supervisors in
event processing contributed significantly to the uniformity
of event processing by different supervisors.
The investigation of the incidents was addressed individually by a small number of key supervisors. A full team
investigation is used for the more serious incidents and for
the incidents where the determination for the basic cause and
identification for the most appropriate learning component
requires more rigorous assessment. Immediately after reporting or monthly all reported incidents were reviewed at the
department multidisciplinary quality management forums.
As we collect a broad spectrum of events, from minor near
misses to potentially serious incidents, the event processing
can vary significantly from one event to another, and this
process evolved over time as our clinic gained experience
with the system. Low severity events that have already been
addressed and at first do not appear to require systematic process changes, such as minor work flow variation or deviation
near misses, are typically only classified and cataloged so that
their statistics can be reviewed at a later time. Events that
are indicative of process weaknesses or the need for training
are immediately addressed by the supervisor, who initiates
procedural changes or schedules the necessary training.
These events are typically judged to be essential for the system
operation and corrective actions are relatively simple. The
event severity is reviewed and assigned by the supervisor
during the investigation and review of the event. Events are
subsequently processed based on the severity classification.
More complex events that require significant corrective
actions are forwarded to the department Process Improvement Committee for further discussion and processing.
All relevant cause/contributory factors were investigated
and assigned by the investigating team, from the aspect of
management, technique, personnel, equipment, and process,
since it is a rare situation in which only one single cause is at
work, so that a full root-cause analysis can be implemented.
In addition, each discipline was further discussed or reflected
on the data within their own specialty, focusing on process
breaks and system errors, intent to find means of preventing
reoccurrence. Departmental no blame policy was advertised
repeatedly encouraging staff to report incidents without fear
of retribution. Blaming an individual or group as the only root

Incident (%)
0.56
0.28
0.40

Near miss (%)

3.54
1.24
2.22

Total (%)
4.10
1.52
2.61

cause may lead to corrective actions that fail to address an

underlying cause. An individuals actions must be viewed in
the context in which they act when performing root-cause
analysis. Each reporter can know about all events that they
have submitted along with the analysis performed by the
supervisor and the ultimate disposition of the events.
Learning actions included focused education sessions on
a specific process to highlight error risk, process mapping,
protocol changes, quality control measures and approaches
of prevention of errors and continuous improvement, discussion with specific workgroups to make them aware of
any error risk that has been identified, and presentations at
chart rounds. The incident learning was used for guidance in
reengineering clinical processes. The data and the statistics
collected through the established database were presented
at the department quality improvement conference with the
corrective actions, thus allowing easy communication of
problem details and corresponding process modifications.
The reported external beam radiotherapy near misses and
incidents were presented from February 2012 to February
2014. The reporting trends for the second year were compared
to the events that were reported in the first year.

3. Results
A total of thirty-three reports were analyzed, including 28
near misses and 5 actual incidents. During this period, a total
of 1262 patients were treated. The summary of reported near
misses and incidents are given in Table 1. The average incident
rate per 100 patients treated was 0.4; this rate fell to 0.28% in
the second year from 0.56% in the first year. The rate of near
miss fell to 1.24% from 2.22%.
Table 2 gives the incident category of reported error
type. The rates of all types of incidents/near miss decreased
except the suboptimal plan quality type. Among them, the
wrong patient incident occurred due to the wrong patient
plan calling out with the same last name of two patients,
which was found when the other patient was treated in the
same afternoon. Of the 7 wrong site near misses/incidents,
2 occurred due to the field shape unchanged to conform
to the target projection after adding wedge and adjusting
the collimator angle, not noticed by the planner, found and
corrected during the plan physics review. Three incorrect
shifts from computed tomography (CT) reference marks
were reported, due to the wrong/missing setting the fiducial
reference point when planning, which were caught during
physics check and CT repositioning before treatment. One
geographic miss resulted from forgetting shifting the treatment isocenter after modifying plan with the prescription
dose delivered to the wrong volume. One electronic transfer

BioMed Research International

Table 2: Summary of reported incidents (in parentheses) and near

misses category.

Table 3: Summary of the origin of reported incidents (in parentheses) and near misses.

Category
Wrong patient
Wrong site
Wrong laterality
Wrong dose
Wrong prescription
Suboptimal plan quality
Total

Category
Imaging for planning
(simulation)

First year
(1)
4 (2)
4
4
5
2
19 (3)

Second year
0
0 (1)
2
1
2
4 (1)
9 (2)

Total
1
7
6
5
7
7
33

error of treatment parameters from the treatment planning

system was reported.
Of the 6 wrong laterality near misses/incidents, 4
occurred due to the contouring and planning at the wrong
(healthy) side of the body, 2 due to wrong patient position
selected in the CT scan protocol (head first supine, instead of
foot first supine), which were found during the image guided
verification before treatment.
Of the 7 dose prescription near misses/incidents, 2 near
misses occurred due to the erroneous planning instructions in the prescription, 1 inconsistent prescription in
terms of total dose, dose per fraction, and fractions (PTV
50.40 Gy/1.8 Gy/25 f). The other 4 occurred due to planner
not following the physicians prescriptions, including significant error of prescription of 60.00 Gy/2.0 Gy/30 f, plan with
60.00 Gy/0.3 Gy/200 f.
Of the 5 wrong dose near misses/incidents, 1 dose calculation error occurred due to the misunderstanding of source
axis distance (SAD) factor for manual dose calculation by
the junior physicist, which was found and corrected during
physics review. The second near miss was an error in absolute
dose calibration of electron beam during commissioning
of Trilogy which would result in approximately 60 patients
treated with a dose 5% below that prescribed per year if not
found. This near miss was caused by wrong understanding
of the calibration protocol, found, and corrected in physics
review. As the cause was a single error, this incident was
categorized as one incident despite the impact on a large
number of patients if not detected. The other reasons of
wrong dose error included insufficient scan field of view, not
enough air flash in planning.
The reasons of the 7 reported suboptimal plan quality
near misses/incidents included incomplete contouring of
organs at risk, not optimal in selection of beam energy,
margin setup, intermediate, and low dose distribution.
The incident breakdown by origin is shown in Table 3. The
proportion of incidents is highest at the planning stage. One
near miss and two incidents occurred in the safety barrier of
the radiotherapy process, including one incident in verification of patient ID before treatment delivery, one incident in
image-guided verification (we used CT simulator for couch
shift and isocenter determination in our department.) after
plan modification, and one near miss in commissioning for
electron dose calibration.

First year

Second year

Total

treatment planning
Plan transfer
Commissioning
Treatment delivery

13 (1)
(1)
1
(1)

9 (2)
0
0
0

25
(1)
1
(1)

Total

19 (3)

9 (2)

28 (5)

Table 4: Dosimetric severity distribution of reported incidents (in

parentheses) and near misses.
Category
1/2
3/4
5/6
7/8
9/10
Total

First year
5 (3)
2
3
3
6
19 (3)

Second year
5 (2)
1
3
0
0
9 (2)

Total
10 (5)
3
6
3
6
28 (5)

Table 5: Medical severity distribution of reported incidents (in

parentheses) and near misses.
Category
0
1
2
8/9
10
Total

First year
(1)
7 (2)
1
9
2
19 (3)

Second year
0
6 (2)
0
3
0
9 (2)

Total
(1)
13 (4)
1
12
2
28 (5)

Of the 33 incidents and near misses reported, 5 originated

in imaging for planning, including error/miss marking reference point on patient in software for three patients, error
in input patient position (head first instead of foot first) in
CT scanner during CT image acquisition for two patients.
Of the 25 incidents/near misses originated in treatment
planning, seven originated in the preliminary prescription
parameters, including 3 wrong prescriptions from oncologist
and 4 due to planners not following physicians prescription
in terms of fractions, dose per fraction, and total dose. 11 near
misses/incidents originated in dose distribution optimization
leading to suboptimal plan. The other near misses/incidents
included 4 due to targets contoured and planned on the
wrong side (healthy side), 1 organ at risk (spinal cord)
incompletely contoured leading to overdose if delivered, 1
treatment plan parameter transfer error due to the network
issue, and 1 manual dose calculation error for a patient
simulated with conventional X-ray simulator, due to the
wrong understanding of SAD factor.
The dosimetric and medical severity distribution of
reported near misses and incidents are given in Tables 4 and
5. Five reported incidents were all classified as grade 1/2 of

BioMed Research International

Table 6: Summary of causes/contributory factors for reported incidents (in parentheses) and near misses.
Category
1biv
1cii
1ciii
1civ
1div
2bii
2biv
3f,
6aiii
6div
6eii
7
Total

policy not followed

Appropriate skills not acquired from vendor provided training
inadequate periodic assessment of staff competencies
lack of continuing education
Communication
equipment design and construction issues
networking problems
Negligence
failure to detect a developing problem
failure to develop an effective plan
failure to execute the planned action
Other

First year
7 (5)
2 (1)
3
3
2
1
(1)
12 (3)
1
1
(1)
0
32 (11)

Second year
3
1
1
1
1
0
0
4
0
4
0
2
17

Total
10 (5)
3 (1)
4
4
3
1
(1)
16 (3)
1
5
(1)
2
49 (11)

Comments: The data indicate the number of causes/contributory factors, not the number of near misses/incidents.

dosimetric severity, 1 as grade 0, and the other 4 as grade 1 of

medical severity.
The summary of the root causes and contributory factors
to the near misses/incidents are given in Table 6. Among
them, negligence, policy not followed, training and failure to
develop an effective plan, and communication contributed to
19, 15, 12, 5, and 3 near misses/incidents, respectively.

4. Discussion
Our results demonstrated that incident learning can be
used for the safety and quality management of radiotherapy,
even for a department with new established program with
advanced technology, new equipment from different vendors,
no much safety, and quality management experience. Implementation of an effective incident learning system may serve
to reduce the occurrence of actual incidents and enhance the
culture of safety at the individual health care professional
level and at the multidisciplinary team level by addressing
quality improvement initiatives collaboratively with transparent accountability. Incident learning also improved event
communication and identification of clinical areas which
needed process and safety improvements and encouraged
the reporting of potential incidents as a proactive means of
enhancing safety and quality in a radiation treatment program. The reported data were also useful for the evaluation of
corrective measures and recognition of ineffective measures
and efforts.
Implementing Incident learning in radiotherapy is a
systematic and complicated project. A rigorous system of
learning, feedback, and action are required for this approach
to have a meaningful impact on patient care. The corresponding departmental infrastructure and facilities, organization,
and culture are needed. The related academic society, organization, and state health administrative department should
encourage and protect the reporting and learning of near
misses/incidents by advocacy, regulation, and legislature.

Step by step improvements are needed for incident learning

fully implemented in radiotherapy.
AAPM recommended database structure provided a very
good reference for establishing individualized database for
individual institution. A reporting system easy to use, file,
report, respond, and analyse could be established based
on that considering of the context of individual department. The individual events could be quickly and easily
reported without disrupting clinical work. The specially
designed incident reporting systems in a radiotherapy setting
can provide valuable data for process and patient safety
improvement. There are some differences in the work process
map and quality control program for different institutions,
such as using CT simulator for couch shift and isocenter
determination before treatment in our department, in which
7 near misses were found and corrected. There is no this
process in the AAPM recommended database structure. Only
one near miss and two incidents occurred in the safety
barrier of the radiotherapy process of the 33 reported near
misses/incidents. Erroneous prescription instructions and
failure to follow the physicians prescriptions lead to 7 dose
prescription near misses/incidents, but no safety barriers
were set from prescription and planning to plan review
whether for physician, planner, or physicist. Only physics
review was set as safety barrier in the AAPM recommended
structure. So, the setting of safety barrier should be catered
to the individual context of the department. Safety barriers
should be added before plan review to reduce errors and delay
and improve efficiency. Cox et al. [14] found modification
of 36% contouring or prescription in the round after several
treatment fractions. So, they suggest that the contouring and
prescription review should be completed before planning.
Although Ford et al. [15] found that physics review was the
most effective safety barrier before treatment delivery, which
can find approximately 60% errors, image-guided verification
before treatment delivery is another important safety barrier,
which found 7 near misses/incidents in our study.

6
In addition, the severity assignment of an actual incident
or near miss is difficult, especially for the near-misses since
one has to estimate the harm that would have reached
the patient several steps down the chain of events. The
dosimetric severity scale could not fully be expressed by
dose; it will be better if evaluated with biological effective
dose. For a near miss reported in our study, in which the
physicians prescription of 60.00 Gy/2.0 Gy/30 f was planned
with 60.00 Gy/0.3 Gy/200 f, the dose was the same, but it
would result in severe effect if not found.
33 near misses/incidents were reported in two years in
this study. The volume of reports varies with the report
criteria, quality and safety culture, equipment, and techniques
used for different institutions. Mutic et al. have observed
an incident report rate of 1 per 1.6 patients treated (this
includes both incidents that reach the patient and near-miss
incidents that are intercepted before reaching the patient)
[16]. We collected a large number of incidents including near
misses with very limited or zero clinical impact on patients,
such as the plans of suboptimal plan quality, efficiency, beam
energy, beam orientation, susceptibility to setup error, and
organ motion, which were found and corrected prior to
treatment. Such an approach facilitates continuous proactive
improvement which can lead to the correction of small and/or
latent system weaknesses before they result in much more
severe events, to improve the safety and quality of care by
supporting the systematic learning from errors [17, 18].
The rate of reported near misses/incidents decreased
significantly in the second year, although more near misses of
suboptimal plan quality were reported due to the enhanced
quality and safety culture in the second year. The high
rate of reported near misses/incidents in the first year may
be attributed to the introduction of new equipment, new
techniques, and new staff. Through the analysis of the
occurrence of near misses/incidents, the cause/contribution
factors, we found the weaknesses in the clinical process and
implemented series of proactive measures to enhance the
safety and quality. At the start of 2013, we introduced a more
comprehensive check list in an attempt to reduce these errors
and the error rate fell dramatically. Morganti et al. report a
reduction in error through the use of independent checks
[19]. Clear communication program et al. also contributed
to the decrease, which included the efforts focusing on communication ambiguities (e.g., technicians not informed of
cancelled treatment, changed plan or special appointment,
or change in number of treatments, doctor unavailable for
assigned appointment, etc.) and unclear physician directive,
and additional policy and procedure changes addressing
verbalization of treatment parameters prior to treatment
delivery, laterality and documentation of change in prescription, and planning. Other examples of such interventions
include changes in staffing levels to concentrate effort in more
vulnerable parts of the process and modification of the quality
assurance processes to focus on weaknesses in the treatment
preparation process, enhance staff training, competency evaluation and supervision, strictly follow clinical protocol, and
standardization of processes.
An important feature of the incident learning system is
that it requires an investigation of sufficient depth to discover

BioMed Research International

the basic causes of an incident according to a predetermined
categorization and thereby facilitate the determination of
meaningful corrective actions. An added benefit of such
an approach is that it enables the identification of system
problems or basic causes that could precipitate a range of
different incidents. If the basic causes are addressed, it can be
expected that overall system safety is enhanced and not just a
particular weakness associated with a particular incident. The
distribution of basic causes demonstrated that negligence and
training contributed the most to the errors. It is understood
for the new established program with new staff. The decrease
of the number of these factors also demonstrated the effect
of enhanced training and continuing education. Following
that was policy not followed which was associated with
a significant proportion of errors. The number of near
misses/incidents assigned a basic cause relating to policy
not followed was 12 and 3 for the first and second year,
respectively. The reducing of these numbers indicated that
the culture of following policies and procedures may be
improving, which was also shown as the rising number
of suboptimal plan quality near misses. There were still a
significant number of near misses occurred due to failure
to develop an effective plan, which showed the challenge
of long term continuing education and personnel capacity
improvement.
Systematic, scientific, efficient, and feasible management
tool and approach is needed for the safety and quality management of radiotherapy. The aim of the incident learning was
to capture of potential as well as actual incidents, engender a
culture of safety, and support process improvement in patient
care and safety. Incident learning can be used individually or
combined with Failure Modes and Effects Analysis (FEMA),
PDCA circle, and six sigma. Incident learning permits the
implementation of both proactive and reactive error management initiatives, leading to quality improvement in all
aspects of the clinics operation and this strategy. A clear
understanding of the potential consequences and relationships between different incident types will guide incident
reporting, resource allocation, and risk management efforts.
Additional work is needed to develop methods that can more
effectively utilize reported data for process improvement.

5. Conclusions
Our results show that implementation of an effective incident
learning system may serve to reduce the occurrence of
actual incidents, enhance the culture of safety, and encourage
the reporting of potential incidents as a proactive means
of enhancing safety and quality in a radiation treatment
program. Incident learning can be used for the safety and
quality management of radiotherapy according to our primary experience.

Conflict of Interests
None of the authors have any actual or potential conflict of
interests related to this work.

BioMed Research International

References
[1] P. J. Pronovost, C. A. Goeschel, K. L. Olsen et al., Reducing
health care hazards: lessons from the commercial aviation safety
team, Health Affairs, vol. 28, no. 3, pp. w479w489, 2009.
[2] J. Cunningham, M. Coffey, T. Knoo s, and O. Holmberg, Radiation oncology safety information system (ROSIS)profiles of
participants and the first 1074 incident reports, Radiotherapy
and Oncology, vol. 97, no. 3, pp. 601607, 2010.
[3] E. C. Ford, L. F. de los Santos, T. Pawlicki, S. Sutlief, and P.
Dunscombe, Consensus recommendations for incident learning database structures in radiation oncology, Medical Physics,
vol. 39, no. 12, pp. 72727290, 2012.
[4] M. V. Williams, Radiotherapy near misses, incidents and
errors: radiotherapy incident at Glasgow, Clinical Oncology, vol.
19, no. 1, pp. 13, 2007.
[5] M. S. Huq, B. A. Fraass, P. B. Dunscombe et al., A method
for evaluating quality assurance needs in radiation therapy,
International Journal of Radiation Oncology Biology Physics, vol.
71, no. 1, pp. S170S173, 2008.
[6] F. Rath, Tools for developing a quality management program:
proactive tools (process mapping, value stream mapping, fault
tree analysis, and failure mode and effects analysis), International Journal of Radiation Oncology Biology Physics, vol. 71, no.
1, pp. S187S190, 2008.
[7] E. C. Ford, K. Smith, K. Harris, and S. Terezakis, Prevention
of a wrong-location misadministration through the use of an
intradepartmental incident learning system, Medical Physics,
vol. 39, no. 11, pp. 69686971, 2012.
[8] L. J. Peters, B. OSullivan, J. Giralt et al., Critical impact of
radiotherapy protocol compliance and quality in the treatment
of advanced head and neck cancer: results from TROG 02.02,
Journal of Clinical Oncology, vol. 28, no. 18, pp. 29963001, 2010.
[9] A. Arnold, G. P. Delaney, L. Cassapi, and M. Barton, The use
of categorized time-trend reporting of radiation oncology incidents: a proactive analytical approach to improving quality and
safety over time, International Journal of Radiation Oncology
Biology Physics, vol. 78, no. 5, pp. 15481554, 2010.
[10] B. G. Clark, R. J. Brown, J. L. Ploquin, A. L. Kind, and
L. Grimard, The management of radiation treatment error
through incident learning, Radiotherapy & Oncology, vol. 95,
no. 3, pp. 344349, 2010.
[11] E. U. Ekaette, R. C. Lee, D. L. Cooke, K. Kelly, and P. B.
Dunscombe, Risk analysis in radiation treatment: application
of a new taxonomic structure, Radiotherapy and Oncology, vol.
80, no. 3, pp. 282287, 2006.
[12] T. K. Yeung, K. Bortolotto, S. Cosby, M. Hoar, and E. Lederer,
Quality assurance in radiotherapy: evaluation of errors and
incidents recorded over a 10 year period, Radiotherapy and
Oncology, vol. 74, no. 3, pp. 283291, 2005.
[13] J. Bissonnette and G. Medlam, Trend analysis of radiation
therapy incidents over seven years, Radiotherapy and Oncology,
vol. 96, no. 1, pp. 139144, 2010.
[14] B. W. Cox, A. Sharma, L. Potters, and A. Kapur, Prospective
contouring rounds: a novel, high-impact tool for optimizing
quality assurance, International Journal of Radiation Oncology,
Biology, and Physics, vol. 87, no. 2, p. S118, 2013.
[15] E. C. Ford, S. Terezakis, A. Souranis, K. Harris, H. Gay, and S.
Mutic, Quality control quantification (QCQ): a tool to measure
the value of quality control checks in radiation oncology,
International Journal of Radiation Oncology Biology Physics, vol.
84, no. 3, pp. e263e269, 2012.

7
[16] S. Mutic, R. S. Brame, S. Oddiraju et al., Event (error and nearmiss) reporting and learning system for process improvement
in radiation oncology, Medical Physics, vol. 37, no. 9, pp. 5027
5036, 2010.
[17] P. Dunscombe, Recommendations for safer radiotherapy:
whats the message? Frontiers in Oncology, vol. 2, article 129,
2012.
[18] M. V. Williams, Improving patient safety in radiotherapy by
learning from near misses, incidents and errors, British Journal
of Radiology, vol. 80, no. 953, pp. 297301, 2007.
[19] A. G. Morganti, F. Deodato, S. Zizzari et al., Complexity
index (COMIX) and not type of treatment predicts undetected
errors in radiotherapy planning and delivery, Radiotherapy &
Oncology, vol. 89, no. 3, pp. 320329, 2008.