Cover Image

211


Chapter 13. Identifying Issues and Investigating Incidents Case
Study


Biomedical technology professionals are often the first people to identify issues with
a technology. Sometimes they identify them proactively, during regular inspections, and
sometimes they find them retrospectively, when they are contacted by front line staff for
assistance. In both of these scenarios, biomedical technology professionals are well
positioned to apply human factors methods to identify human factors issues. This chapter
will use case studies to showcase two human factors informed methods that can be used to
help identify human factors issues with technology either retrospectively (HFRCA), or
prospectively (HFFMEA).


Section 13.1. Retrospective Incident Investigations: HFRCA
To illustrate how the biomedical technology professional can approach incident


investigations from a human factors perspective, the human factors informed root cause
analysis (HFRCA) framework will be applied to a case study (Case Study 3) that occurred in
Canada in 2006.


Chemotherapy Overdose Results in Patient Death


On July 31, 2006, a 43-year old woman underwent her first cycle of adjuvant
intravenous (IV) chemotherapy treatment to reduce the likelihood of recurrence of
nasopharyngeal cancer. Previously, she had received two months of combined
chemotherapy and radiation treatment, and although her cancer was advanced, the
planned treatment was expected to be effective.


On the morning of her first cycle of adjuvant treatment, she arrived at the cancer
centre and received hydration and anti-nausea medications intravenously, followed by the
intravenous chemotherapy drug, cisplatin. This drug was followed by a post-hydration
medication and another chemotherapy drug, fluorouracil. A high dose of fluorouracil was to
be given to the patient slowly, over the course of four days. So the patient would not have
to stay in the hospital for this infusion, the fluorouracil was to be given intravenously using
an ambulatory infusion pump.


The nurse calculated the required medication delivery rate for the fluorouracil,
programmed the ambulatory infusion pump, and asked a second nurse to double check her
calculation and that the pump had been programmed correctly. Both nurses signed off on
the required documentation, and the patient’s nurse connected the patient to the
ambulatory infusion pump and started the infusion. The nurse instructed the patient to




212


return in four days, at which point she would be disconnected from the ambulatory
infusion pump.


The patient left the clinic, and about four hours later she heard the infusion pump
beeping. When she checked the pump she saw that the entire bag of fluorouracil was
already empty. Instead of infusing over four days, the medication had been delivered to the
patient in just four hours.


The patient returned to the cancer centre where the pump was disconnected and
her line was flushed. The physician on call was notified and he indicated that unfortunately
nothing could be done to reverse the overdose, as there was no antidote. Tragically, the
patient died on August 22, 2006 from “complex causes, including a failure of multiple
organs, as well as widespread internal bleeding”.


From: ISMP Canada Fluorouracil Incident Root Cause Analysis, May 2007. Available
at https://www.ismp-canada.org/download/reports/FluorouracilIncidentMay2007.pdf


Case Study 3. Chemotherapy overdose resulting in patient death


Shortly after this tragic incident, ISMP Canada was asked to investigate and identify
underlying factors that could have contributed to the event, in the hope that other similar
events could be prevented in the future. Correspondingly, when a sentinel event occurs in
your own healthcare organization, it is highly recommended that an HFRCA be completed,
not only to fulfill legislative or accreditation requirements, but also to reduce the likelihood
of other similar events occurring again.


Prior to conducting an HFRCA, the biomedical technology professional should ensure
they have the support and buy-in of senior level management to increase the chance for
positive change as a result of the analysis. It may be helpful to share RCA reports from
other investigations, such as the ISMP Canada RCA cited in Case Study 4, to illustrate the
possible output and impact. In the case of the fluorouracil incident described above, the
Chief Medical Officer of the organization took immediate action by declaring the tragedy a
systems failure, apologizing to the family, and requesting that ISMP Canada conduct a
formal RCA. Senior management in this case was keenly aware of the importance of
managing risk at a systems level, rather than at a person level.


The ISMP Canada RCA [94] of the fluorouracil incident is an excellent example of
how to conduct a RCA. While ISMP Canada does not refer to it as an HFRCA, a human factors
specialist was asked to participate on the ISMP Canada RCA team, and he performed
several HF methods covered in this book including Observations (Chapter 4), Interviews
(Chapter 5) and Usability Testing (Chapter 8). The report, therefore, included a description




213


of many human factors issues that were deemed to have contributed to the identified
systems failures.


Section 13.1.1. Determine Whether HFRCA is Required
To determine whether it is appropriate to conduct a RCA in this case (i.e., there are


systems issues that must be identified and addressed) the Incident Decision Tree (Figure
26) should be applied. It is clear in this case that conducting a HFRCA is appropriate
because: the nurse did not intend to cause harm, there was no evidence of ill health or
substance abuse, the nurse does not appear to have departed from agreed protocols or safe
procedures, and others in a similar situation could make the same mistake. Consequently,
this sentinel event is a systems failure rather than a person-centered issue and thus
requires an analysis of the risks in the system to identify mitigating strategies.


Section 13.1.2. Secure Items
Based on the description of the case, the items considered important to secure were:


• The ambulatory infusion pump
• The bag of chemotherapy
• The tubing sets used
• All medication labels
• The medication order
• The patient’s records
• Any notes or papers used for the calculation


After securing these items, photographs were taken, and lot numbers, serial
numbers, and expiration dates were recorded. Pump logs were saved for future review.


Section 13.1.3. Establish the Team
If a HFRCA were conducted by an internal team at the healthcare facility, it should


include a pharmacist, one or more oncology nurses, an oncologist, a risk manager, a
biomedical technology professional, and someone with human factors expertise (could be
the biomedical technology professional). The team should be assembled at the request of a
senior hospital administrator who will also receive reports on the activities of the team.


In the case of the RCA conducted by ISMP Canada, the team consisted of five health
care professionals: three pharmacists with expertise in medication safety; an oncology
nurse; and a physician who was also a human factors engineer.


A senior leadership representative from the healthcare institution would be an
excellent addition to the team. In the case of this incident, the Chief Medical Officer fully
supported the completion of this RCA. Support and awareness of RCA activities by senior
leadership is essential to realizing positive change at the healthcare organization following
a sentinel event.




214


Section 13.1.4. Develop Initial Understanding of Incident
The initial understanding of the incident described in the ISMP Canada RCA [52] as


follows:


“A woman in her 40s died last week after she was mistakenly given a lethal overdose of a
standard chemotherapy drug while undergoing treatment at the XXXX Cancer Institute
Instead of receiving the intravenous drug continuously over four days, the woman received
the dose over four hours on July 31 from a pump that had been programmed in error. She
died Aug. 22 at the University of XXXX Hospital from complex causes, including a failure of
multiple organs, as well as widespread internal bleeding.”


From: XXXX. We cannot eliminate human error. XXXX Journal, Thursday, August 31, 2006.


To help create this succinct statement, a process flow diagram was created to
support the development of the initial understanding of the incident (Figure 43). The
information used to create this type of process flow diagram is usually generated by
conducting observations of the work environments responsible for all tasks related to
ordering, preparing and administering ambulatory intravenous chemotherapy and
conducting interviews with staff about what happened on the day of the incident.
Conducting interviews following an incident can be difficult, both for the interviewer and
especially for the interviewees. Careful consideration should be given to who conducts the
interviews, where the interviews take place, the specific questions that are asked, and how
the interview is positioned to the interviewee.




215




Figure 43. Process flow diagram based on the teams’ initial understanding of the incident


Creating this diagram was just the first step in understanding the incident. Once
created, it was enhanced iteratively through subsequent interviews, an examination of the
physical environment, usability testing, and a search for information about other similar
incidents.


In the case of the RCA conducted by ISMP Canada, the initial understanding of the
incident was informed by:


• Interviews with:
• Corporate executive team members
• Senior leadership
• Pharmacy administrators
• Internal critical incident review team members
• Nursing and medical staff directly involved in the incident
• Nursing and medical staff indirectly involved in the incident
• Nursing and medical staff knowledgeable about the typical care


process




216


• Front line staff
• Biomedical engineering manager
• Medical staff from the Intensive Care Unit where the patient was


transferred following the incident
• Staff from the patient residence in the community, where the patient


had stayed during the ambulatory portion of her chemotherapy
treatment


• A representative from the provincial Health Quality Council, who also
conducted an external review of the incident


• An examination of the physical environment where the incident took place


• Observations of typical work processes in the Medical Clinic, Treatment Area,
and Pharmacy


• A usability test of the tasks associated with setting up and programming the
ambulatory infusion pump.


• A search for information about similar incidents that may have occurred
nationally, or internationally.


In addition to incorporating information from these data collection exercises, other
contextual information was included in the process flow diagram such as notes about
artefacts, timing, and a comparison of the actual, typical and expected workflows (Figure
44). In this case, one of the many factors identified as contributing to the incident was a
missed step in the calculation resulting in the programming of a medication delivery rate
that was 24 times too fast. Figure 44 shows this missed calculation step on the diagram, but
does not tie it in with the actual events. Time data indicated on the process flow diagram in
Figure 44 are not accurate, and have been included for illustrative purposes only.




217




Figure 44. Updated process flow diagram based on data collected through interviews,
observations, usability testing, and information searches.


In addition to a process flow diagram, a factual description of the events leading to
the incident should then be created. These descriptions are useful to aid in systematically
thinking through potential failure modes across the entire workflow. The following list
provides an abridged summary of the events leading to the adverse event, with a more
complete summary available in the ISMP RCA report [94].


• The patient received her pre-hydration, pre-medications, cisplatin, and post-
hydration according to the typical prescribed protocol.




218


• Following the post-hydration infusion, nurse #1 calculated the required
medication delivery rate for the patient’s fluorouracil infusion. To do so, she
used the dose ordered over four days (5,250 mg), the total duration of the
infusion (4 days), and the final concentration (45.57 mg/mL). A rate of 28.8
mL/h was calculated, and was observed to match a number printed on the
pharmacy drug label. The calculation was done using a calculator available
on a computer.


• Nurse #1 entered the rate of 28.8 mL/h into the ambulatory infusion pump.


• Nurse #1 requested a second check to verify the correctly calculated rate of
drug delivery and pump programming.


• Nurse #2 came to do the check but could not find a calculator, so she did the
calculation both mentally and on paper. Nurse #2 confirmed the calculation
and pump programming before locking the pump.


• Nurses #1 and #2 each signed the handwritten medication administration
record, documenting the total dose of fluorouracil as 5,250 mg.


• Nurse #1 signed off electronically on the total dose in the computer.


• Nurse #1 started the infusion, reviewed the pump functionality with the
patient, and instructed her to return to the cancer centre in 4 days.


• About four hours after the patient left the cancer centre the pump started
beeping because the bag of fluorouracil was empty.


• The patient contacted the cancer centre, and later returned to the cancer
centre, where the evening shift Nursing Supervisor disconnected the pump
and flushed the patient’s line.


• The Nursing Supervisor contacted the physician on call, who advised that
nothing could be done. The Nursing Supervisor completed a paper incident
report and submitted it, with the pump, to the Chemotherapy Treatment
Clinic.


• The following morning, the Unit Manager and Nurse #1 reviewed the pump
history and verified that the pump had been programmed at the incorrect
rate. The pump should have been programmed at a rate of 1.2mL/h, but was
programmed at 28.8mL/h - a rate that was 24 times higher than intended.




219


Section 13.1.5. Identify Contributing Factors
The factual description of events highlights that several contributing factors across


the system led to the occurrence of this incident. The human factors framework adapted
from Reason’s Swiss Cheese Model and Vicente’s Human-tech ladder (Section 10.6.5.1) is
helpful for identifying and documenting contributing factors across the levels of the
system. The Swiss Cheese/Human-tech illustration for this incident is included in Figure
45.




Figure 45. Using Reason’s Swiss Cheese Model and Vicente’s Human-tech ladder to identify
contributing factors to a sentinel event.


Additionally, the Joint Commission RCA Action Plan Tool helps to systematically
identify contributing factors by asking a series of questions. A brief excerpt of the analysis
questions from the Joint Commission RCA Action Plan Tool that were used to help identify
additional contributing factors are included below in Table 28.




220


Table 28. Excerpt of analysis questions from the Joint Commission RCA Action Plan Tool for
the fluorouracil incident


Analysis Question Description based on incident


2 Were there any steps in the process
that did not occur as intended


Nurse #2 looked for a calculator to do
her calculation, but could not find one




3 What human factors were relevant to
the outcome?


Confirmation bias:


Information about rate per hour and
rate per 24 hours was available on
label, and matched what both nurses
calculated




… … …




Other tools that are helpful for analyzing and document contributing factors include
an Ishikawa diagram, a tree diagram, or a constellation diagram (Section 10.6.5.2).


In this case, the ISMP Canada RCA team also created a number of causal statements
to summarize the contributing factors leading to the incident. A selection of the most
critical factors that contributed to (1) the miscalculation, (2) the false confirmation of the
information on the label and (3) the pump being programmed in accordance with the
miscalculation, are included here.


(1) Factors contributing to the miscalculation


• Nurses were used to performing complex calculations involving multiple
dimensions, even though the information was available on the medication label. The
nurses at this institution did this calculation as a double check to catch any issues
that might have been introduced upstream.


• Nurse #1 had never administered fluorouracil in this way before and so was not
suspicious of the calculated value; this was the first time the nurse had ever
administered this protocol.


• The calculated rate of 28.8 mL was not unusual for other intravenous infusions
administered in the chemotherapy treatment clinic.




221


• Nurse #1 did not verify the calculated rate with a mental approximation (e.g., if the
total volume in the bag was 130 mL, and the infusion was to be given at a rate of
approximately 30 mL/h, the infusion would only last for about 4 hours rather than
the intended 96 hours).


(2) Factors contributing to a false confirmation of the miscalculation


• The medication label (Figure 46) contained information about two different rates,
including a rate per hour and a rate per 24-hours, increasing the opportunity for a
false confirmation of the miscalculated rate.


• Ambulatory infusion pumps used previously at the institution were programmed in
mL/24 h.




Figure 46. Medication label for fluorouracil infusion containing two
different rates


• The double-checking process was not standardized to support an independent
check by Nurse #2, and there was no checklist or documentation required to
support the calculation.


• The double-checking process was not truly independent with documentation of
independent mathematical calculations.



• Nurse #2 did not verify the calculated rate with a mental approximation. (e.g., if the


total volume in the bag was 130 mL, and the infusion was to be given at a rate of
approximately 30 mL/h, the infusion would only last for about 4 hours rather than
the intended 96 hours).


• There was no calculator readily available to Nurse #2, so the calculation was done
on a scrap piece of paper.




222


• The format of the medication label reflected pharmacy’s interpretation of the legal
requirements and professional guidelines for labeling medications. Human factors
principles were not taken into account to ensure the contents reflected pump
programming requirements and that other factors, such as optimal font size, style,
appropriate use of white space, etc., were incorporated.


(3) Factors contributing to the pump’s inability to detect the calculation error


• The pump used at the cancer centre did not have built-in safeguards to prevent
users from programming a rate exceeding a specified maximum value for a
particular drug. This was true of all electronic ambulatory infusion pumps
available on the market at the time.


Section 13.1.6. Develop Mitigating Strategies
Once the contributing factors are identified, mitigating strategies to address those


factors must be identified. There is no single approach for developing mitigating strategies,
and often this is an iterative process, and one that requires careful consideration of
resources, feasibility, accountability and, most importantly, effectiveness. Section 10.6.6
describes several approaches to developing mitigating strategies and potential pitfalls
associated with this task.


Several recommendations were put forth to address the contributing factors
identified through the HFRCA. For a complete listing of the recommendations identified by
ISMP Canada, see the ISMP Canada RCA [69]. One such recommendation was that in the
absence of “smart pump” technology for ambulatory infusion pumps, other safeguards
should be put in place to ensure that programming parameters fell within a safe range for
high-risk medications. Since no electronic pumps on the marketplace had this capability at
the time, another option was to migrate to the use of elastomeric, rather than electronic
pumps. When this solution was considered in the context of the Hierarchy of Effectiveness
(Chapter 3) it was determined to be a systems-focused solution, and so likely to be more
effective than some of the other person-focused solutions the team had identified. It turned
out, however, to be less of a fail-safe solution than anticipated, as is discussed in the FMEA
case study in the next section.


In a typical RCA, a list of prioritized RCA action items (Table 28) is captured with
progress being tracked using a spreadsheet outlining necessary follow through actions and
timing (Table 29).




223


Table 29. Excerpt of prioritized list of HFRCA actions


Recommended
Action


(category)


Risk
(severity


assessment)


Hierarchy of
Effectiveness


(high,
medium, low
leverage)


Predictors of
Success


(alignment,
existing


mechanisms,
quick wins)


System
Level


Targeted


Evidence
Available?


What
Type?


Confirm
Validity,


Feasibility


Order of
Priority
or Time
Frame


Migrate from
electronic AIPs
to elastomeric
AIPs


Medium High Recent pump
procurement
project in
another unit
could be used
as a basis for
evaluation


Unit-wide
rather
than
organizat-
ion wide


Little
evidence
found


Talk with
other
hospitals
using
elastomer
-ics


Inter-
mediate




Table 30. Excerpt of follow through actions and timing for HFRCA


# Recommendation
Source
and ID


#


Date
Entered


Progress
Status


Timefram
e (end
date)


Target
Area


Risk
Level


Individual
Responsible


1 Migrate from
electronic AIPs to
elastomeric AIPs


1A 09/09/06 03/01/07 Chemo
Treatment
Area


Medium Peter




A report is created to summarize the process, findings, and action items stemming
from the HFRCA. The final ISMP Canada RCA report [94] is an excellent resource that
contains further detail.


Section 13.2. Proactive Systems Improvement Following an Incident
(HFFMEA)
Following a retrospective analysis such as an HFRCA, it can be beneficial to conduct a


prospective analysis using HFFMEA. When a prospective technique is applied following a
retrospective analysis, the opportunity to identify general risks, not directly involved in the
incident, is presented. Further, a prospective analysis method like HFFMEA can be applied
following an HFRCA to examine the potential for new risks to be introduced into the system
as a result of planned changes and mitigating strategies. Case Study 4 expands on Case
Study 3 and will be used to illustrate how the biomedical technology professional can use
HFFMEA to conduct a prospective risk analysis.




224


Implementing Elastomeric Ambulatory Infusion Pumps


Following the incident described in Case Study 3(Chemotherapy Overdose), a
retrospective analysis was conducted using HFRCA (Chapter 10). One of the root causes
identified through the analysis was that the electronic ambulatory infusion pumps in use at
the time of the incident did not have built-in safeguards to prevent programming errors
from occurring. Based on this issue, a recommendation was put forth that the healthcare
organization should start using pumps with built-in safeguards to prevent programming
errors.


At the time of the incident, there were no electronic ambulatory infusion pumps
with built in safeguards available on the market. Consequently, the healthcare organization
considered other options, such as elastomeric pumps (Figure 47). Unlike electronic pumps,
elastomeric pumps are mechanical and do not require any pump programming. However,
prior to moving from electronic to elastomeric ambulatory infusion pumps, the healthcare
organization wanted to understand what risks were associated with the use of these
devices, and so an HFFMEA was to be undertaken at the healthcare organization.


Case Study 4. Identifying Risks Associated with Elastomeric Ambulatory Infusions Pumps




Figure 47. An elastomeric ambulatory infusion pump


Section 13.2.1. Select a Process
The chosen process for this particular HFFMEA was administering chemotherapy


using an elastomeric ambulatory infusion pump. This process was chosen because from the
fluorouracil incident it was known that the electronic ambulatory infusion pumps in use at
the time did not contain any built-in safeguards to ensure the parameters entered for pump
programming fell within acceptable ranges. The reason the institution was considering




225


switching to an elastomeric pump was to prevent these types of pump programming errors
from occurring as chemotherapy was set up and administered using the pump.


Consequently, the starting point for the process was chosen to be as the nurse
received the pump filled with chemotherapy from pharmacy, and the ending point for the
process was chosen to be as the patient left the chemotherapy treatment chair.


To keep the analysis focused and scope manageable, the following inclusion and
exclusion criteria were defined (Table 31):


Table 31. Inclusion and exclusion criteria to define process scope


Inclusion and Exclusion Criteria


Patient Population


Inclusions: Adult patients receiving chemotherapy


Exclusions: Adult patients not receiving chemotherapy, paediatric patients, clinical trials
patients, and other special cases


Location/Environment:


Inclusions: Outpatient treatment clinic of cancer centre


Exclusions: Inpatient cancer treatment areas, pharmacy, physician’s clinics, community,
home


Staff Population:


Inclusions: Chemotherapy nurses working in the outpatient treatment clinic


Exclusions: Chemotherapy nurses not working in the outpatient treatment clinic,
pharmacists, physicians/oncologists, community health care workers, home care workers


Tasks:


Inclusions: Receive filled elastomeric pump from pharmacy, check five rights, connect
pump to patient, start infusion, check the pump is infusing


Exclusions: Ordering chemotherapy, checking order, picking supplies to make chemo order,
mixing chemotherapy order, checking chemotherapy mix


Equipment:


Inclusions: Elastomeric ambulatory infusion pumps and associated tubing/supplies


Exclusions: Large volume infusion pumps, electronic ambulatory infusion pumps




226


Thus, the process scope was to include adult patients receiving chemotherapy
treatment at the outpatient treatment clinic of the cancer centre, from the time the nurse
receives the mixed chemotherapy and elastomeric ambulatory infusion pump from the
pharmacy to the point at which the patient leaves the treatment chair with the infusion
running.


Section 13.2.2. Assemble a Team
Although the process scope only included chemotherapy nurses and processes


contained within the chemotherapy treatment clinic, it was essential that people from
outside of this process scope were included as part of the HFFMEA team.


To complete this HFFMEA, the following team members were chosen and recruited:


Work Team:


• Front line chemotherapy nurse
• Biomedical technology professional
• Human factors specialist


Advisory Team:


• Nursing manager for outpatient chemotherapy treatment clinic
• A second front line chemotherapy nurse
• Oncology pharmacist
• Pharmacy technician
• Oncologist
• Clinic nurse
• Clerk
• Risk manager
• Cancer centre chief nursing officer


Team Meeting # 1:


Attendees: work and advisory teams


Purpose: meet and greet; review the process scope


Date: May 2, 2007


Time: 12:00-14:00


Meeting Notes:


-Roundtable introductions




227


-Decided on responsibilities including: team leader (biomedical technology professional),
scribe (front line chemotherapy nurse), and facilitator (human factors expert)


-Explained the difference between the work and advisory groups and set expectations for
frequency of meetings for advisory group (about 7 meetings of varied length over
the course of the analysis)


-Gave overview of planned process scope, start/end points, and inclusion and exclusion
criteria


-Had a group discussion about whether the scope should be expanded to include pharmacy;
decided to keep it the same for now, but to revisit this at next meeting once the
work team has conducted observations and created a draft of the process flow
diagram


Section 13.2.3. Document the Process
An initial process flow diagram was created based on an understanding of the tasks


that would be required to administer chemotherapy to an adult patient using an
elastomeric ambulatory infusion pump (Figure 48).




Figure 48. Initial process flow diagram for administering chemotherapy to an adult patient
with an elastomeric ambulatory infusion pump


After creating the initial process flow diagram, several questions and areas of
uncertainty remained. Work team members raised questions about how to control the rate
of medication delivery, what type of tubing to use, and whether there were any special




228


considerations for nursing for the elastomeric pump. Since these devices were not
currently being used at the healthcare organization, the work team contacted the vendor to
get more information about the devices. The vendor agreed to provide samples of the
elastomeric pump to the healthcare organization so they could better determine how they
might fit with the in-house workflow.


The work team also decided to contact another local organization that was using the
elastomeric pumps to see if they could come and observe staff, to see how the pumps fit
into their workflow. The work team planned a visit to this institution and learned through
observations and interviews of nurses that the stated flow rate of the pump seemed to
depend on a number of physical factors including a patient’s temperature, the head height
between the pump and the infusion port, and the patient’s catheter size. They also learned
that several models of elastomeric ambulatory infusion pump had to be purchased and
stocked because each model of pump delivered medication to the patient at a different flow
rate.


Based on this information, the work team decided to schedule a second visit to the
local institution to learn how the pumps were stored, and how the pharmacy made sure the
proper elastomeric pump model was chosen for a particular patient’s chemotherapy.


During this second visit, the work team learned through observations and
interviews in the pharmacy that pump storage and selection was sometimes challenging
because there were many models that looked similar, with the only differences being a
small printed label on the device, and differently coloured plastic top. One pharmacist also
mentioned that the flow rate stated on the side of the pump could be affected depending on
the diluent that was used by the pharmacy technician when mixing a patient’s
chemotherapy.


The work team updated the initial process flow diagram based on the information
learned through observations and interviews (Figure 49). Through the visits to the field, it
also became apparent to the team that it would be important to expand the scope of the
analyzed process to include: (1) tasks in pharmacy related to picking the right pump, and
(2) mixing with the correct diluent to ensure chemotherapy is delivered to the patient at
the intended rate.




229




Figure 49. Updated process flow diagram based on information learned through
observations and interviews in the field




230


Team Meeting # 2:


Attendees: work and advisory teams


Purpose: review process flow diagram


Date: May 25, 2007


Time: 8:00–12:00


Meeting Notes:


-Asked team members to help themselves to coffee and snacks


-Verified that all team members received a copy of the updated process flow diagram for
review


-Updated advisory team on what the work team has been doing since the last meeting.
Work team created an initial process flow diagram, contacted the manufacturer of the
elastomeric ambulatory infusion pump to get more information and samples, and
contacted another local institution using elastomeric pumps. Work team conducted two
field visits where observations and interviews were conducted. Learned that several
factors affect the flow rate of these pumps and that many models of pumps need to be
purchased and stored as each model delivers medication at a different rate.


-Reviewed updated process flow diagram with the advisory team


-Discussed and came to consensus that process scope should be expanded to include
pharmacy based on observations and interviews


-Reviewed membership of work and advisory team to ensure pharmacy expertise was
accounted for; since pharmacist and pharmacy technician are already part of team, agreed
that no new members are required at this point


-Feedback acquired from the advisory team about updated process flow diagram steps;
several minor modifications were agreed upon based on subject matter expert input


-Next steps: work team will update process flow diagram based on feedback from today’s
meeting and will recirculate within two weeks for independent review and approval by
advisory team members.


Section 13.2.4. Identify Failure Modes and Effects
The work team converted the final process flow diagram approved by the advisory


team into a spreadsheet format, a portion of which is shown in Table 32.




231


Table 32. Part of the spreadsheet created based on the approved process flow diagram


The work team then met to systematically identify failure modes and effects for each
task step and sub-step as shown in Table 33.


Table 33. Portion of the spreadsheet showing failure modes and effects based on the process
flow diagram





232


After coming up with a preliminary list of failure modes and effects, the work team
developed rating scales for review at the next advisory team meeting. The severity and
probability rating scales are shown in Table 34 and Table 35, respectively.


Table 34. Severity rating scale developed by work team for this HFFMEA




Table 35. Probability rating scale developed by work team for this HFFMEA





233


Team Meeting # 3:


Attendees: work and advisory teams


Purpose: review and expand upon potential failure modes and effects


Date: June 15, 2007


Time: 8:00-16:00


Meeting Notes:


-Reviewed agenda including scheduled breaks and lunch.


-Confirmed team members received a copy of the spreadsheet containing failure modes
and effects for the approved process flow diagram.


-Facilitator led the group through the process steps and associated failure modes and
effects row-by-row and asked for input from the group. The facilitator reminded the group
that with this type of analysis, even if a failure mode seems unlikely or has not happened
previously, it could happen and should be included on the spreadsheet.


-Group discussions about many of the listed failure modes, with several new failure modes
being added by the team. The scribe documented this discussion in real-time so the team
could see edits that will be made to the spreadsheet. Team members brought forth several
causes, but this was not the focus of this meeting. So the scribe captured these in a separate
file for later review.


-Some changes to the approved process flow diagram were suggested; these will be made
by the work team following the meeting.


-Once the failure modes and effects were reviewed, the facilitator share the severity and
probability scoring matrices with the group for discussion. The group agreed the scoring
matrices did not need further modifications.


-Next steps: work team to update the process flow diagram and failure modes and effects
spreadsheets and send both documents to the advisory team for review.


Section 13.2.5. Rate Failure Mode Effects and Determine Key Failure Modes
Using the severity and probability scoring matrices agreed upon by the advisory


team, the work team rated each failure mode and effect (Table 36). Whenever there were
disagreements about how an item should be scored, they were discussed until a consensus
was reached.




234




Table 36. Portion of the spreadsheet showing scores assigned for severity and probability





235


Team Meeting # 4:


Attendees: work and advisory teams


Purpose: reach consensus about severity and probability ratings for failure modes and
effects


Date: June 26, 2007


Time: 14:00-17:00


Meeting Notes:


-Asked team members to help themselves to coffee and muffins


-Confirmed team members received a copy of the HFFMEA spreadsheet containing severity
and probability scores and copies of the severity and probability scoring matrices


-Facilitator reminded the team to think about human limitations including cognitive bias,
and limitations in memory and attention when scoring severity and probability of each
failure mode. The facilitator made the point that staff have the best of intentions when they
come to work, but they can’t be expected to be superhuman.


-Facilitator worked through each failure mode, starting by sharing the work teams’
scoring assignments and then invited discussion from the advisory team.


-Advisory team members agreed with many of the pre-assigned scores, however, some
changes were requested and discussed by the group, with the scribe editing scores in real
time.


-Advisory team reviewed the scores for failure modes and effects and discussed cut-off
thresholds for severity and hazard scores; decided on a severity threshold of 3 or higher
and a hazard score threshold of 8 or higher.


Once the advisory team had agreed upon severity and probability scores, the work
team met again and applied the Three Tests (Section 9.5.6.1) to determine whether each
failure mode was a key failure mode (Table 37).




236


Table 37. Portion of the spreadsheet showing determination of key failure modes




The work team created a new spreadsheet that included only those failure modes
considered to be key failure modes. These failure modes carried with them a risk that was
higher than the risk threshold that was predefined by the advisory team. These became the
failure modes that required further consideration in the event the healthcare organization
decided to move forward with implementing elastomeric pumps.


Section 13.2.6. Identify Causes
For those failure modes determined to be key failure modes, the work team met to


discuss possible root causes and contributing factors. Causes that were brought up during
past advisory team meetings, and kept track of by the work team member in the scribe role,
were re-examined to determine whether they might be contributing factors to any of the
key failure modes. A selection of key failure modes and possible causes are included in
Table 38.




237


Table 38. Portion of the updated HFFMEA spreadsheet showing possible causes of key failure
modes




As part of the HFFMEA, the work team was careful to think beyond factors like
compliance with established protocols and procedures, and other more human-centric
causes. Instead, the work team focused on system-level causes and contributing factors,
knowing that only when the system factors were addressed would meaningful
improvements to patient and staff safety be achieved.




238


Team Meeting # 5:


Attendees: work and advisory teams


Purpose: finalize root causes for each key failure mode


Date: July 20, 2007


Time: 9:00-12:00


Meeting Notes:


-Asked team members to help themselves to coffee and muffins


-Confirmed team members received a copy of the HFFMEA spreadsheet containing causes


-Facilitator reminded the group not to focus on human-centric causes and a failure to
follow procedures as root causes. Instead, group should be thinking about system level
causes contributing to potential failure modes.


-Facilitator walked the group through each key failure mode and the potential causes that
had been identified by the work team. The group discussed these and other potential
causes for each key failure mode. The scribe captured the discussion in real time so all
advisory team members could follow along.


-A number of system level causes that could contribute to several key failure modes were
identified during the discussion. These types of contributing factors may be good to focus
on when it comes to developing mitigating strategies as fixing even just one of these
contributing factors would have the potential to mitigate several key failure modes.


Next Steps: The work team will meet to update and refine the list of causes based on this
meeting. The updated spreadsheet will be circulated to the advisory team within the next
three weeks for review and feedback.


Section 13.2.7. Develop and Implement Mitigating Strategies
Based on the causes identified by the advisory team, the work team met and


brainstormed a number of possible mitigating strategies to address system issues at the
root of each key failure mode. The work team referred to the Hierarchy of Effectiveness
(Chapter 3) while developing potential mitigating strategies to ensure solutions addressed
system-level, rather than person-level factors.


If the healthcare organization decided to implement elastomeric pumps, in parallel
they would also want to consider implementing a number of the identified mitigating
strategies to proactively prevent any potential errors as identified through the analysis,
from occurring.




239


Team Meeting # 6:


Attendees: work and advisory teams


Purpose: develop mitigating strategies to address root causes for each key failure mode


Date: August 17, 2007


Time: 13:00-16:00


Meeting Notes:


-Asked team members to help themselves to coffee and muffins


-Confirmed team members received a copy of the HFFMEA spreadsheet containing finalized
causes, and preliminary ideas for mitigating strategies.


-Facilitator circulated copies of the Hierarchy of Effectiveness (Chapter 11) to each
advisory team member and described the model to help ensure recommendations
generated were more systems focused rather than person focused.


-Ideas for how to mitigate root causes for each failure mode were discussed by the team
and the Hierarchy of Effectiveness was referred to throughout the discussion.


-Advisory team discussed possible criteria that could be used to highlight those mitigating
strategies likely to be the most feasible. Considered several different aspects such as (1)
how effective (Hierarchy of Effectiveness), (2) required resources, (3) available resources.
The team agreed that it would be preferred to implement fewer high-impact mitigating
strategies, than many lower-impact mitigating strategies.


-Advisory team looked for and identified possible areas of overlap where implementing a
single recommendation would address more than one cause.


-Scribe recorded discussion in real time so team members could see and follow along.


-Next Steps: work team to circulate cleaned version of HFFMEA spreadsheet containing
ideas about mitigating strategies


Based on the preliminary ideas for mitigating strategies, and discussion during
Team Meeting #6, the work team updated the HFFMEA spreadsheet (Table 39), and
circulated it to the advisory team for review and feedback.




240


Table 39. Part of the updated HFFMEA spreadsheet showing ideas for possible mitigating
strategies







241


Team Meeting # 7:


Attendees: work and advisory teams


Purpose: prioritize mitigating strategies, create implementation plans, conclude HFFMEA


Date: August 31, 2007


Time: 8:00-12:00


Meeting Notes:


-Asked team members to help themselves to coffee and muffins


-Confirmed team members received a copy of the updated HFFMEA spreadsheet containing
mitigating strategies


-Facilitator reminded the advisory team to refer to the Hierarchy of Effectiveness during
the discussion, and reviewed the criteria discussed during the previous meeting for
highlighting those mitigating strategies likely to be the most feasible.


-Facilitator presented the work groups’ thoughts about which mitigating strategies would
be most feasible and have the highest impact based on criteria chosen.


-Discussion took place among advisory group members about the pros and cons of
eventually trying to implement the proposed mitigating strategies, and the discussion was
opened up to consider whether other strategies should also be considered in more detail.


-The advisory team prioritized mitigating strategies considering the criteria chosen to
determine (1) how effective, (2) the resources likely to be required, and (3) the resources
likely to be available.


-Based on the long list of possible mitigating strategies, the advisory team chose the top 10
priority strategies and developed an implementation plan for each (Section 9.6, What to Do
With a Completed HFFMEA). In the event the healthcare organization decides to move to
elastomeric pumps, each implementation plan will be assigned to a staff member who will
see the plan through so the associated key failure modes can be mitigated before causing
harm.


-The team leader concluded the meeting by thanking everyone for their participation as
part of the HFFMEA advisory team. Although implementation work would continue if the
organization decided to go ahead with elastomeric pumps, this was the last official meeting
of the HFFMEA team.


Following the final HFFMEA meeting, the work group met to create a summary
document outlining (1) the HFFMEA process followed, (2) team members on both the work
and advisory groups, (3) key decisions made, (4) lessons learned, (5) implementation




242


strategies developed, and (6) an appendix containing the key failure modes, their causes
and effects. Before providing the report to upper management, it was shared with the
advisory team for feedback.


The report was shared with management at the healthcare organization in a timely
manner so that information about key failure modes and potential means of mitigating
risks associated with implementing elastomeric pumps could be integrated with the
healthcare organizations’ decision-making process. This resource provided insight to
potential risks associated with implementing elastomeric pumps, which could then be
compared with the inherent risks associated with keeping the electronic ambulatory
infusion pumps uncovered as part of the HFRCA. In this way, management was able to make
a more informed decision by weighing the residual risk associated with keeping the
existing electronic pumps versus implementing the new elastomeric pumps.










243


Human Factors Resources
HumanEra


This book is based on the collective experience of the team members of HumanEra.
HumanEra is a healthcare human factors research team based out of the Centre for Global
eHealth Innovation in Toronto, Canada with over a decade of experience conducting
applied research and implementation projects to improve healthcare system safety. To
contact or learn more about HumanEra visit our website at www.HumanEra.ca.


Human Factors Books
There is no shortage of books and texts on the subject of human factors, its methods,


and specific applications across individual domains, but the following two books provide
an excellent primer on the topic and are filled with relevant examples.


1. Kim J. Vicente. The human factor: Revolutionizing the way we live with
technology: Vintage Canada; 2004.


2. Steven M. Casey. Set phasers on stun: And other true tales of design,
technology, and human error: Aegean Publishing Co.; 1998.


Human Factors Organizations/Events


• The Human Factors and Ergonomics Society (HFES); www.hfes.org.


This American organization hosts an annual conference (produces published
proceedings) and publishes the journals Human Factors, Ergonomics in Design
and the Journal of Cognitive Engineering and Decision Making. HFES has a
Healthcare Technical Group (http://hctg.wordpress.com) and organizes an
annual Symposium on Human Factors and Ergonomics in Healthcare. HFES also
has a European Chapter.


• SIGCHI; www.sigchi.org.


This international organization hosts an annual conference (produces published
proceedings and publishes the journal TOCHI (ACM Transactions on Computer-
Human Interaction).


• The Institute of Ergonomics and Human Factors; http://iehf.org.


This UK-based organization hosts an annual conference, accredits professionals,
and has a Healthcare special interest group.


• The International Conference on Applied Human Factors and Ergonomics
and the Affiliated Conferences; www.ahfe2014.org.




244


Held concurrently as part of this event is an International Conference on Human
Factors and Ergonomics in Healthcare. AHFE publishes post-conference edited
books with accepted and peer reviewed papers.


Healthcare Human Factors Guidance Documents
1. The FDA has developed a draft guidance document to assist industry in
conducting appropriate human factors testing and identifying device features
that manufacturers should optimize throughout the total product life cycle.
Available at:


http://www.fda.gov/medicaldevices/deviceregulationandguidance/guidancedo
cuments/ucm259748.htm


2. The World Health Organization has produced a document that reviews ten
topic areas related to organizational and human factors influencing patient
safety. Available at:


http://www.who.int/patientsafety/research/methods_measures/human_factor
s/human_factors_review.pdf?ua=1


3. The clinical human factors group (www.chfg.org) has influenced the first
volume and produced the second volume of a document titled Implementing
Human Factors in healthcare.


Volume 1 (published by Patient Safety First) available at:


http://www.patientsafetyfirst.nhs.uk/ashx/Asset.ashx?path=/Intervention-
support/Human%20Factors%20How-to%20Guide%20v1.2.pdf


Volume 2 available at:


Http://www.chfg.org/wp-content/uploads/2013/05/Implementing-human-
factors-in-healthcare-How-to-guide-volume-2-FINAL-2013_05_16.pdf





245


References
1. To Err Is Human: Building a Safer Health System: The National Academies Press;


2000.
2. Baker GR, Norton PG, Flintoft V, Blais R, Brown A, Cox J, Etchells E, Ghali WA, Hébert


P, Majumdar SR et al: The Canadian Adverse Events Study: the incidence of
adverse events among hospital patients in Canada. Canadian Medical Association
Journal 2004, 170(11):1678-1686.


3. Sari ABA, Sheldon TA, Cracknell A, Turnbull A, Dobson Y, Grant C, Gray W,
Richardson A: Extent, nature and consequences of adverse events: results of a
retrospective casenote review in a large NHS hospital. Quality & Safety in Health
Care 2007, 16(6):434-439.


4. Davis P, Lay-Yee R, Briant R, Ali W, Scott A, Schug S: Adverse events in New
Zealand public hospitals I: occurrence and impact. The New Zealand medical
journal 2002, 115(1167):U271.


5. Wilson RM, Harrison BT, Gibberd RW, Hamilton JD: An analysis of the causes of
adverse events from the Quality in Australian Health Care Study. The Medical
Journal of Australia 1999, 170(9):411-415.


6. James JT: A New, Evidence-based Estimate of Patient Harms Associated with
Hospital Care. Journal of Patient Safety 2013, 9(3):122-128.


7. FDA: Draft Guidance for Industry and Food and Drug Administration Staff -
Applying Human Factors and Usability Engineering to Optimize Medical
Device Design. In.; 2011.


8. American College of Clinical Engineering (ACCE)
[http://www.accenet.org/default.asp?page=about&section=definition]


9. Vicente K: The human factor: Revolutionizing the way we live with technology:
Vintage Canada; 2004.


10. Norman DA: The Design of Everyday Things: Basic Books; 2002.
11. United States. Army Air Forces. Headquarters AMCED: Analysis of Factors


Contributing to 460 "pilot-error" Experiences in Operating Aircraft Controls:
Army Air Forces Headquarters, Air Material Command, Engineering Division; 1947.


12. Wiener EL, Nagel DC: Human factors in aviation: Access Online via Elsevier; 1988.
13. Misumi J, Miller R, Wilpert B: Nuclear safety: A human factors perspective: CRC


Press; 1998.
14. Stanton NA: Human factors in nuclear safety: Taylor & Francis; 1996.
15. Carayon P: Human factors in patient safety as an innovation. Applied ergonomics


2010, 41(5):657-665.
16. Cooper JB, Newbower RS, Long CD, McPeek B: Preventable Anesthesia Mishaps: A


Study of Human Factors. Anesthesiology 1978, 49(6):399-406.
17. Leape LL: Error in medicine. JAMA-Journal of the American Medical Association-US


Edition 1994, 272(23):1851-1856.
18. Leape LL, Brennan TA, Laird N, Lawthers AG, Localio AR, Barnes BA, Hebert L,


Newhouse JP, Weiler PC, Hiatt H: The nature of adverse events in hospitalized
patients: results of the Harvard Medical Practice Study II. New England Journal
of Medicine 1991, 324(6):377-384.




246


19. Hollands JG, Wickens CD: Engineering psychology and human performance:
Prentice Hall New Jersey; 1999.


20. Weissman JS, Rothschild JM, Bendavid E, Sprivulis P, Cook EF, Evans RS, Kaganova Y,
Bender M, David-Kasdan J, Haug P et al: Hospital workload and adverse events.
Med Care 2007, 45(5):448-455.


21. Harrison MI, Koppel R, Bar-Lev S: Unintended consequences of information
technologies in health care—an interactive sociotechnical analysis. Journal of
the American Medical Informatics Association 2007, 14(5):542-549.


22. Jennings BM: Work Stress and Burnout Among Nurses: Role of the Work
Environment and Working Conditions. 2008.


23. Flewwelling CJ: No Fault Found Reporting and its Relation to Human Factors
Related Design Faults of Medical Devices. 2012.


24. Smith EE, Kosslyn SM: Cognitive Psychology: Mind and Brain: Pearson/Prentice
Hall; 2007.


25. Mack A, Rock I: Inattentional Blindness: MIT Press; 1998.
26. Alert IMS: Medication error prevention “tool box”. Retrieved June 1999, 30:2004.
27. McCarney R, Warner J, Iliffe S, van Haselen R, Griffin M, Fisher P: The Hawthorne


Effect: a randomised, controlled trial. BMC Med Res Methodol 2007, 7:30.
28. Bailey RW: Performance vs. preference. In: Proceedings of the Human Factors and


Ergonomics Society Annual Meeting: 1993. SAGE Publications: 282-286.
29. Crabtree BF, Miller WL: Doing Qualitative Research: SAGE Publications; 1999.
30. Glaser B: The Constant Comparative Method of Qualitative Analysis. Social


Problems 1965, 12(4):436-445.
31. Kirwan BE, Ainsworth LK: A guide to task analysis. 1992.
32. Nielsen J: Usability inspection methods. In: Conference Companion on Human


Factors in Computing Systems; Boston, Massachusetts, USA. ACM 1994.
33. Shneiderman SB, Plaisant C: Designing the user interface 4 th edition. In.:


Pearson Addison Wesley, USA; 2005.
34. Zhang J, Johnson TR, Patel VL, Paige DL, Kubose T: Using usability heuristics to


evaluate patient safety of medical devices. Journal of Biomedical Informatics
2003, 36:23-30.


35. Nielsen J, Molich R: Heuristic evaluation of user interfaces. In: Proceedings ACM
CHI'90 Conf: 1990; Seattle, WA, 1-5 April. ACM: 249-256.


36. Food, Administration D: Draft Guidance for Industry and Food and Drug
Administration Staff. Applying Human Factors and Usability Engineering to
Optimize Medical Device Design. US Food and Drug Administration, Silver Spring,
MD 2011.


37. DeRosier J, Stalhandske E, Bagian JP, Nudell T: Using health care Failure Mode
and Effect Analysis: the VA National Center for Patient Safety's prospective
risk analysis system. Jt Comm J Qual Improv 2002, 28(5):248-267, 209.


38. Failure Mode And Effects Analysis (FMEA): A Tool to Help Guide Error
Prevention Efforts [https://www.ismp.org/tools/FMEA.asp]


39. ISMP Canada Failure Mode and Effects Analysis [http://www.ismp-
canada.org/fmea.htm]


40. Institute for Healthcare Improvement: Failure Mode and Effects Analysis
(FMEA) Tool [http://app.ihi.org/Workspace/tools/fmea/]




247


41. 2014 Hospital Accreditation Standards. Oakbrook Terrace, Illinois: Joint
Commission; 2013.


42. Required Organizational Practices 2012. In., vol. 2013. Ottawa: Accreditation
Canada; 2012.


43. Safety VNCfP: The basics of Healthcare Failure Mode and Effect Analysis. In., vol.
2013: Veterans Affairs National Center for Patient Safety; 2001.


44. Commission J: Sentinel Events. Comprehensive Accreditation Manual for Office-
Based Surgery Practices 2011:SE-1 to SE-18.


45. Smetzer J, Baker C, Byrne FD, Cohen MR: Shaping systems for better behavioral
choices: lessons learned from a fatal medication error. Joint Commission Journal
on Quality and Patient Safety 2010, 36(4):152.


46. Cohen MR: An Injustice Has Been Done: Jaile Time for an Error. Patient Safety
and Quality Healthcare 2009(September/October 2009):6-7.


47. Wu AW: Medical error: the second victim. BMJ 2000, 320(7237):726-727.
48. Anderson P, Townsend T: Medication errors: Don’t let them happen to you.


American Nurse Today 2010, 5(3):23-27.
49. Clancy CM: Alleviating "second victim" syndrome: how we should handle


patient harm. J Nurs Care Qual 2012, 27(1):1-5.
50. Framework for Conducting a Root Cause Analysis and Action Plan | Joint


Commission
[http://www.jointcommission.org/Framework_for_Conducting_a_Root_Cause_Analy
sis_and_Action_Plan/]


51. Root Cause Analysis - VA National Center for Patient Safety
[http://www.patientsafety.va.gov/media/rca.asp]


52. ISMP Canada - Root Cause Analysis [https://www.ismp-canada.org/rca.htm]
53. Canadian Incident Analysis Framework


[http://www.patientsafetyinstitute.ca/English/toolsResources/IncidentAnalysis/D
ocuments/Canadian%20Incident%20Analysis%20Framework.PDF]


54. Patient Safety - Root Cause Analysis toolkit | conditions of use
[http://www.nrls.npsa.nhs.uk/resources/rca-conditions/]


55. Meadows S, Baker K, Butler J: The Incident Decision Tree: Guidelines for Action
Following Patient Safety Incidents, vol. Advances in Patient Safety: From Research
to Implementation (Volume 4: Programs, Tools, and Products). Rockville MD; 2005.


56. Miller R, Griffith S, Vogelsmeier A: A statewide approach to a just culture for
patient safety: The Missouri story. Journal of Nursing Regulation 2010, 1(1):52-
57.


57. Marx DA: Patient safety and the" just culture": a primer for health care
executives: Trustees of Columbia University; 2001.


58. Doran GT: There’sa SMART way to write management’s goals and objectives.
Management review 1981, 70(11):35-36.


59. ECRI: Burns and Fires from Electrosurgical Active Electrodes. Health Devices
1993, 22(8-9):2.


60. ECRI: ESU Burns from Poor Return Electrode Site Preparation. Health Devices
1987, 16(1):2.




248


61. Donato M: Family Still Coming To Grips With Electrocution Of Baby. In: Chicago
Tribune. 1993.


62. Newborn Is Electrocuted. In: New York Times. 1993.
63. Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH: Types of unintended


consequences related to computerized provider order entry. J Am Med Inform
Assoc 2006, 13(5):547-556.


64. Saathoff A: Human factors considerations relevant to CPOE implementations. J
Healthc Inf Manag 2005, 19(3):71-78.


65. Connolly C: Cedars-Sinai Doctors Cling to Pen and Paper. In: The Washington
Post. 2005.


66. FDA: Preventing Tubing and Luer Misconnections > Examples of Tubing and
Luer Misconnections. In. Edited by Health C-CfDaR. Rockville; 2015.


67. FDA: Guidance for Industry and Food and Drug Administration Staff - Total
Product Life Cycle: Infusion Pump - Premarket Notification [510(k)]
Submissions - Draft Guidance. In. Edited by Health C-CfDaR. Rockville; 2010.


68. Johnson TR, Zhang J, Patel VL, Keselman A, Tang X, Brixey JJ, Paige D, Turley JP: The
role of patient safety in the device purchasing process. In.: DTIC Document;
2005.


69. Cassano AL: Applying human factors to the procurement of electrosurgical
medical devices: a case study. In: Proceedings of the Human Factors and
Ergonomics Society Annual Meeting: 2003. SAGE Publications: 1815-1819.


70. Cassano-Piché A, Cafazzo JA, Chagpar A, Easty AC: Choosing Safer Medical
Technologies: How Human Factors Methods Can Help in the Procurement
Process. Biomedical Instrumentation & Technology 2010, 44(s1):49-56.


71. Ginsburg G: Human factors engineering: A tool for medical device evaluation in
hospital procurement decision-making. Journal of Biomedical Informatics 2005,
38(3):213-219.


72. Namshirin P, Ibey A, Lamsdale A: Applying a multidisciplinary approach to the
selection, evaluation, and acquisition of smart infusion pumps. Journal of
Medical and Biological Engineering 2011, 31(2):93-98.


73. Instrumentation AftAoM: ANSI/AAMI HE75: 2009: Human factors engineering:
Design of medical devices. AAMI, Arlington 2009.


74. Israelski E: Overview of International Medical Device Human Factors
Standards. In: Symposium on Human Factors and Ergonomics in Health Care:
Bridging the Gap: 12-17 March 2012 2012. 2015.


75. Holden RJ, Karsh B-T: The Technology Acceptance Model: Its past and its future
in health care. Journal of Biomedical Informatics 2010, 43(1):159-172.


76. Lapointe L, Rivard S: A multilevel model of resistance to information technology
implementation. Mis Quarterly 2005:461-491.


77. Markus ML: Power, politics, and MIS implementation. Communications of the ACM
1983, 26(6):430-444.


78. Zuboff S: In the age of the smart machine: the future of work and power:
Heinemann Professional; 1988.


79. Lorenzi NM, Riley RT: Managing change: an overview. J Am Med Inform Assoc
2000, 7(2):116-124.




249


80. Smith MJ, Carayon P: New technology, automation, and work organization:
Stress problems and improved technology implementation strategies.
International Journal of Human Factors in Manufacturing 1995, 5(1):99-116.


81. Lærum H, Ellingsen G, Faxvaag A: Doctors' use of electronic medical records
systems in hospitals: cross sectional survey. BMJ 2001, 323(7325):1344-1348.


82. Davis FD: Perceived usefulness, perceived ease of use, and user acceptance of
information technology. Mis Quarterly 1989:319-340.


83. Davis FD, Bagozzi RP, Warshaw PR: User acceptance of computer technology: a
comparison of two theoretical models. Management Science 1989, 35(8):982-
1003.


84. Vicente KJ, Kada-Bekhaled K, Hillel G, Cassano A, Orser BA: Programming errors
contribute to death from patient-controlled analgesia: case report and
estimate of probability. Canadian Journal of Anesthesia 2003, 50(4):328-332.


85. Weinger MB, Pantiskas C, Wiklund ME, Carstensen P: Incorporating human
factors into the design of medical devices. Jama 1998, 280(17):1484.


86. Abbott PCA Plus II patient-controlled analgesic pumps prone to
misprogramming resulting in narcotic overinfusions. Health Devices 1997, 26(9-
10):389-391.


87. Evidence builds: lack of focus on human factors allows error-prone devices. In:
ISMP Medication Safety Alert. vol. 2013. Horsham, PA: Institute for Safe Medication
Practices; 1999.


88. Lin L, Isla R, Doniz K, Harkness H, Vicente K, Doyle DJ: Applying Human Factors to
the Design of Medical Equipment: Patient-Controlled Analgesia. Journal of
Clinical Monitoring and Computing 1998, 14(4):253-263.


89. Fairbanks Rollin J, Caplan S: Poor Interface Design and Lack of Usability Testing
Facilitate Medical Error. Joint Commission Journal on Quality and Patient Safety
2004, 30(10):579-584.


90. Gagnon R, Laberge J, Lamsdale A, Histon J, Hudson C, Davies J, Caird J: A User-
Centered Evaluation of three Intravenous Infusion Pumps. Proceedings of the
Human Factors and Ergonomics Society Annual Meeting 2004, 48(15):1773-1777.


91. Kaye R, Crowley J: Medical Device Use-Safety: Incorporating Human Factors
Engineering into Risk Management. 2000.


92. Elliott M, Liu Y: The nine rights of medication administration: an overview.
British journal of nursing (Mark Allen Publishing) 2010, 19(5):300-305.


93. Ontario CoNo: Practice Standard: Medication. In.; 2015.
94. Fluorouracil Incident Root Cause Analysis Report. In., vol. 2014. Toronto:


Institute for Safe Medication Practices Canada; 2007.
95. Furniss D, Randell R, O’Kane AA, Taneva S, Mentis H, Blandford A: Fieldwork for


healthcare: guidance for investigating human factors in computing systems.
Synthesis Lectures on Assistive, Rehabilitative, and Health-Preserving Technologies
2014, 3(2):1-146.





248


Appendix A: Confidentiality and Anonymity
Most of the human factors methods presented in this book require the time and


participation of the end-users of the technology. It is important that information gathered
using these methods be treated as confidential and anonymous, to protect the reputation
and credibility of participants, and respondents are usually more honest under these
conditions. Those responsible for carrying out human factors methods should get
agreement from clinical managers, and all members of the team or committee involved in
reviewing the human factors data, that all data gathered will be kept confidential and will
not be used in any way to evaluate clinical competence or expose them to professional risk.
No disciplinary action should ever result from participation in human factors testing.


Best practice is to not collect or record any unnecessary participant information that
is identifying (e.g., names of participants). Participant numbers can be assigned for
comparison and reference purposes. Additionally, to help communicate your commitment
to confidentiality and/or anonymity, it is important to get informed consent. This purpose
of informed consent is to ensure that participants understand [95]:


• the aims and methods of the study/project;


• that their participation is voluntary, and they can withdraw at any time
without any consequences (and how their data will be handled);


• any risks and benefits of their participation;


• that their data will be anonymized and kept confidential;


• how the results of the study/project will be used and shared (e.g., to
make a procurement decision);


A sample consent form is provided at the end of this section for you to use as a
template.


The process of obtaining informed consent usually involves ensuring the participant
understands the points listed above by reading through a consent document that explains
each of the points and allowing the participant as much time as they need to review the
document and ask questions before deciding whether or not to participate. If they choose
to participate, they must sign the consent document.


If you are conducting an internal project that involves participants and it is not part
of a research study and there are no plans to disseminate the findings outside of the
organization, it is likely sufficient to use a consent form that covers the points described
above without requiring research ethics approval. However, it is recommended that you




249


investigate whether ethics approval is required for the work you are undertaking prior to
collecting any data to ensure the data can be used for its intended purpose, and without
any restrictions, as a result of not obtaining research ethics approval. Generally speaking,
research ethics approval is required if:


• the data being collected is part of a research study.


• there is a possibility that the data will be used for research at a later date.


However, even quality improvement initiatives can present ethical risks and should
be managed by a formal research ethics process. To help determine whether this is indeed
the case, you can use the online ARECCI Ethics Screening Tool found at
http://www.aihealthsolutions.ca/arecci/screening/30863/6d62b234cf1570caeb290708c
af72dd3, or inquire directly to your organization’s research ethics committee.


A detailed discussion of the research ethics approval process in healthcare as it
relates specifically to conducting human factors studies is covered in the book Fieldwork
for Healthcare: Guidance for investigation human factors in computing systems[95]. A free
chapter of the book, containing the sections related to research ethics approval and
informed consent is available at:


http://www.morganclaypool.com/doi/suppl/10.2200/S00606ED1V02Y201410ARH007/s
uppl_file/Furniss_Ch1.pdf.





Sample Consent Form


CONSENT TO PARTICIPATE IN A USABILITY STUDY


Introduction


You are being asked to take part in a usability study. Please read this explanation about the study and its
risks and benefits before you decide if you would like to take part. You should take as much time as you
need to make your decision. You should ask the study staff to explain anything that you do not
understand and make sure that all of your questions have been answered before signing this consent
form. Before you make your decision, feel free to talk about this study with anyone you wish. Participation
in this study is voluntary.


Purpose


The purpose of this project is to [insert purpose here]. Your participation helps us to determine [insert
benefit here such as “identify which product is the safest for your unit”].


Procedures


If you agree to participate in the study your demographic information (e.g., age, sex, years nursing
experience) will be collected and you will be asked to complete a series of clinical tasks in a simulated
clinical environment. In other words, you will be in a room with clinical equipment and scenarios but no
real patients or patient care. You will be taught how to use the devices not in routine use on your unit prior
to starting the simulations. After training you will be oriented to the simulated environment, and asked to
perform various tasks to a simulated patient (mannequin and/or actor). After each scenario, we will ask
you for your feedback based on our observations to further understand the risks and benefits of the
devices being tested. The session will last no more than 3 hours, and will be videotaped for later analysis.
Your performance/competency is NOT being evaluated in a way that will impact your employment, but
rather the results of this study will be used to better understand issues relating to the devices we are
evaluating.


Risks


There are no anticipated or known medical risks associated with this study. You may experience
discomfort in sharing your opinions with the researchers. You only have to share as much about your
opinions as you wish. Your participation will have NO impact on your employment.


Benefits


You may or may not receive direct benefit from participating in this study. Information from this study may
help to increase your knowledge about [insert the type of device here].


Voluntary Participation


Your participation in this study is voluntary. You can choose not to participate or you may withdraw at
any time. Whether you choose to participate or not has no impact on your employment. In no way does
signing this consent form waive your legal rights nor does it relieve the investigators, sponsors or involved




institutions from their legal and professional responsibilities. You do not give up any of your legal rights by
signing this consent form.


Confidentiality


All information obtained during the study will be held in strict confidence. You will be identified with a
subject number only. No names or identifying information will be used in any reports, publication or
presentations that may come from this study. No information identifying you will be transferred outside
the investigators of this study. If the videos from the research are shown outside the research team, your
face will be blurred and all identifying information will be made anonymous. However, despite best
efforts, there is a very small possibility that you may still be identified. Data from the study (e.g.,
videotapes, paper records) will be kept for a minimum of two years, and a maximum of seven years, after
the completion of the study. Any personal identifiable information will be stored and protected on secured
servers or kept in a locked filing cabinet and then destroyed by shredding of paper or erasing of digital
information.


Reimbursement


You will not receive any financial reimbursement for your participation.


Questions


If you have any questions, concerns or would like to speak to the study team for any reason, please
contact [insert contact name and information of person responsible for the study.


Consent


This study has been explained to me and any questions I had have been answered. I know that I may
leave the study at any time. I agree to take part in this study.






Study participant’s name (please print) Participant’s Signature Date
(You will be given a signed copy of this consent form)


My signature means that I have explained the study to the participant named above. I have answered all
questions.






Name of person obtaining consent Signature Date







252


About the Authors


Andrea Cassano-Piché, MASc, PEng
Andrea is a human factors engineer. She received her degrees in Industrial
Engineering at the University of Toronto. Since 2002, she has been applying
human factors methods to the design and evaluation of health technologies
and conducting proactive and retrospective investigations of patient safety
issues. Andrea’s research interests include proactive risks management of
complex healthcare systems and building human factors capacity within
healthcare as a means of transforming the culture of healthcare safety.


cassanopiche@gmail.com


https://ca.linkedin.com/in/cassanopiche




Patricia Trbovich, PhD
Patricia is the Research Lead for the HumanEra Team at the University
Health Network. She is Assistant Professor in the Institute of Biomaterials
and Biomedical Engineering, and of Health Informatics in the Department of
Health, Policy, Management and Evaluation, Faculty of Medicine, University
of Toronto. Her areas of expertise include human factors engineering and
patient safety.


patricia.trbovich@uhn.ca


https://ca.linkedin.com/in/patriciatrbovich



Melissa Griffin, MHSc, PEng
Melissa is a clinical engineer and human factors specialist. She received her
Master of Health Science degree in Clinical Engineering from the University
of Toronto. Her work has focused on applying human factors methods to a
wide range of healthcare environments, including home care and oncology.
She is passionate about improving patient safety across our health system
and teaching others to apply this way of thinking in their daily work.


Melissa.griffin@uhn.ca


https://ca.linkedin.com/in/melissagriffin4











253


Ying Ling Yin, MEng, PhD candidate.

Ying Ling received her Bachelors and Masters Degrees in engineering from
McGill University. She continues her graduate studies at the University of
Toronto where she researches human factors in complex, high-technology
medical environments. Her interests include patient safety in limited
resource settings and human factors methods as a means to design safer and
appropriate healthcare systems.


ylin@metalin.ca


https://ca.linkedin.com/in/yinglinglin





Anthony (Tony) Easty, Ph D, PEng, CCE
Tony is a Senior Scientist and the Inaugural Chair-holder of the Baxter Chair
in Health Technology at the University Health Network/University of
Toronto. He is also an associate professor at the Institute of Biomaterials
and Biomedical engineering at the university. Dr. Easty’s research focus is on
the usability and safety of medical technologies throughout the health care
system with a special focus on medication safety and safety in home care. He
has established the HumanEra Team, educated at the PhD and Masters levels
in human factors engineering, cognitive psychology and biomedical
engineering.


tony.easty@utoronto.ca


https://ca.linkedin.com/in/tonyeasty









Copyright © 2017 Engineering World Health. All Rights Reserved.