Content uploaded by Karim Keshavjee
Author content
All content in this area was uploaded by Karim Keshavjee on Oct 21, 2016
Content may be subject to copyright.
DOI: 10.4018/IJHCR.2016040103
Copyright © 2016, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Volume 7 • Issue 2 • April-June 2016
Pannel Chindalo, InfoClin, Toronto, Canada
Arsalan Karim, InfoClin, Toronto, Canada
Ronak Brahmbhatt, InfoClin, Toronto, Canada
Nishita Saha, InfoClin, Toronto, Canada
Karim Keshavjee, InfoClin, Toronto, Canada
The mobile health (mhealth) app market continues to grow rapidly. However, with the exception
of fitness apps and a few isolated cases, most mhealth apps have not gained traction. The barriers
preventing patients and care providers from using these apps include: for patients, information that
contradicts health care provider advice, manual data entry procedures and poor fit with their treatment
plan; for providers, distrust in unknown apps, lack of congruence with workflow, inability to integrate
app data into their medical record system and challenges to analyze and visualize information
effectively. In this article, the authors build upon previous work to define design requirements for
quality mhealth apps and a framework for patient engagement to propose a new reference architecture
for the next generation of healthcare mobile apps that increase the likelihood of being useful for and
used by patients and health care providers alike.
App Design, Architecture, Behavior Change, Health App, mHealth App, Patient Engagement, Patient Reported
Experience Measures (PREM), Patient Reported Outcome Measures (PROM), Self-Management
The popularity and usage of mobile technology continues to boom (Research2guidance, 2015).
Increasingly, people are inclined to seek guidance from smartphones than from other persons (Elias,
2015). Smartphones’ ascendance to this level is highly associated to its practicality in communicating,
information resourcefulness, portability and flexible costs for most people, regardless of their economic
status (Silow-Carroll & Smith, 2013). Mobile health (mhealth) care applications are seeing a similar
boom. Of the millions of apps in circulation, about 45,000 are mhealth apps (Research2guidance,
2015). More than half of these mhealth apps are new on the market. Thirteen percent of these
apps were introduced in the first quarter of 2015. However, most mhealth apps are not used in
healthcare despite their growing popularity (in terms of downloads) and potential medical purposes
(Research2guidance, 2015).
Researchers from various backgrounds have proposed ways that consumers can select useful
mhealth apps for their health and health information needs (Albrecht, 2013; Boudreaux, 2014; Powell,
2014; Kumar, 2013). Several studies have identified hurdles that challenge wide usage, including
poor user interface designs, differing user literacy levels, implementation issues and organizational
structures (Bailey, 2014, Boudreaux 2014; Brown 2013; Caburnay, 2015; McMillan, 2015; McCurdie,
34
Volume 7 • Issue 2 • April-June 2016
35
2012). These efforts have so far not been successful at wooing patients and care providers to greater
use of mhealth apps. Health app developers are caught in a dilemma of not knowing how to overcome
these hurdles.
In this paper, we build upon Albrecht et al.’s “synopsis for apps” in health care (Albrecht, 2014)
to propose a novel reference architecture for mhealth apps that can overcome current barriers. Albrecht
et al provide a comprehensive framework for mhealth app publishers to describe their compliance
with a variety of pragmatic and evidence-informed criteria that are worth considering when evaluating
apps. We found their framework useful as a scaffold for considering important elements of an app
during the design process.
We also build upon the patient engagement framework developed by Balouchi et al. that describes
an enabling environment for engagement and communication between patients and providers
(Balouchi, 2014). We propose a refined approach for engaged communication between patients and
care providers. This approach focuses on the patient and care provider relationship as the starting
place to add value that is likely to grow exponentially in ways we can only now imagine. Our approach
considers the constraints identified by the studies cited above in order to identify the critical functions
that can elevate how apps can deliver added value. Further, we propose an architecture for mhealth
apps that arranges the critical functions identified in order to accomplish the following: (a) capture,
validate and communicate data about the processes and outcomes of a disease; and (b) enable on-
going communication during treatment to enhance the patient-care provider relationship and ensure
patients get the support they need to implement the advice and interventions prescribed by their
health care provider.
We conducted a literature search in PubMed and Google Scholar to identify articles that described
methods to evaluate mobile apps. We utilized the related articles feature to find additional articles.
We also identified articles on mhealth architecture and patient engagement with apps. We conducted
a narrative synthesis of the studies we identified and applied a critical analysis by way of identifying
common hurdles that restrict wide usage of mhealth apps. Our process of deliberation comprised
distributing the studies we identified as critical to the research topic. We had three workings sessions:
one concept development meeting which led to designating responsibilities for drafting the study’s
sections; a rethinking and refinement session and a final interdisciplinary discussion to finalize the
story arc of the study. We also utilized Google Drive file sharing to coordinate our communication.
We used a gap analysis that drew on philosophy, data science, education, life science and business
analyses methods to develop a concept that would overcome the constraints and meet the goals
identified in the introduction. Through analysis, discussion and iteration, we arrived at a proposed
architecture that is evidence-informed, uses validated tools effectively and is situated in a philosophy
that puts a high value on the patient-physician relationship.
Other than fitness tracker apps and some tethered apps provided by insurance companies and
by integrated delivery systems, most mhealth apps in app stores have not gained much traction
(Research2guidance, 2015). We hypothesize that this is due to multiple factors: a) apps may provide
information that conflicts with information received from health care providers (Bierbrier, Lo & Wu,
2014); b) the language and terminology of the app may not be compatible with the patient’s health
Volume 7 • Issue 2 • April-June 2016
36
literacy (Caburnay, 2015); c) the patient has to enter the data himself or herself (Gruman, 2013); d)
the patient has no way to use the information in a meaningful way; e.g., they cannot order diagnostic
testing for or prescribe medications to themselves; e) daily use of the app is not required for most
diseases and therefore the patient does not get into the habit of using it; f) lack of incentives to use,
such as cost savings or social approval; g) providers may not value or use the data collected by patients
in apps downloaded from an app store whose provenance and pedigree is not known or established
(Terry, 2015); h) there is no way for providers to consume (i.e., visualize, analyze, derive meaning
from) the large amounts of data that are collected in apps (Terry, 2015) and i) there is no way for
providers to integrate the app data into their own electronic medical record system (EMR) for analysis
or follow-up or share the data in their EMR with their patient’s apps (Abebe, 2013).
In our search, we identified authors who are pioneering a paradigm that is supportive of a
scientific validation process and a technology convenient platform. The Open mHealth group (Chen,
2012) has developed some sophisticated tools to assist app publishers in designing and developing
useful mhealth apps. Their open source tools allow developers to obtain standardized data from other
systems and display them in high quality visualizations. However, the “architecture” they present is
focused entirely on an information technology view. There is no proposed business architecture or
information architecture. Thus, developers are left to their own devices on how they will connect the
various pieces of technology to create a viable technology. Silow-Carroll and Smith (Silow-Carroll,
2013) describe apps that have had some success in healthcare. They point to several features which
have gained some success, including having apps prescribed by health care providers who monitor
their use and patient outcomes. These exemplars are mostly found in leading health care systems like
Geisinger and Kaiser Permanente. However, they do not propose a comprehensive architecture that
can be used by mhealth developers as a reference architecture.
Our approach, illustrated in Figure 1, is an adaptation of Balouchi et al.’s “enabling environment”
for patient engagement. Successful use of mhealth apps is predicated on having an enabling
environment, including the functions shown in the margins of the diagram (Balouchi, 2014). To this
enabling environment, we propose that mhealth apps must, in addition, provide useful functions to have
value for health care providers and patients. At a high level, this means that the app must be able to
capture clinical information about the patient that providers can use for their clinical decision-making
and treatment recommendations. And, the app must also be able to communicate advice, education,
information and treatment recommendations from providers to the patient. This simple architecture, if
applied correctly, can be a powerful driver of change through on-going feedback, accountability and
congruence between patient and provider; something that is sorely missing in current mhealth apps.
As previously mentioned, some of the leading health systems in the US, such as Geisinger and
Kaiser Permanente, have already started along this path. Other care models such as the Accountable
Care Organizations (ACOs), the Patient Centred Medical Homes (PCMH) and the Ontario Family
Health Teams (FHTs) are all care models that have elements of the enabling environment in place
for chronic disease management. Health care providers working within these organizations already
have the infrastructure and the reimbursement model (salaried or capitation funding) to facilitate use
of mhealth apps within their practice offerings.
Patients and healthcare providers have for centuries operated in a relationship that demonstrates
trust, skills, dependence and knowledge. Yet, most mhealth apps try to avoid getting entangled in
this relationship. It would be helpful to introduce mhealth apps into the patient and health provider
relationship (Edelman & Singer, 2015).
Most commercial mhealth apps do not take into consideration the critical role of the patient-
healthcare provider relationship. A patient and health provider relationship is important for patients
to understand the difference between a gadget for entertainment and a tool for improving health. In
order to understand how seemingly this small difference can have an outsize impact, we have to see
it from patients’ and healthcare providers’ viewpoints. On the one hand, a patient’s perspective is
Volume 7 • Issue 2 • April-June 2016
37
informed by a culture of excitement for hi-tech gadgets that has become part of daily life experiences.
Consumers have developed an insatiable appetite for hi-tech gadgets that compete for their attention.
This high expectation not only drives business, but it also reduces the shelf-life of hi-tech gadgets.
This cultural expectation could explain the lack of retention of mhealth apps. On the other hand, a
healthcare provider’s perspective on hi-tech is very different when it comes to professional practice.
A health provider makes a clear distinction between a gadget for entertainment and how to evaluate
its use in a professional environment. There are a lot of apps that give health information for personal
use, but this type of information may not be credible for a licensed healthcare provider to use when
advising a patient in a healthcare binding relationship. A licensed healthcare provider is trained to
advise patients by using information that is acceptable by a professional code of standards as they
evaluate a patient’s condition and a patient’s overall health story. An evidence-based assessment that
a health provider performs before advising a patient requires an understanding of approved evidence-
based knowledge about the problem, transforming this knowledge in view of the patient’s condition
based on clinical experience and arriving at a final treatment decision based on discussion with the
patient about their values and preferences. For our discussion purpose, it means that both the patient
and the healthcare provider do not consider standalone mhealth apps as credible resources in their
healthcare relationship. In practice, it simply means that an app fails a reality-test to patients and
health providers for reasons of entertainment and unreliable measurements respectively. In order to
gain acceptance, mhealth apps need to be designed to enter, meet, and enhance the patient and the
health provider relationship. A patient and a health provider relationship is based on trust and skills,
as such apps should be designed in a way that they could become valuable resources that fill-in the
gaps in knowledge and support in a patient-provider relationship.
The task of advising patients to perceive apps as a form of “prescription” in order to enhance
treatment and communicate a sense of accountability depends on the health providers’ willingness
to use apps. Patients are likely to associate an mhealth app as a treatment plan when health providers
introduce to it to them in that way. Educating healthcare providers about how an app meets the scientific
standards of measurements and demonstrating how user-friendly designs account for their diverse
clientele’s educational levels and age groups is crucial to their acceptance. User-friendly designs,
Figure 1. Patient-Health provider centered approach (Adapted from Balouchi)
Volume 7 • Issue 2 • April-June 2016
38
ability to capture important information between visits, ability to monitor follow-throughs on patients,
reduced time pressure during encounters and higher productivity are likely to self-advertise apps into
the patient and healthcare provider relationship. Licensed health providers have considerable moral
and psychological influence within the patient-provider relationship to rebrand and introduce mhealth
apps as critical communication devices during monitoring or treatment phases.
Many apps are proving that useful algorithms can be generated without a scientific validating process.
So why do we need a scientific validating process for mhealth apps? We use an illustration from
the sporting industry. It is intriguing to observe how excited people become after a major sporting
event like tennis or soccer. In excitement, people go to soccer fields and tennis courts to emulate the
skills they watched on television. Soon, most fans realize that it requires above average skills and
body conditioning to perform like conditioned athletes. Technology has a way of exciting people
who don’t understand it to think big but are soon disappointed by their lack of skills. Just because
an app can take a blood pressure measurement and produce information about our bodies does not
make patients experts in healthcare. Scales and measurements used for assessing and treating patients
should be left to the experts.
Human ailments are diverse and specializations in health have developed validated ways of
measuring and assessing diseases. mHealth apps could customize these scales and make them available
for patients and healthcare providers. Success in patient treatment has depended on healthcare providers
and patients as “managers” of treatment plans. It makes sense that this relationship be supplied with
first class resources. The expediency of technology could enable treatment resources with an added
value of communicating, monitoring and encouraging the patient in real-time from any location.
It is important to realize that the worth of specialized scales comes to make sense only when a
patient’s information is aligned with it to enable interpretation. The need to make precise interpretation
about a patient’s condition by using verifiable measurements that have been scientifically validated
is critical for making a diagnosis and individualizing a treatment plan to the unique needs of each
patient. The ease of appearance that technology gives healthcare makes it appear flatteringly simple,
but the disciplined body of knowledge it takes to diagnose and set a treatment plan is complex.
Figure 2 describes in further detail the types of information required for appropriate management
of a patient in the context of the patient-provider relationship. More description is provided in the
section entitled, Process and Outcomes.
Inherent within the patient-provider relationship, there is a power and information asymmetry.
Physicians have access to and control access to information generated by the health system, such as
laboratory results, specialist reports, imaging reports and other health care data. Introducing mhealth
apps in this relationship could empower patients and health providers in ways that could improve
health care systems. The current structure positions the patient in a subservient role. The reasons
are obvious, the patient describes the problem, and the expert interprets what it means according to
the patient’s medical history. In practice a healthcare provider may request tests in order to identify
the cause of the problem. As expected the healthcare provider informs and advises a patient about
a diagnosis. In most cases patients have little or superficial knowledge about their diagnosis. When
patients are given an opportunity to ask questions, they are still at a disadvantage of lacking knowledge
to ask informative questions.
Many healthcare providers are well trained in delivering diagnosis news and proceeding further
into treatment plans. In summary patients’ dependence on healthcare providers is necessitated by a
structure that limits patients and healthcare providers to spaces in buildings. There are many ways
that apps would elevate the nature of this relationship. The potential of apps to collect information
from patients during treatment is critical for empowering patients to speak about their experiences
Volume 7 • Issue 2 • April-June 2016
39
in real-time. This process can lead to collecting data based on patients’ experiences and treatment
outcomes in order to improve quality, patient awareness and professionalism.
A trusted structure or framework speaks to the idea that content of apps need to be evidence-based
(Albrecht, 2014). It is well-known that information technology exerts its effect through automating
work flows and work processes; it is not a magic wand. An app is unlikely to provide valuable
information without a defined process on how information should flow.
A validated evidence-based process requires a purposive procedure for ensuring the accuracy of
patient information that is collected. For example, Coleman (2009) provides a chronic disease model
that exemplifies a process of collecting information on a chronic disease.
Figure 2 describes a simple, but powerful information and business workflow that is centered
on the patient-provider relationship. A mhealth app that meets evidence-based content and process
requirements would embody the following properties:
1. It would explicitly identify the patient’s diagnosis to ensure that the patient is eligible for guideline
or other evidence-based content. Further, the physician should make the diagnosis and the app
should be ‘prescribed’ by the physician as part of a course of treatment for the patient (see Figure
3). The app should interoperate with the EMR to allow this prescription functionality to activate
it only when authorized by a licensed provider.
2. It would identify and track the process and proxy metrics for that disease (e.g., frequency of
visits required for monitoring the disease, the biomarkers and/or patient questionnaires used to
monitor disease severity for a particular disease such as HbA1c, blood pressure, PHQ9 scores,
pain scores, etc.). Modern apps and Big Data approaches require as much data as possible to
monitor and evaluate a patient’s progress. Gamification data and other user behavior data is
useful to an extent where its metric is assessed within a context of other validated metrics that
correlate with the disease.
3. Identify and track important outcome measures. This includes clinical outcomes such as heart
attacks, specialist consultations, medical procedures, hospitalization, strokes and death, which
will come from the EMR. Patient reported outcome measures (PROMs) of all types need to be
captured for comprehending patient specific patient health profile and data collection. Patient
reported experience measures (PREMs) are also necessary to understand patient’s subjective
experiences that may have relevance to treatment adherence or goals. Any and all of the above
outcome measures may be relevant for a particular disease and predictable based on process and
proxy measures collected as described above. These need to be part of the mHealth app, since
they are unlikely to be made available by the EMR vendor.
The concepts described above fit very nicely into Albrecht et al.’s framework (Albrecht 2014)
into Item 4 Validity and Reliability, creating new Sub Items which we would label 4.3 Process and
Proxy Measures and 4.4 Outcomes and Outcome Measures.
Figure 2. Communicating process and outcomes in mHealth apps
Volume 7 • Issue 2 • April-June 2016
40
It is important to understand why this functionality must come from the mHealth app and not
from the EMR. Most EMR vendors do not have the clinical knowledge or expertise to develop the
type of functionality required for the mHealth apps described in this paper. mHealth app developers
need to understand the entire value chain for the management of a particular disease and then make
that available to patients and providers in their respective ehealth platforms –the PHR and the EMR.
In order to ensure credibility of the data, which is critical for patient follow-up, the provenance
of information (i.e., where it originates) is very important. For example, a physician should enter
a diagnosis. A patient entered diagnosis, although useful, could be a sign that the physician lacks
commitment to patient engagement, patient self-management and field professionalism. It could also
be erroneously interpreted or understood by the patient. This means that the health care provider
should activate the app to ensure that it is part of a “prescribed” intervention.
The app also needs to have interoperability with EMRs. This would provide the patient with
important information that is recorded in the EMR by a clinician during a patient encounter or that is
reported by various health care organizations to the physicians EMR (Figure 4). Interoperability is also
required for patients to report important information such as process measures and self-monitoring
information back to the physician or their designate for action or to update their records. There are
an increasing number of interoperability platforms available to support this, including Open mHealth
(Open mHealth, 2015), SMART Health IT (Smart Health IT, 2015) and the Apple HealthKit.
For the patient to be able to get appropriate support for questions about new information received
or new symptoms or issues that arise from time to time, the app should allow for messaging back to
Figure 3. Sequence of Interactions between provider and patient in an mHealth app
Volume 7 • Issue 2 • April-June 2016
41
the physician or their designate. The app should also allow the provider to communicate important
updates and disease relevant information to the patient (Figure 3).
The concepts described in this section also fit well into Albrecht et al.’s (Albrecht, 2015)
framework as additions to Item 5 Data requisitioning and management as new Sub Item 5.2 EMR-
App Interoperability and 5.3 App-EMR Messaging.
This paper discusses a novel, interoperable architecture that can help mhealth apps become widely
used as an asset for the enhancement of health care by strengthening the patient-provider relationship.
The paper presents the difficulties that are impeding the acceptance of apps in health care. Although
the hurdles for apps usage in health care are several, the paper focused on introducing apps into the
patient and healthcare provider relationship because of the respect it commands among patients.
By introducing mhealth into the patient and healthcare provider structure, especially physicians,
it would communicate the importance apps can bring to healthcare. It is possible that patients and
culture could eventually woo healthcare practitioners into accepting mhealth apps but the risks of this
approach would require more work. In contrast, most healthcare providers are likely to be persuaded
if apps were to embrace age-old cultural norms and values of the patient-provider relationship, core
scientific methods of using validated measures in order to have confidence in the measurements that
apps generate and interoperability that ensures two way communications. In an effort to differentiate
our approach from providers ‘prescribing’ mhealth apps from the app store, we incorporate validating
capabilities, communication and interoperability as keystones to facilitate patient empowerment,
engagement and follow-up within the context of the patient-provider relationship.
Figure 4. Reference architecture for mHealth app integration with EMRs
Volume 7 • Issue 2 • April-June 2016
42
Abebe, N. A., Capozza, K. L., Des Jardins, T. R., Kulick, D. A., Rein, A. L., Schachter, A. A., & Turske, S. A.
(2013). Considerations for community-based initiatives: Insights from three Beacon Communities. Journal of
Medical Internet Research, 15(10), e221. doi:10.2196/jmir.2803 PMID:24128406
Albrecht, U. V. (2013). Transparency of health-apps for trust and decision making. Journal of Medical Internet
Research, 15(12), e277. doi:10.2196/jmir.2981 PMID:24449711
Albrecht, U. V., Pramann, O., & von Jan, U. (2014). Synopsis for health apps–transparency for trust and decision
making. In Social Media and Mobile Technologies for Healthcare. Hershey, PA, USA: IGI Global.
Bailey, S. C., Belter, L. T., Pandit, A. U., Carpenter, D. M., Carlos, E., & Wolf, M. S. (2014). The availability,
functionality, and quality of mobile applications supporting medication self-management. Journal of the American
Medical Informatics Association, 21(3), 542–546. doi:10.1136/amiajnl-2013-002232 PMID:24163156
Balouchi, S., Keshavjee, K., Zbib, A., Vassanji, K., & Toor, J. (2014). Creating a Supportive Environment
for Self-Management in Healthcare via Patient Electronic Tools. Social Media and Mobile Technologies for
Healthcare, 109.
Bierbrier, R., Lo, V., & Wu, R. C. (2014). Evaluation of the accuracy of smartphone medical calculation apps.
Journal of Medical Internet Research, 16(2), e32. doi:10.2196/jmir.3062 PMID:24491911
Boudreaux, E. D., Waring, M. E., Hayes, R. B., Sadasivam, R. S., Mullen, S., & Pagoto, S. (2014). Evaluating
and selecting mobile health apps: Strategies for healthcare providers and healthcare organizations. Translational
Behavioral Medicine, 4(4), 363–371. doi:10.1007/s13142-014-0293-9 PMID:25584085
Brown, W. III, Yen, P. Y., Rojas, M., & Schnall, R. (2013). Assessment of the Health IT Usability Evaluation
Model (Health-ITUEM) for evaluating mobile health technology. Journal of Biomedical Informatics, 46(6),
1080–1087. doi:10.1016/j.jbi.2013.08.001 PMID:23973872
Caburnay, C. A. (2015). Evaluating diabetes mobile applications for health literate designs and functionality,
2014. Preventing Chronic Disease, 12. PMID:25950568
Chen, C., Haddad, D., Selsky, J., Hoffman, J. E., Kravitz, R. L., Estrin, D. E., & Sim, I. (2012). Making sense of
mobile health data: An open architecture to improve individual-and population-level health. Journal of Medical
Internet Research, 14(4), e112. doi:10.2196/jmir.2152 PMID:22875563
Coleman, K., Austin, B. T., Brach, C., & Wagner, E. H. (2009). Evidence on the Chronic Care Model in the new
millennium. Health Affairs, 28(1), 75–85. doi:10.1377/hlthaff.28.1.75 PMID:19124857
Edelman, D., & Singer, M. (2015). Competing on customer journeys. HBR.org. Retrieved from https://hbr.org/
product/competing-on-customer-journeys/R1511E-PDF-ENG
Elias, J. (2015). In 2016, Users Will Trust Health Apps More Than Their Doctors. Forbes. Retrieved from http://
www.forbes.com/sites/jenniferelias/2015/12/31/in-2016-users-will-trust-health-apps-more-than-their-doctors/
Grumman, J. (2013). What Patients Want from Mobile Apps. Retrieved from http://www.kevinmd.com/
blog/2013/04/patients-mobile-apps.html
HONcode. (n. d.). Operational definition of the HONcode principles. Retrieved from http://www.hon.ch/
HONcode/Webmasters/Guidelines/guidelines.html
Kumar, S., Nilsen, W. J., Abernethy, A., Atienza, A., Patrick, K., Pavel, M., & Hedeker, D. etal. (2013). Mobile
health technology evaluation: The evidence workshop. American Journal of Preventive Medicine, 45(2), 228–236.
doi:10.1016/j.amepre.2013.03.017 PMID:23867031
McCurdie, T., Taneva, S., Casselman, M., Yeung, M., McDaniel, C., Ho, W., & Cafazzo, J. (2012). consumer
apps: The case for user-centered design. Biomedical Instrumentation & Technology, 46(Suppl. 2), 49–56.
doi:10.2345/0899-8205-46.s2.49 PMID:23039777
McMillan, B., Hickey, E., Patel, M.G., & Mitchell, C. (2015). Quality assessment of a sample of mobile app-
based health behavior change interventions using a tool based on the National Institute of Health and Care
Excellence behavior change guidance. Patient education and counseling. Open mHealth. Retrieved from http://
www.openmhealth.org/
Volume 7 • Issue 2 • April-June 2016
43
Peeples, M. M., Iyer, A. K., & Cohen, J. L. (2013). Integration of a mobile-integrated therapy with
electronic health records: Lessons learned. Journal of Diabetes Science and Technology, 7(3), 602–611.
doi:10.1177/193229681300700304 PMID:23759392
Powell, A. C., Landman, A. B., & Bates, D. W. (2014). In search of a few good apps. Journal of the American
Medical Association, 311(18), 1851–1852. doi:10.1001/jama.2014.2564 PMID:24664278
Pruitt, J., & Grudin, J. (2003, June 5-7). Personas: practice and theory. Proceedings of the conference on Designing
for User Experiences DUX ‘03, San Francisco, CA, USA. doi:10.1145/997078.997089
Research2guidance. (2015). App Developer Economics 2015. Retrieved from http://research2guidance.com/
r2g/r2g--App-Developer-Economics-2015.pdf
Silow-Carroll, S., & Smith, B. (2013). Clinical management apps: creating partnerships between providers and
patients. Commonwealth Fund Issue Brief. SMART Health IT. Retrieved from http://smarthealthit.org/
SMART Health IT. (n. d.). Retrieved from http://smarthealthit.org/
Terry, K. (2015). Prescribing mobile apps: What to consider. Retrieved from http://medicaleconomics.
modernmedicine.com/medical-economics/news/prescribing-mobile-apps-what-consider?page=full
