Giving Voice To Patients Through Interviews and Qualitative Analysis-A Pilot Study
Giving Voice To Patients Through Interviews and Qualitative Analysis-A Pilot Study
Giving Voice To Patients Through Interviews and Qualitative Analysis-A Pilot Study
The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
WBB Inc
• Ebele Okoli
WBB Inc
Abstract
Modern healthcare is drowning in data, and burdened by quality, safety, financial, and operational
metrics, but few relate directly to how patients experiences their care. This has a direct bearing on
patient safety, and whether the care they receive meets their needs and goals. As such, a key concept in
quality management, is to view all processes in terms of whether, and to what degree, these meet
patient goals. However, the literature lacks sufficient specificity on how care processes are seen through
the eyes of the patient. A thick account of patient experience of their care processes could provide us
with a typology of what patients are seeing, how they conceptualize what they experience, and what
To fill a gap in awareness of the patient experience of the radiology processes, we used a mixed
methods qualitative approach to elicit the patient view of their radiology experiences, and attempt to
develop a typology and insights from the patient voice. We developed a typology of patient experiences
of the radiology processes that centered on communication gaps, and reflected opacity, fragility, and
unpredictability of administrative and care processes in radiology. Although care and administrative
processes frequently failed to interconnect efficiently or effectively, and did not work well as an end-to-
end patient journey. Care processes were described by participants as fragile, solitary, and opaque, and
required constant vigilance, supervision, and assistance by patients. Participants described a need for
improved communication between radiology staff and patients that focuses on the patient journey and
Introduction
One of the key recommendations of the National Patient Safety Foundation (NPSF) is to “create a
common set of safety metrics that reflect meaningful outcomes.” [1] This paper attempts to look at
patient outcomes from the broad perspective of how patients might experience the end-to-end journey
through radiology, and generate a typology of potential measures of quality through the lens of process
quality.
From a quality improvement perspective, patients are the “customers” for radiology care, and their
perspective is valuable in improving radiology processes and interactions. “Voice of the customer” is a
concept that prioritizes how a customer would view the services, products, and outcomes of a workflow
or process. [2] In this qualitative study, we used a contextual approach to examine radiology from the
voice of the customer perspective, report on the patient view of their radiology experiences, and
attempt to develop a typology and insights from the patient voice. To provide insight into the patients’
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
perspective on their care experiences, we look at elements of the patient voice from structured
interview responses, and identify opportunities for improving specialty-specific processes and
interactions.
The interviews speak to the failure of radiology care processes in many, and perhaps predictable, ways.
Processes were often opaque, fragile, or extremely siloed, and frequently failed in ways that
undermined patient confidence, caused anxiety or alarm, or resulted in mental or physical harm.
Method
This study was evaluated as IRB exempt, and interviewing, coding, and analysis was carried out by two
researchers, both with mixed-methods experience in monitorting & evaluation and process
deployment, and workflow improvement assessments. The study uses mixed methods, [3] and draws
from postpositivism (an external reality), constructivism (participants form own views of healthcare) ,
phenomenalism (how participants experienced healthcare), and partial use of grounded theory (coding
built mainly from what participants revealed). We did so to establish an asymetrical but “conversational
partnership” in which both the researcher and participant jointly shaped the path of discovery. [4]
We recruited nine participants through Twitter and on the Blue Faery Liver Cancer forum group on the
Health Unlocked website. The Blue Faery Liver Cancer Association is a non-profit organization that
provides information to patients and carers on research, and education, and performs advocacy related
to primary liver cancer, specifically Hepatocellular Carcinoma (HCC). We invited patients and carers who
had previous experience of radiology in a message on Twitter in which we described the aim of the
study, and referred interested parties to a Physician’s Weekly Blog entitled “Micro-Study: Discovering
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
the Patient Voice.” Prospective participants were invited to contact the lead researcher by direct
message on Twitter, email, or through LinkedIn. Ten individuals made contact, and each provided an
email for further contact. The ten who expressed the wish to be interviewed were emailed details
outlining the purpose of the study, topic areas, the level of privacy they should expect, and an offer of a
$25 gift card for participation. Of the ten people who responded, one was lost to contact, and nine
completed the study. Participant identity was not verified, and we did not verify experiences or events
The use of online patients is an important limitation of this study, as it is difficult to verify the true
nature of their radiology experience, and also limits respondents to those patients and carers who have
internet access and are active on social media. Since this is a qualitative study, it seeks to answer the
question "what can happen" not "what is typical," and therefore does not attempt to have a large or
representational respondent panel. Participants were interviewed over the phone, and sessions were
recorded with explicit participant permission. The interviews were transcribed using an online
transcription service with a machine learning system, and was coded in a qualitative data analysis (QDA)
tool. The transcription results were manually compared to the audio track, and were found to have a
Due to the small sample size used for this study and no intention to attempt to generalize the findings,
there was no attempt to stratify or declare participant demographic data such as age, gender, race, or
whether they visited urban, rural, or inner-city radiology facilities. Likewise, although participants
mentioned how they funded their care, we did not use this as a stratification. The participants
Interview questions were framed in the context of the things that surprised, confused, or frustrated the
patients in any of their experiences with radiology. The framework of surprise, confusion, and
frustration was selected as a result of previous focus group sessions held with patients on their care
experiences, and as a result of our experience with hospital processes during quality improvement and
assessment work. Although partially overlapping, the three constructs were found to be sufficient to
capture experiences related to care and administrative processes encountered by patients during their
care journey. Participants were encouraged to relate the questions to any experience, from the very first
moment they were referred for radiology, to getting the results explained to them or receiving a bill.
1. In your experience with radiology, can you recall something that surprised you?
b. What part of the radiological process did it relate to? E.g. getting an x-ray, booking an
c. How did this affect you? i.e. did something bad happen, like missing an appointment,
etc.
2. In your experience with radiology, can you recall something that confused you?
b. What part of the radiological process did it relate to? E.g. getting an x-ray, booking an
c. How did this affect you? i.e. did something bad happen, like missing an appointment,
etc.
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
3. In your experience with radiology, can you recall something that frustrated you?
b. What part of the radiological process did it relate to? E.g. getting an x-ray, booking an
c. How did this affect you? i.e. did something bad happen, like missing an appointment,
etc.
Participants were further asked to reflect on the entire experience, and describe how parts of the
process worked for them – anything from the way appointments were booked or parking was provided,
To gain an initial insight into what our participants were saying, the word cloud (Figure 1) was generated
in the QDA from the text across all participants. In the word cloud, word frequency dictated font size,
and color used merely to provide optical contrast. We used the word cloud as an initial orienting device
prior to coding.
Although we started analysis with a code structure matching the question elements of things that were
surprising, confusing, or frustrating, these elements were a vehicle needed for sufficiency of data
collection, rather than a conceptual taxonomy for analysis, and were not used as part of the analytical
process of constructing a code system . With the mindset of building on our understanding of what the
participants felt worked well or did not work well, we grew the code system using a grounded theory
approach. [6, 7] Using a card-sorting method, codes were grouped, split, or merged according to their
affinity to each other, and through multiple iterative sorting sequences, and re-examination of the
transcriptions or the associated audio tracks, we terminated analysis with four major code families,
described further under Results. The full code system, provided as a typology of radiology patient
Results
Communication was the most strongly discussed topic in the interview responses. It may seem trite to
say that ineffective communication was a central theme reported by participants, and perhaps it would
be more meaningful to say that communication about care processes and the lived experience of care
failed participants in very specific ways. The major area of ineffective communication related to
processes; what the patient should expect from them, what the experience would be like, and what to
While participants did report some problems with care delivery, by far the majority of negative
outcomes they reported related to the predictability and reliability of care and administrative processes,
rather than the care itself. A complicating factor was that participants viewed the care journey in far
broader terms than clinicians might expect, and experienced the bulk of their risks and issues with
regard to weaknesses in care processes in the interstices. Where clinicians may think in terms of the
care journey starting at the unit door or even at first encounter with a clinician in the unit, participants
were more likely to see the journey as starting with the motivating health event or symptom that led
them to seek care, and ending only with resolution of a health condition.
There are indeed islands of excellence of well-executed process fragments that moved efficiently and
expertly. Emergency Departments, for example, were related as being effective in maintaining process
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
integrity. Once a patient had navigated themselves to the radiology department, the processes wer4e
described as generally working effectively, but processes did not connect at the beginning or end with
The overall picture developed from analyzing the interview responses was of narrow and partially
overlapping clinical domains with poor integration and frequent handoff and process failures, for which
exhausted and frightened, but thankful patients, were acting as the overall process managers.
Analysis of the interview text resulted in four major code families, namely “Patient Needs,” “Patient
The “Patient Needs” code encompassed the accomodations that participants expressed regarding the
mental and physical space in which care took place. Participants described physical barriers to accessing
care, such as expensive or inconvenient parking, and the architectural design of facilities, as well as
needs such as care environment that is not noisy, cluttered, or smelly. A notable code was “Atypical
Need”, which describes patient conditions that do not fit the standard care protocols well, resulting in
care that was wasted, or that did not address the patient complaint.
Atypical Need 6
Care Environment 12
Disability 7
Parking 4
Physical Build 6
Participants descibed needs and experiences that reflected what we termed “Psychological Security”.
Several participants had mental health comorbidities that complicated the patient journey, and acted as
barrier to their ability to access and use care services effectively. Care environments that were low on
“Compassion”, for example, acted as a barrier to access, because participants were less motivated to
engage with facilities or staff that seemed less compassionate. Likewise, a sense of low compassion
reduced the efficacy of care provided, because participants attached less significance to advice from
people they regarded as not compassionate. In comparison, participants who developed coping
Comfort 2
Compassion 8
Coping Methods 4
Patient Confidence 4
Patient's Trust 2
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
The “Process” code was the most active code family, and was threaded through almost every topic of
discussion. It was perhaps unsurprising that issues and missed opportunities related to poor
communications featured as strongly as it did, but perhaps given popular sentiments regarding medical
insurance, surprising that insurance and fatigue with processes did not feature more strongly in how
Administrative Processes 13
Communication 30
Wayfinding 2
Health Records 5
Information Filtering 0
Insurance admin 2
Missed Opportunity 3
Organizational Learning 5
Patient Participation 11
Process Fatigue 1
Process Fragility 12
Process Opacity 30
Process Ownership 3
Process Silo 5
Rework 1
Standardized Language 2
Undocumented Processes 5
Unexpected Process 8
Unnecessary Variation 13
Waiting 14
Finally, the major codes included the negative outcomes that participants encountered in their journeys
through radiology. These codes reflected any negative experience or outcome, including any risks,
issues, events, sequelae, or missed opportunity that the participants identified during their radiology
journeys. Perhaps perculiar to radiology more than other specialties is the high frequency of
“Scanxiety”, which denotes the anticipatory stress and worry resulting from the delay between having a
radiology scan, and having the results explained and contextualized by a physician.
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
Abandonment 1
Confusing 4
Documentation Errors 1
Frustrating 8
Incomprehensive Assessments 1
Lost productivity 1
Scanxiety 11
Side Effects 10
Unexpected Effects 7
Unmet Expectations 1
Unprofessional Behavior 3
Discussion
The radiology patient population has unique, yet reasonable needs and expectations. These include
those tied to the care environment, their psychological well-being, and care requirements. The care
environment significantly influences the patient’s perception of quality and thus is a key determinant of
Process Failures
Participants spoke of fragile care processes that frequently did not start as expected, frequently ground
to a halt, or often failed to mesh well with the processes of other disciplines, unless the patient or their
caregiver were actively monitoring and mitigating them. In this sense, participants were continuously
acting as their own case manager, quality assessor, and process guide. Participants described how siloed
processes resulted in in negative outcomes such as being left alone over lunch breaks in an empty
consultation room, aborted sessions due to ineffective coordination of care, or radiology teams
performing scans that fit narrow protocols, but did not address the patient issues. In general, when
processes failed, the patient was left carrying the burden of integrating or restarting them.
Participants reported rare occurrences where care processes within a specific radiology treatment or
exam were at fault, but related frequent, persistent, and ubiquitous process failures whenever
processes between different specialties had to mesh, but failed to do so. Participants had exhaustive
examples of inter-specialty process failures, whether those were radiology results never getting to the
oncologist, physical therapist, or family doctor, or whether they were scheduling requests, referrals, or
queries from myriad specialties never arriving intact in radiology. In some cases, participants waited for
months for a radiology appointment that was never completely scheduled, or spent days trying to use
the provider scheduling system, only to have to call and manually book an appointment.
Participants reported ongoing unease and uncertainty about whether appointments had been booked
for the day, time, and location specified. Even when they were finally sure that a firm appointment was
scheduled, these might be moved or cancelled by unknown parties without notice. Likewise, even when
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
patients went to extraordinary lengths to specify their special needs, they would often arrive only to find
out that the service could not be completed as booked because the accommodations were not in place.
For one participant, this meant costly and worrisome rescheduling for a dependent with special needs,
and whose academic year dictated when lengthy radiology sessions could be scheduled.
Participants described being left alone in consultation rooms over lunch and having lights turned off or
doors locked as processes failed to mesh. One team placed a patient in a room without another team
knowing that this had occurred. In these cases, it was not that anyone was malicious or even negligent,
but that the processes were simply fragile, poorly engineered, and gave no warning to anyone that they
had failed.
Participants described how there was ultimately nobody other than themselves to take ownership of
end-to-end processes, to report on broken processes or those that had failed, or to even notice that
processes had failed. From scheduling never completed, to results never sent, participants felt that they
alone were the actors most likely to be hurt, and the ones most often making the processes work.
Participants shared psychological needs including mental comfort and the confidence that processes and
events will be predictable, be carried out with compassion, and uphold the trust that patients have in
Participants described physical needs that included sensory environment, seating location and
ergonomics, and storage. Participants described the negative effects of having television news at high
volume, unpleasantly high or low office temperatures, and inappropriate use of incense and other
odorants. Participants described issues relating to lack of effective storage for hand luggage and
personal items, and frequent lack of space for wheelchairs. Participants remarked on the benefits they
experienced where reclining chairs and adequate lockable storage for personal baggage and effects was
available.
Psychological needs were frequently unmet, and participants reported situations in which they felt
confused or anxious. These needs included predictable wayfinding to reach the radiology unit, and
dietary restrictions, or difficulties in finding the radiology department undermined participants’ trust in
their radiology teams. There was a strong sense that if simple expectations failed, the patient would
have future doubts about anything else the radiology team did. In essence, patients conveyed their
perceptions that “if you cannot get this simple thing right, how can I trust anything else you tell me.”
Participants reported being bewildered by confusing or contradictory signage and lack of clear
wayfinding markers to locate radiology departments. They were surprised by sudden changes in their
procedure preparations, unexpected sounds from equipment, or the lack of any sensation during
treatment.
Communication
Communication was the most discussed factor influencing the participant experience. Participants spoke
of confusion, frustration, and surprise as result of insufficient preparatory discussion about what to
expect, how it would feel, and what the possible post-treatment outcomes would be. In one example,
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
the participant had expected the radiological treatment to “feel like something,” and the lack of any
sensation of burning, tingling, or heat led them to doubt that the machine was on or working. In another
example, confusing signage and presence of a television news crew for an event led the participant to be
in a highly agitated state by the time they found the radiology department.
Nearly all of the participants referred to having anxiety during the waiting periods for receiving results.
During several of the interviews, this type of anxiety was referred to as “scanxiety,” the anxiety
experienced when waiting for the result of imaging or tests. Participants felt that communications of
In some cases, participants expressed a perceived lack of full disclosure by providers about clinical
findings. This resulted in feelings of distrust for some, and for others, reduced confidence in the
competence of their provider. Lack of disclosure raised suspicions that providers missed key details, or
lacked a full understanding of the connection of the findings to the specific condition.
Participants also felt disoriented by the failure of their providers to disclose the full range of possible
side effects, especially likely long-term effects of treatment. Unexpected side effects were often
associated with feelings of extreme anxiety and resulted in distrust of their providers. Although
participants acknowledged that it may be cumbersome to run down the full list of possible side effects,
but they believed that the onus is on the provider and their care team to provide this information prior
to discharge. Participants described side-effects and long-term effects not discussed with them, that
Given the nature of radiological services, patients are often accustomed to certain procedural flows.
Departures from the usual care procedures, without forewarning or explanation, was a concern for
many of the participants. Many reported that changes were often unnecessary and only compounded
the feeling of nervousness in an already stressful process. To mitigate their concerns, the participants
felt that providers should 1) avoid departures from their usual approach if not clinically necessary and 2)
Emotional Support.
Participants spoke of a need for emotional support to be overtly demonstrated by provider staff, and
participants reported often judging whether an encounter went well by how much compassion they
experienced. Participants spoke of how comforting it was to interact with care teams that knew them by
name, appeared to be concerned about their wellbeing, and expressed compassion when they needed
reassurance. Many participants made specific mention of such experiences with the technicians,
assistants, and administrative staff who consistently provided a personal touch. One participant
expressed gratitude for being given a moment to cry due to being overwhelmed during an encounter.
In contrast, participants also reported a feeling, due to the routine nature of radiology care, of being
rushed through treatments or scans without consideration for their personal psychological well-being
and security. To elaborate, participants reported that some providers failed to talk them through the
process and failed to ask about their patients’ comfort or their comprehension of the procedures in
progress.
In addition, participants spoke of missed opportunities during visits. Providers may be focused strictly on
the radiology aspects of an encounter, and not notice or react to signs that should drive up the index of
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
suspicion. One participant was receiving radiological care that exposed ample evidence of severe
domestic violence, but no one on the care team engaged the participant to see if they needed additional
resources or were in an unsafe situation. The participant advised that they would have greatly benefited
from the offer of additional support and resources at that time, and might have escaped an abusive
extensive coping mechanisms. Patients developed personal health networks (PHN), used mindfulness
methods such as yoga, built up their own health literacy, or scheduled encounters at times in the
Participants reported leveraging social media and friendship circles to build PHNs. They used their PHNs
to solicit assistance from radiologists, physicians, nurses, and technicians on social media in an effort to
explain procedures, processes, or findings. These PHNs were effective in helping participants to self-
manage their care journey, plan for encounters, or make sense of experiences or findings. While it may
be laudatory that patients are taking control of their care in this manner, it also reveals gaps and risks in
In one example, the instructions to drink no fluids prior to a scan were confusing, and the participant
solicited help through Twitter physician connections to explain the restrictions in the context of the
participant’s chronic condition. In another example, the participant used their social media connections
with a radiologist to help understand the findings of a scan. In a further example, a participant
exchanged spinal images with a radiologist in their PHN by sending a compact disk, and the PHN
radiologist spent over an hour on the phone to discuss the images and their implications.
Physical accommodations
Participants with and without physical limitations emphasized the importance of quick, seamless access
to the care facility. This included conveniently located parking, direct access to the department of
One participant specifically called out the arrangement of chairs in the waiting rooms. Many patients are
accompanied by parents and care givers, and in some cases are wheelchair bound. Given the layout of
most waiting rooms, wheelchair-bound patients are often unable to sit with those accompanying them
during a visit. Another participant suggested that the type of seating is often not well designed to
accommodate the physical challenges of the typical radiology patient. Participants appreciated
entertainment and informational resources tailored to patient needs, welcoming front desk staff, and
settings that are calm, but were not in favor of television news or political programming in the waiting
areas.
References
[1] D. M. Berwick, K. G. Shojania and B. K. Atchinson, "Free from harm: Accelerating patient safety
improvement fifteen years after To Err Is Human," Boston, MA National Patient Safety Foundation,
2015.
medRxiv preprint doi: https://doi.org/10.1101/2020.03.16.20030684. The copyright holder for this preprint (which was not peer-reviewed) is the
author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
[2] N. Proudlove, C. Hannibal and R. Boaden, "Lessons for Lean in Healthcare from Using Six Sigma in the
NHS," Public Money & Management, vol. 28, pp. 27-34, 2 2008.
[3] J. W. Creswell and J. D. Creswell, Research design: Qualitative, quantitative, and mixed methods
[4] H. J. Rubin and I. S. Rubin, Qualitative interviewing: The art of hearing data, sage, 2011.
[5] M. Loxton, "How to Use AI Transcription Services with MAXQDA," MAXQDA Research Blog, 29 4
2019.
[7] J. Saldaña, The coding manual for qualitative researchers, Sage, 2015.
Table of Figures
Figure 1 Patient Voice ...................................................................................................................................... 2