Prescribing an App?

Oncology Providers’
original report

Views on Mobile Health Apps for Cancer

Care
Introduction Although there are over 500 mobile health (mHealth) applications (apps) available
abstract

for download in the field of oncology, little research has addressed their acceptability among
health care providers. In addition, the providers’ perspectives regarding patient app use has been
largely unexamined. We conducted a qualitative study to explore opportunities and barriers for
mHealth app use for oncology care.
Methods We developed a structured interview guide focusing on acceptability, appropriateness,
feasibility, and sustainability of the use of apps in cancer care. We interviewed 15 oncology pro-
viders about their attitudes and preferences. De-identified audio recordings were transcribed and
coded for emerging themes.
Results Providers interviewed included physicians (n = 8) and advanced practice (n = 3) and
supportive services (n = 4) providers who care for a wide range of cancer types; ages ranged
from 32 to 68 years. Interviews lasted approximately 30 minutes. Oncology providers reported
limited exposure to mHealth apps in patient care, but were generally open to recommending or
prescribing apps in the future. Key themes included opportunities for mobile app use (including
general health promotion, tracking symptoms, and engaging patients) and barriers to implemen-
tation (including access to technology, responsibility, workflow, and the source of the app itself).
Conclusion Our results show openness among oncology providers to using mHealth technology as
part of patient care, but concerns regarding implementation. Designing acceptable apps may be
challenging and require involvement of key stakeholders, partnering with trustworthy institutions,
and outcome-based research.
Clin Cancer Inform. © 2017 by American Society of Clinical Oncology

INTRODUCTION Integrating apps into scalable interventions re-

Callie M.Berkowitz
Leah L.Zullig
Bridget F.Koontz
Sophia K.Smith
Author affiliations and
support information (if
applicable) appear at the
end of this article.
Corresponding author:
Sophia K. Smith, PhD,
MSW, School of Nursing,
Duke University, 307
Trent Dr, DUMC 3322,
Durham, NC 27710;
e-mail: sophia.smith@
duke.edu
Over the last decade, mobile devices have become
nearly ubiquitous; an estimated 77% of US
adults own a smartphone, and 51% own a tablet
computer.1 Mobile health (mHealth) applications
(apps) have been touted as a means to reduce
health disparity and improve patient care. Bil-
lions of dollars have been invested in mHealth
apps for IOS (internetwork operating system)
and Android.2-4 A systematic review published in
2016 found 539 mobile apps related to oncology,
117 of which were targeted toward patients.5
However, few (5.8%) were explicitly supported
by industry, and the majority (63.5%) made no
reference to scientific validation.5 At the same
time, patient interest in health apps is growing;
for example, an estimated 20% of US adults
use a wearable health device (eg, fitness bands,
smart watches).6
quires the commitment and support of both
patients with cancer and providers. In Europe,
there is early evidence that oncology providers
support the use of apps in clinical practice. In a
survey of 108 health care providers in Munich,
Germany, most respondents (84.3%) supported
the idea of an app to complement treatment. Cli-
nicians not in favor cited legal uncertainty, data
privacy, and possible problems with data stor-
age as potential barriers.7 It is unclear whether
oncology providers in the United States would
find mHealth apps appropriate for prescription
to their patients. Barriers may exist when recom-
mending mHealth apps; in a qualitative study
of gastroenterologists, all participants were reluc-
tant to endorse a digital inflammatory bowel dis-
ease tool because of concerns about the factual
material presented.8

© 2017 by American Society of Clinical Oncology ascopubs.org/journal/cci JCO™ Clinical Cancer Informatics 1

Downloaded from ascopubs.org by 181.46.139.130 on August 16, 2021 from 181.046.139.130

Copyright © 2021 American Society of Clinical Oncology. All rights reserved.
 Understanding provider perceptions is also
Table 1. Participant Characteristics (N = 15)
important as academic institutions, which are
Characteristic No. %
positioned to study the efficacy of apps via clin-
Sex
ical trial, begin to play a role in app develop-
ment. With the launch of Apple’s ResearchKit,  Male 7 47
institutions including Duke University, Stanford  Female 8 53
University, and Massachusetts General Hospital Provider type (credentials)
have launched mobile apps for research.9 Given   Physician (MD) 8 53
the commercial, academic, and health system   Advanced practice provider 3 20
interest in mHealth, the purpose of this qualita- (PA/NP)
tive study was to explore health care provider–   Supportive services (MS, PhD, 4 27
identified opportunities and barriers regarding MSW, RN)
app use for oncology care. Clinical days/week
 1 4 27

METHODS  2-3 6 40
 4-5 5 33
After reviewing the existing literature, our multi-
Age (years)
disciplinary team (including an oncologist, an on-
 30-39 4 27
cology social work and digital health researcher,
 40-49 7 47
and a health services researcher) developed
a structured interview guide inquiring about  50-59 3 20

mHealth apps in cancer care along with general   ≥ 60 1 7

demographic questions. Because our goal was to
assess potential implementation in clinical prac-
tice, we framed our guide around implementa-
tion outcomes of acceptability, appropriateness,
feasibility, and sustainability.10 Interviews were
piloted with clinicians (n = 2) and designed to
last 30 minutes.
We purposively sampled practicing oncology
providers (medical, radiation, or surgical) affiliat-
ed with Duke University Health System, includ-
ing physicians, advanced practice providers (eg,
nurse practitioners, physician assistants), and
supportive service providers (eg, social work-
ers, nursing support staff). The research team
identified potential participants (n = 23) meeting
these eligibility criteria on the basis of sugges-
tions of the senior author. The research team
then contacted participants via university e-mail.
A trained research team member conducted
the interviews in February and March 2017
in person or via telephone, in accordance with
participants’ preference. Interviews were record-
ed using Sparky recording equipment (USB re-
corder, Weston, CT) and manually transcribed.
Two researchers independently coded four de-
identified transcripts for emerging themes, which
were integrated and reconciled. Senior authors
reviewed and approved the resulting codebook,
which was iteratively revised. A single researcher
coded the remaining transcripts using the ap-
proved codebook. NVivo (version 11.4; QSR
Abbreviations: MS, master of science; MSW, master of social
work; NP, nurse practitioner; PA, physician assistant; RN,
registered nurse.
International, Doncaster, Australia) was used to
assist in coding and data management. Each
theme we highlighted was endorsed by seven to
10 participants; a limited number of represen-
tative quotes were selected by authors to pro-
vide a richer context. No personally identifying
information was captured in the interview. The
Duke University Medical Center Institutional
Review Board deemed this project exempt from
research.
RESULTS
Participant Characteristics
Fifteen providers participated in our study. Par-
ticipants were physicians (n = 8) and advanced
practice (n = 3) and supportive services (n = 4)
providers who cared for a wide range of cancer
types, including breast, hematologic, and gas-
trointestinal; ages ranged from 32 to 68 years
(Table 1).
Prescribing Apps: Present and Future
Few providers had experience discussing, rec-
ommending, or prescribing mobile apps to pa-
tients. Some providers were involved in mHealth

2ascopubs.org/journal/cci JCO™ Clinical Cancer Informatics

Downloaded from ascopubs.org by 181.46.139.130 on August 16, 2021 from 181.046.139.130

Copyright © 2021 American Society of Clinical Oncology. All rights reserved.
research, yet did not discuss apps with patients
in clinical care. A few had searched for mobile
apps to assist their patients, but did not find any
that met their criteria. Exceptions to this included
the use of Apple’s FaceTime (a video-conferencing
tool) for remote monitoring in bone marrow trans-
plantations and limited patient discussions about
activity motoring apps (eg, FitBit).
Most providers could see themselves recom-
mending apps to patients in the future if a clini-
cally appropriate and useful app existed.
I think [it] highly likely in the future [to recom-
mend or prescribe an app], I just feel, not quite
there yet, and maybe because I haven't put in
the due diligence to find the ones that might
work, but I think it’s more that they are not quite
there....They don't have the functionality.
I think it depends on the indications, obviously.
In my mind its more…what is the patient’s
problem and what is the best way to address
that, and if the answer happens to be an app,
then I guess I would be very interested, but not
just an app for the sake of it.
Many providers prioritized usability and reliabil-
ity over clinical trial outcomes when deciding
whether to recommend apps. Most felt mHealth
interventions to be generally low risk and consid-
ered patient preferences in their clinical decision
making.
My assumption or gut sense is, that the poten-
tial for harm is relatively low....I don't view this as
a medical device quote unquote that has to be
FDA [Food and Drug Administration] approved to
be able to put into clinical use for things around
education or patient tracking of their own course.
I think as a clinician, I’m just as likely to recom-
mend something if [patients] say “I just found
this useful” but that might be communicated in a
way different than, you know, classical outcome
measurements we may use for other things.
A minority of providers required more concrete
data about an app’s effectiveness or at least that
of its components:
if I knew of [an app] that had been pretty rigor-
ously tested and shown to be beneficial for cer-
tain things then I would be likely to prescribe it
for those certain things. But if it's not one that's
pretty rigorously testing and clearly improves out-
comes that I think are important to patients and
families, then I actually dissuade people from us-
ing some of them because I would wonder and
worry whether [the app is] even helpful.
Downloaded from ascopubs.org by 181.46.139.130 on August 16, 2021 from 181.046.139.130
Copyright © 2021 American Society of Clinical Oncology. All rights reserved.
Opportunities in Care Delivery
Health promotion and survivorship. When asked
for gaps in care that mHealth apps could ad-
dress, many providers cited general wellness
topics, including managing chronic conditions
(obesity, diabetes), monitoring diet and exer-
cise, improving compliance with medications,
and managing stress. The possibility of apps
addressing the financial burden of cancer treat-
ment, including navigating copay assistance
and financial planning, was also raised. These
functions were felt to be of particular importance
after treatment when patients have less direct
contact with providers.
I also think there’s an opportunity for people who
want apps around how to support behaviors,
especially in cancer…to get more engaged and
understand risk factors, so how should I modify my
diet, how should I modify my calorie count, how do
I exercise, how do I become more mindful….
I do think [apps] could help a lot with…medi-
cation compliance, a lot of people that we care
for are totally overwhelmed once they leave here,
from all the information. Medication changes or
dose changes, anything like that. So just some-
thing to keep them you know, on the right track
once they leave here, especially if we are not go-
ing to see them for a few weeks.
Symptom tracking and remote monitoring. Many
clinicians expressed interest in using apps to
collect patient-reported outcomes (PROs) related
to therapies. Clinicians noted that patients often
have difficulty describing symptom trajectories
over time, and real-time data input and trending
tools available within an app could assist. One
clinician described how such functionality could
be helpful in detecting the cumulative impact of
less severe or acute medication adverse effects
within clinical trials. Some saw extending further
into a means of monitoring patients remotely.
In oncology…patients’ symptoms [are] really im-
portant, and there’s a real interest in making sure
treatment is well tolerated and patient symptoms
improve with treatment and not just that they’re
living longer but also they’re feeling better…if a
person had an app that they put in that infor-
mation on a daily basis…then reports could be
generated that could come back in real-time to
an oncologist or other members of the team to
help improve patient symptoms in real time, as
opposed to just waiting for the patient to come
back and then remembering, how did I feel for
the past months?”
ascopubs.org/journal/cci JCO™ Clinical Cancer Informatics 3
 Patient engagement. Providers noted that apps
could help patients feel engaged at all stages of
treatment, from helping them communicate with
providers to bringing in caregivers.
I think there are a lot of ways that apps could
help engage patients both before and after, and
potentially even during clinical visits. There are
certainly ways to bring their caregiver and others
into that conversation.
Although not a concrete function, apps were
identified as being able to foster a sense of own-
ership and control during a challenging and
overwhelming time.
I think [an app] gives them some tools to use in-
stead of feeling like you know, we ripped the rug
out from under their lives and tell them what they
have to do, and any semblance of control that
they can have, I think that would be wonderful.
Implementation Challenges
Even in areas where they saw opportunity for mo-
bile apps, providers noted significant concerns
regarding implementation. Primary concerns
surrounded access to technology, workflow, re-
sponsibility, and source of the app itself (eg,
app development team, company, or institution;
Fig 1).
Access to technology. Although providers gen-
erally acknowledged growing smartphone use
among their patients, they remained concerned
about those who would not have access because
of a lack of finances or technologic literacy.
Advanced age and lower socioeconomic status
are both risk factors for cancer and potential bar-
riers to app use.
The reality is we are dealing with some folks that do
not have lot of sophistication. They may not have
access even to computers and [using mobile apps]
would be a very difficult thing for them to do.
Patients who might benefit most from an app’s
services might lack access, raising concerns of
equity. Related to access was the importance
that the app did not pose cost or financial bur-
den to patients.
Where I find it challenging, the people who need
the resources the most are the ones who typically
don’t have access to the resource…phones are
getting cheaper, but still.
Involving caregivers was cited as one means to
address concerns about technology access, with
4ascopubs.org/journal/cci JCO™ Clinical Cancer Informatics
Downloaded from ascopubs.org by 181.46.139.130 on August 16, 2021 from 181.046.139.130
Copyright © 2021 American Society of Clinical Oncology. All rights reserved.
Access
• Technological sophistication
• Cost and socioeconomics
• Equity
Source
• Commercial interests
• Instiutional branding/integration
• Involving key stakeholders
Usability
• Trust and fidelity
Responsibility
• Privacy/data security
• Responding to inputs
• Liability
Workflow
• Time burden/efficiency
• Clinical utility
Fig 1. Implementation challenges.
clinicians noting that oncology patients often
have more tech-savvy family members.
Workflow. In a health care system with increased
documentation requirements and compressed
patient visits, providers were concerned about
how apps would affect their clinic workflow.
Clinicians were concerned that any new tech-
nology would be burdensome in terms of time,
both in introducing the app to patients and
in any role it might play in clinic visits or data
collection.
That the doctor pretty much has to do nothing
[would make incorporating apps more feasi-
ble]. That’s the reality. I mean, the whole med-
ical system, there’s more and more stuff you’re
supposed to do and more and more paperwork
and more and more time stuff, and so every new
thing feels like a burden.
I think workflow is the big issue…that [apps] are
easy to access for patients, and any data that’s
generated from those apps and from those in-
teractions are integrated into the existing EHR
[electronic health record].
Related to workflow concerns was the idea that
apps need to provide true clinical utility and
usability to mitigate demands on clinician time.
An unreliable or poorly designed app could cre-
ate more work, with providers acting as tech
support. However, some clinicians saw the ability
of apps to make clinic visits more efficient, for
example, by improving clinical data collection.
The app could have some functionality that helps
set the agenda or moves forward the clinic visit
in a positive way.
 I can envision that coupled with mobile devices
that are sensor, that we can probably cut our
clinic visits by three fourths [time].
Responsibility. Providers recognized that apps
might generate large amounts of data via report-
ing functions and sensors. Although providers
were concerned with protecting patient privacy
and the implications of collecting sensitive
health information, they also raised the issue of
how to respond to that data. For example, clini-
cians were concerned that patients might report
critical symptoms to an app, from which data are
not being actively monitored, instead of calling a
triage line or seeking care.
If [the app is] a reporting tool like patient-report-
ed outcomes or if patients are entering any data
into it regarding their health state, I would be
really concerned about who looks at that data
and who responds to it, so not so much from
the security side, but more who’s responding to
patients that they’ve got a particular problem,
whether it’s a worsening symptom or you know,
some sort of psychological problem or psycho-
social issue. The timeliness of the response is
important.
Providers were concerned that apps could have
unintended outcomes, leading to issues of pa-
tient care and provider liability. Clinicians often
wondered what medicolegal implications recom-
mending an app could pose.
If [patients] report on [a mobile app] they are do-
ing well and I just assume that's correct, I may be
missing something.
There's always the risk of misdirection, there's
always the risk of misunderstanding, things that
would lead patients to do quite the opposite of
what we hope the intent would be.
Source of app. Many providers cited the source
of the mobile app, whether it was developed in-
ternally or externally or by a commercial compa-
ny or respected institution, as being important.
Some clinicians were concerned about mitigat-
ing conflicting interests that commercial apps
may pose, acknowledging that some apps may
be created primarily for profit.
So I think clinicians want to be very careful about
endorsing something that may have undue in-
terests within it, that are not well known. So, I
don’t think clinicians are against a company, a
for-profit entity for example sponsoring funding
or otherwise, you know, maintaining the sustain-
ability of something as long as there are really,
Downloaded from ascopubs.org by 181.46.139.130 on August 16, 2021 from 181.046.139.130
Copyright © 2021 American Society of Clinical Oncology. All rights reserved.
really, really clear assurances that there’s no
influence on the content of the app, for example.
The source of the app may also provide assur-
ance to the provider that the content of the app
is credible and up to date.
If I knew it was created by an entity that was…
trustworthy and not going to be swayed or bi-
ased…then I would feel comfortable [recom-
mending an app].
Institutional development and branding, although
not a requirement, could help enhance patient
engagement and connect with other patient tools
(eg, EHR patient portals). The team creating the
app may be important for patient adoption as well.
I would say there is a lot of validity and probably
uptake downstream when users know that clini-
cians, patients like them, and other aligned with
their interest are part of the development and
testing process, I think that’s a really important
message.
DISCUSSION
Health care providers are important and under-
studied stakeholders in mHealth app design
and implementation. Despite the commercial
enthusiasm for mHealth apps, we found that
discussions about mobile tools rarely occur in
the clinical setting for oncology. However, most
providers could see themselves prescribing
apps to patients in the future. Several desired
app functions have been explored in the liter-
ature, including apps for oral chemotherapy
adherence,11 collecting PROs,12 and promoting
healthy behaviors.13 Providers were concerned
about equitable access, data safety and respond-
ing to patient inputs, and navigating commercial
interests. Although most concerns were found
across provider groups, physicians in particular
were concerned with how an app might affect
clinic workflow.
Providers may lack the time and resources to fully
explore and assess available apps. Although the
vast majority of mHealth apps currently available
for patients to download have no scientific valida-
tion, formal randomized clinical trials may not be
necessary to promote adoption among oncology
providers. At present, the FDA oversight focuses
on apps that “present a greater risk to patients if
they don’t work as intended.”14 For some provid-
ers, studies of usability and patient satisfaction,
and thoughtfulness toward core functionality,
ascopubs.org/journal/cci JCO™ Clinical Cancer Informatics 5
 may be sufficient. However, evidentiary require-
ments vary among providers on the basis of the
type of app.
Providers expressed concerns regarding access
among older and lower socioeconomic status
patients. Although US smartphone adoption
rates are lower for individuals older than 65 years
of age (42%) and individuals making < $30,000
a year (64%), usage has increased dramatically
since 2011. Smartphone adoption rates are
slightly lower among African Americans (72%)
and Hispanics (75%) versus whites (77%).1
Our study has several limitations. The provid-
ers interviewed were all affiliated with the same
academic institution, although practice envi-
ronments varied. The providers who agreed to
participate in our study may have been more
interested in digital health technology than the
average provider. Six providers were involved
in mHealth research studies. These providers
were more likely to question roles and responsi-
bilities regarding app usage and data and were
less likely to discuss technology access issues.
AUTHOR CONTRIBUTIONS
Conception and design: All authors
Provision of study material or patients: Sophia K. Smith
Administrative support: Leah L. Zullig
Collection and assembly of data: Callie M. Berkowitz,
Sophia K. Smith
Data analysis and interpretation: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
AUTHORS' DISCLOSURES OF
POTENTIAL CONFLICTS OF INTEREST
The following represents disclosure information provided
by authors of this manuscript. All relationships are
considered compensated. Relationships are self-held
unless noted. I = Immediate Family Member, Inst = My
Institution. Relationships may not relate to the subject
matter of this manuscript. For more information about
ASCO's conflict of interest policy, please refer to
www.asco.org/rwc or ascopubs.org/jco/site/ifc.
Affiliations
Callie M. Berkowitz, Leah L. Zullig, and Sophia K. Smith, Duke University; Leah L. Zullig, Durham Veterans Affairs Health
Care System; and Bridget F. Koontz and Sophia K. Smith, Duke Cancer Institute, Durham, NC.
Support
Supported by Duke Institute for Health Innovation, Durham, NC. C.M.B. is supported by the Duke Institute for Health
Innovation. L.L.Z. is supported by Veterans Affairs Health Services Research and Development Career Development
Awards (CDA 13-025).
6ascopubs.org/journal/cci JCO™ Clinical Cancer Informatics
Downloaded from ascopubs.org by 181.46.139.130 on August 16, 2021 from 181.046.139.130
Copyright © 2021 American Society of Clinical Oncology. All rights reserved.
Although technology literacy may affect behav-
ior, we did not formally access these skills in
our providers; however, all routinely use EHRs
and conversed comfortably about mobile apps.
Participants may have also responded more
favorably knowing that our research team is in-
volved in developing mHealth tools. However,
insight from providers interested in mHealth
apps still showed potential areas for growth and
barriers that limit app development in clinical
care.
Enthusiasm for mHealth apps among oncology
providers is tempered by significant challenges,
spanning technology, implementation science,
and ethics. Concerns are greatest for apps that
collect data from patients, although symptom
tracking is a sought-after functionality given
emphasis in oncology on PROs. Partnering with
trustworthy institutions to develop apps can help
mitigate quality concerns.
DOI: https://doi.org/10.1200/CCI.17.00107
Published online on ascopubs.org/journal/cci on
December 21, 2017.
Callie M. Berkowitz
No relationship to disclose
Leah L. Zullig
Honoraria: Novartis
Bridget F. Koontz
Consulting or Advisory Role: Blue Earth Diagnostics, Bayer
Research Funding: Janssen
Patents, Royalties, Other Intellectual Property: UpToDate
Other Relationship: Sanofi, Algeta ASA, Medivation, Exelixis
Sophia K. Smith
Research Funding: Pfizer Independent Grants for Learning
and Change
Acknowledgment
We thank Cristina Van Sant for her help with qualitative
coding.
Prior Presentation
Presented at the Duke Digital Health Conference, Durham, NC, April 12, 2017.

REFERENCES
1. Pew Research Center: Mobile fact sheet.http://www.pewinternet.org/fact-sheet/mobile/
2. Eapen ZJ, Peterson ED: Can mobile health applications facilitate meaningful behavior change? Time for answers.
JAMA 314:1236-1237, 2015
3. Schoenfeld AJ, Sehgal NJ, Auerbach A: The challenges of mobile health regulation. JAMA Intern Med 176:704-
705, 2016
4. Quintiles IMS Institute: Patient adoption of mHealth. http://www.imshealth.com/en/thought-leadership/
quintilesims-institute/reports/patient-adoption-of-mhealth
5. Brouard B, Bardo P, Bonnet C, et al: Mobile applications in oncology: Is it possible for patients and healthcare
professionals to easily identify relevant tools? Ann Med 48:509-515, 2016
6. Agboola SO, Bates DW, Kvedar JC: Digital health and patient safety. JAMA 315:1697-1698, 2016
7. Kessel KA, Vogel MM, Schmidt-Graf F, et al: Mobile apps in oncology: A survey on health care professionals’ attitude
toward telemedicine, mHealth, and oncological apps. J Med Internet Res 18:e312, 2016
8. van Mierlo T, Fournier R, Fedorak R: Don’t forget the doctor: Gastroenterologists’ preferences on the development
of mHealth tools for inflammatory bowel disease. JMIR Mhealth Uhealth 3:e5, 2015
9. Hsin H, Torous J, Roberts L: An adjuvant role for mobile health in psychiatry. JAMA Psychiatry 73:103-104, 2016
10. Proctor E, Silmere H, Raghavan R, et al: Outcomes for implementation research: Conceptual distinctions,
measurement challenges, and research agenda. Adm Policy Ment Health 38:65-76, 2011
11. Fishbein JN, Nisotel LE, MacDonald JJ, et al: Mobile application to promote adherence to oral chemotherapy and
symptom management: A protocol for design and development. JMIR Res Protoc 6:e62, 2017
12. Rowe L, Dong T, Armstrong T, et al: Abstract 1433: A mobile app for health related quality of life and symptom
assessment in patients with primary brain tumors in an outpatient oncology clinic. Cancer Res 76:1433-1433,
2016 (14 suppl)
13. Vollmer Dahlke D, Fair K, Hong YA, et al: Apps seeking theories: Results of a study on the use of health behavior
change theories in cancer survivorship mobile apps. JMIR Mhealth Uhealth 3:e31, 2015
14. U.S. Food and Drug Administration: Mobile medical applications. https://www.fda.gov/medicaldevices/
digitalhealth/mobilemedicalapplications/ucm255978.htm

ascopubs.org/journal/cci JCO™ Clinical Cancer Informatics 7

Downloaded from ascopubs.org by 181.46.139.130 on August 16, 2021 from 181.046.139.130

Copyright © 2021 American Society of Clinical Oncology. All rights reserved.