Abstract
This study details a method for mHealth app development and user experience design (UX) evaluation, which generates a comprehensive list of stakeholder-users, acknowledges UX barriers, advocates multiple methods, and argues that developers should address the UX needs of each stakeholder-user in a complex health-care system. A case study of a research project on an mHealth app for women who are considering prevention of or treatment for osteoporosis assists to elaborate and define the method. To find any measure of success, a fully functional app for older users should be integrated into the entire health-care system.
Keywords
Health and medical organizations have increasingly begun to create digital tools supported by mobile (mHealth) technologies. 1 Researchers have found that mHealth apps can maintain and improve patient health, safety, and quality of life while providing meaningful real-time data to both patients and their health-care providers (Azevedo, de Sousa, Monteiro, & Lima, 2015; Isaković, Sedlar, Volk, & Bešter, 2016). Technical communicators have contributed substantially to this growing area of research through their development and application of user experience design (UX) methodologies and methods to interrogate and improve the quality of mHealth apps. UX emphasizes relationships among stakeholder-users and includes usability testing, which assesses a product’s ability to enable the completion of certain tasks to a user’s satisfaction. However, UX is also a comprehensive concept “where tasks and texts no longer exist individually or in a silo, but instead connect across a broad and complex landscape of interfaces and environments” (Lauer & Brumberger, 2016, p. 249; see also Redish, 2010). Technical communicators, with their expertise in UX, are well suited to investigate how developers and stakeholder-users of mHealth tools participate in this complex system of “human intentions, organizational discourses, biological trajectories, and technological possibilities” (Mara & Hawk, 2009, p. 3).
mHealth apps can create new UX challenges for patients and other stakeholder-users such as health-care providers and hospital administrators. Kopelson (2009) found that patients who use information technology (IT) are more knowledgeable about their conditions and want to participate in health-care decision-making but are viewed by physicians “as difficult [and] demanding” (p. 366). Teston (2016) urged scholars to recognize the precarity of individuals in today’s health-care system and the various ways that ostensibly supportive mHealth technologies can increase marginalization in the most vulnerable populations. Furthermore, mobile information technologies can omit familiar genre markers in order to adapt texts to smaller screens, forcing users of mobile devices to “compose their own coherent connections between fragments of information” (Swarts, 2006, p. 196). Nielsen and Budiu (2013) associated this concern with the transition of information from traditional formats and websites to smartphone interfaces: “Reading through a peephole increases cognitive load and makes it about twice as hard to understand text on a mobile device as on a desktop” (p. 26). As a result of these broad communication impediments, mobility and contexts of use should be considered in app development and UX testing (Melonçon 2017a). Researchers who use a more global context to view a product’s usability can discern relationships among stakeholder-users that can inform UX evaluation: “… different groups of people come to understand a situation differently using different mental routes and assigning different priorities to information elements” (Albers, 2011, p. 121). Technical communicators, by aligning their goals with Williamson and Kowalewski’s (2018) rhetoric of advocacy, can take the lead in navigating nuanced UX testing that encompasses this complex set of material challenges.
In this article, I report on a case study that describes the research to develop and prepare for UX testing an mHealth app for older women who are considering treatment or prevention of osteoporosis. 2 Throughout the case study, I outline the contexts associated with fully implementing the app into health and medical systems, and I make UX evaluation recommendations that can extend to other mHealth apps and contexts. For a completely functional tool, I argue that mHealth app developers and UX researchers should consider the embodied personas of multiple stakeholder-users. This complex system of human and nonhuman actors moves UX testing from a simple one-to-one, inductive process of discovery to a collective, phased process that should acknowledge and resolve numerous challenges to gain the acceptance of many stakeholder-users (Mara & Hawk, 2009; Melonçon, 2013). 3
This study is informed by performative phenomenology (Melonçon, 2018), which can assist in defining and exploring the roles of several stakeholder-users in the health-care system and identifying UX contexts to ensure the embodiment of each stakeholder-user is addressed through the research and testing process. Embodiment is an important dimension of UX for older adults with their decreasing dexterity and increasing reliance on younger caregivers for health care and technological access. However, it is equally important to understand how mobility and environmental contexts affect other mHealth stakeholder-users such as health-care workers and IT administrators. As a concept that uses the force of performance to elicit the embodied narrative of the participants in usability testing and UX evaluation, “performative phenomenology captures or records embodied experience in a way that accounts for past and present associations between participants and the multicontextual, multidimensional nature of health care” (Melonçon, 2018, p. 97). As an extension of patient experience design (Melonçon, 2017b), performative phenomenology enables UX researchers to examine and describe the complex position of each stakeholder-user. In the multidimensional world of an mHealth app, each stakeholder-user brings a set of criteria and interprets functions in different ways according to their needs. Recording participants’ embodied knowledge by using multiple research methods before and during development and UX evaluation foregrounds the intentionality of stakeholder-user performance. With the goal of considering the contexts of people in several intertwined parts of a health-care system, I address the following research question: How should technical communicators identify and address the needs of all stakeholder-users in the development and UX evaluation of an mHealth app for older persons?
Methodologies and Methods That Inform mHealth Communication
Scholars have modified and developed methodologies and methods for UX evaluation of health and medical technology beyond the traditional vacuum of the laboratory to address the interconnected contexts and embodied needs of users of wearable medical devices (e.g., Arduser, 2018; Jones, Gouge, & Crilley, 2017; Kennedy, 2018; Kessler, 2016) and a wearable fitness tracker (Welhausen, 2018). Implicit in these studies is the concept of the user as a contributor and practitioner in a complex design process (Dilger, 2006; Howard, 2008; Johnson, Salvo, & Zoetewey, 2007; Simmons & Zoetewey, 2012). Concepts of complex systems and multiplicities pervade research in which scholars consider usability testing and UX for health-care technologies. Researchers and theorists in composition studies and the rhetoric of health and medicine have developed research methods based on Guattari and Deleuze’s (2000) assemblage thinking (e.g., Angeli, 2018; Fox, 2002; Rice, 2008) and Latour’s (2005) actor-network theory (ANT; e.g., Kelly & Maddalena, 2016). 4 Several authors have also used Rivers and Söderlund’s (2016) concept of speculative usability, which applied ANT to usability testing and argued that researchers should consider and address all possible human and nonhuman contributors (Arduser, 2018; Cannon, Walkup, & Rea, 2016; Gouge, 2017). Others have adopted Edbauer’s (2005) concept of rhetorical ecologies to frame research on complex systems in health care (e.g., Ehrenfield, 2018; Jensen, 2015; Scott, 2006; Swarts, 2006).
As advocates of patient-centered communication, these scholars have emphasized the UX needs of patients who interact with health and medical information in both traditional and digital formats. In this research, patients and technologies exist in complex discursive contexts that challenge traditional UX testing methods and methodologies, and researchers describe and call for new methodologies to address the needs of user contexts. Melonçon (2017b) argued, “Current scholarship that insists on larger and more complex contexts suggests that everything in an ecology (or network, etc.) reverberates equally from everything else. But it does not” (p. 22). Instead, patients can be disempowered by systems in which health-care providers, hospital administrators, and health insurance organizations may make decisions with little or no patient input, leading to further patient marginalization that is exponentially more difficult to overcome.
A necessary component to increasing patient participation in health-care decision-making is increased communication among all stakeholder-users. Condit (2006) observed that people never exist outside relationships, and that all relationships are constructed through communication. The task of technical communicator as researcher and UX evaluator then becomes one of considering the patient within a complex system of relationships. Health-care systems stakeholder-users have relationships with one another, many times informed by earlier structures, created by earlier stakeholder-user relationships, and often obscured by material structures and other spatial–temporal degrees of separation (Ehrenfield, 2018; Mol, 2002). By narrowing the context and focusing on single cases in systems, Melonçon (2017a) argued, researchers can begin to build a body of qualitative research methods and methodologies that move toward a sensitive treatment of patients. As I discuss in the following case study, each stakeholder-user has disparate UX needs that should be met using appropriate research methods. This argument does not discount the aims of patient-centered communication. Instead, acknowledging the need for multiple UX evaluations involving multiple stakeholder-users extends the idea of patient experience design to support the many individuals in a health-care system.
Case Study: mHealth App for Osteoporosis Diagnosis and Treatment Information
Production of the mHealth app in this study was driven by a largely unmet need to provide decision-making information for the treatment and prevention of osteoporosis in women 50 years of age and older (Project Hepius, 2018). Approximately half of all women in the United States will break a bone because of osteoporosis (National Osteoporosis Foundation, 2018). By 2025, treatments for the results of osteoporosis will cost $23.5 billion per year (National Osteoporosis Foundation, 2018). In addition, older Americans are using digital technology at increasing rates, including 42% who own smartphones (Anderson & Perrin, 2017). People aged 65 years or older make up 15% of the U.S. population (Anderson & Perrin, 2017). This number is predicted to rise to 22% by 2050 (U.S. Census Bureau, 2016).
Case studies have proved helpful in understanding complex systems and embodied UX during development and evaluation of wearable health-care devices (Angeli & Norwood, 2017; Arduser, 2018; Kennedy, 2018). In particular, case studies in this field can allow retrospective inquiry to occur (Angeli & Norwood, 2017) by “enabl[ing] scholars to explicate a rhetorical orientation” (Moriarty et al., 2019, p. 126). For example, Arduser (2018) used semistructured interviews to provide a case study of participants in an online diabetes support community who created do-it-yourself (DIY) monitoring systems and gathered the data to support the entire group. Kennedy (2018) offered an autoethnographic case study of technological embodiment using close textual reading of traditional and online documents, interviews, and her own experience to provide a multimethodological approach to describe the integration of a hearing aid with a smartphone app. Kennedy (2018) observed: “Valuing rigorous reports of lived experiences also facilitates more fully rounded, more complex, and more ethical social science research of the sort that is frequently undertaken in user experience design” (p. 42).
Research Contexts
Development of the app was guided by agile methodology, which involved a team working in an ethical, client-oriented rapid development cycle and producing necessary iterations of functioning prototypes that move toward a final product (Shore & Warden, 2008). Researchers used the method established by Harrington, Ruzic, and Sanford (2017), who elaborated the work of Ruzic, Lee, Liu, and Sanford (2016) to create two usability categories informed by Universal Design criteria adapted to emphasize the needs of older users: a heuristic for app design and a usability assessment tool. Patient-users may elect to use information from the National Institutes of Health(2016), the results of a decisional self-efficacy survey (O’Connor, 2002), and a fracture risk calculator from American Bone Health(2018) to assist in preparing them to participate in shared decision-making with their physicians. The goal of the app is to integrate it into patients’ electronic health record (EHR) at a large university hospital so that all primary care providers can use the tool to increase treatment and prevention outcomes for patients. Consequently, researchers also developed a back-end program to allow health-care providers to add, delete, and edit the information in the app. Patient survey results, fracture risk calculation results, and inputs for health-care providers such as treatment and progress notes have been coded for insertion into the EHR.
To guide early development, I used semistructured interviews to survey three groups of stakeholder-users as part of the National Science Foundation’s Innovation Corps (Innovation Corps, n.d.) customer discovery exercise. The sample included 10 women aged 50 or older, 3 primary care physicians, and 2 health insurance industry representatives. Participants all agreed that they would like to use the app—or that they would support integration of the app into their health-care systems. However, each group expressed concerns related to UX including usability, deployment, integration, and management barriers. Moreover, the participants also identified three additional stakeholder-user groups that they considered important to the successful integration of the app: caregivers, hospital system IT providers, and hospital administrators. Consequently, this case study focuses on six stakeholder/user groups comprising a complex set of relationships that should be addressed by mHealth developers and UX researchers (see Figure 1).

Stakeholder-user context for mHealth UX evaluation.
Embodied Personas in mHealth App Development and UX Evaluation for Older Adults
With the interview data from stakeholder-users in three of the six stakeholder-user groups, I began to gather information that will be used to create personas to inform future development and UX evaluation of the app. Personas are descriptions of fictional users gathered from initial research and include information such as demographic information, contexts and goals of product use, and technical expertise (Barnum, 2011). Personas have been widely applied for the past 20 years in several fields including human–computer interaction, public policy, and marketing (Salminen, Jansen, An, Kwak, & Jung, 2018). Scholars in technical communication have begun to elaborate and refine practices in persona-building to think about complex user needs. Getto and St. Amant (2015) addressed the growing need for developers and researchers to consider intercultural communication by outlining a four-part framework to determine user needs and create more nuanced personas: local and technological, local and cultural, global and cultural, and global and technological. Researchers and developers have often used personas to create ideal users and have ignored the importance of embodiment and mobility. In response, “technical communicators … need to ensure that they understand that the embodied users are complicated, affective beings with a range of emotions” and that mobility should include “both physical movements as well as movements enhanced by technologies and a combination of both” (Melonçon, 2017a, pp. 55–57).
In the next sections, I use participant interviews and recent literature to outline the criteria necessary to develop embodied personas for mHealth app development and UX evaluation for older adults in a complex health-care system. Throughout, I describe research and evaluation constraints and opportunities that developers and UX researchers should consider in order to move mHealth tools for older users from prototype to fully scalable products that successfully address the needs of all stakeholder-users. This diachronic and synchronic strategy includes several phases and research methods to inform and substantiate not only the patient usability of the app but also its efficacy within a health-care system.
Older adults as patients
Respondents listed several challenges to using mHealth apps, including cognition (see also Becker et al., 2013), physical movement (see also Page, 2014), motivation (see also Ellis & Kurniawan, 2000), and visual perception (see also Hassan & Mathiassen, 2017). In addition, interviewees noted that the mobility of smartphones allows the use of the app away from clinical settings (e.g., at home) and increases the time available to use and comprehend the medical information. The use of the app in other settings also supports the inclusion of caregivers who may not be able to attend the clinical encounter but still participate in both use of the app and decision-making about treatment and prevention strategies.
Caregivers
A pervasive theme in interviews with older women was their desire for technical assistance and decision-making support from caregivers. Caregivers are typically untrained in health and medical issues, are younger relatives or close friends, and live either with the patient or in proximity to the patient (Van Houtven, 2015); however, “their role in health information technology is largely undefined and poorly understood” (Wolff, Darer, & Larsen, 2016, p. 117). Participants often commented that they relied on younger family members to navigate downloading and initial registration of apps. Respondents also depended on family and friends to consider health-care options, but they noted that sharing information was challenging because of privacy stipulations of EHR. Older adults overwhelmingly want to share health and medical information with their caregivers (Zulman et al., 2011). However, “health systems are understandably reluctant to create potential privacy breaches” (Sarkar & Bates, 2014, p. 358). Because access to patients’ mHealth apps can enhance the abilities of caregivers to more actively participate in the support process for older adults, app developers and UX researchers should consider ways to include caregivers as stakeholder-users.
Health-care providers
Physician respondents argued that they needed formal data to prove that an mHealth app would both be helpful in their practices (i.e., save time and money) and increase patient outcomes. To fully gain the support of physicians, developers will not only need to create UX evaluations for physicians and other health-care workers, but they will also need to work with physicians to test the app in a clinical trial that would prove the app’s efficacy. Most reviews of mHealth apps by and for physicians have been based on anecdotal evidence and personal experience (Powell, Landman, & Bates, 2014), which, while helpful, do not meet the standard of a clinical trial. Furthermore, physician respondents confirmed that they would be willing to use them more widely if apps were integrated into EHR (see also Slabodkin, 2013).
IT providers
Physician respondents also noted that developers will need to work with IT coordinators in the system to integrate the app into a system that must adhere to strict regulatory measures, including Health Insurance Portability and Accountability Act mandates for patient privacy and access (see also Istepanian & Woodward, 2017). In addition, the physicians expressed concern about integrating mHealth apps into EHR in their present system and suggested that all stakeholder-users, including IT workers, have a steep learning curve to overcome. One physician gave the example of a year-and-a half time frame to install a relatively simple mHealth app. However, this duration may have been complicated by the transition to a new EHR system that replaced a legacy paper and electronic version in that hospital. In addition, one respondent suggested that the data on individual smartphones could still be accessible to anyone with the proper ID and password (see also Wood et al., 2019). Thus, data continue to be at risk of physical theft or a lack of user oversight by older adults. This challenge further complicates efforts to give access to caregivers.
Health insurance providers
Only after the integration of the app into EHR would it be possible for developers to perform usability testing and UX evaluations on sites supported by insurance providers such as patient portals. Responses from health insurance representatives were similar to the responses from physicians. They were generally supportive of the idea of mHealth apps and gave some examples of apps that were, or had been, available through patients’ digital portals. UX methods for health insurance providers could entail usability testing with their IT staffs. However, interviewees also observed that their organization would support mHealth apps only if the apps proved to be net beneficial in reducing costs, improving patient outcomes, and satisfying patients as digital consumers (see also Malvey & Slovensky, 2014). These data would come from the clinical trial and UX evaluations of health-care providers.
Hospital administrators
After addressing each of the research opportunities for the above stakeholder-users, the last move will be to take the entire set of data to hospital administrators to inform them that patients, caregivers, health-care providers, health insurance providers, and IT coordinators find the app useful in improving patient outcomes and improving health-care provider efficiency. A recent survey by the Health care Division of the American Society for Quality found that 70% of hospital administrators saw “improved workflow efficiency” with “digital health tools” as vital for increased stakeholder satisfaction (Wicklund, 2017, para. 6). However, the administrators surveyed also feared “resistance to change from staff and physicians unwilling to learn new skills or fearing that the technology will impede their workflow” (Wicklund, 2017, para. 7). Developers and UX evaluators can be prepared for this final step by providing to administrators the appropriate data in the voices of the stakeholder-users that will ensure effective integration of the app.
A Heuristic for Systemic mHealth Development and Usability Testing for Older Adults
App developers and UX researchers should address the embodied, structural, and attitudinal barriers that each group of stakeholder-users face and evaluate the mHealth app using mixed methods, which include interviews, usability testing, and clinical trials (see Table 1). Triangulation of data will be especially important in proving the efficacy and efficiency of the app to health insurance providers and hospital administrators.
A Heuristic for Systemic mHealth Development and UX Evaluation for Older Adults.
Note. EHR = electronic health record; HIPAA = Health Insurance Portability and Accountability Act; IT = information technology.
Perform stakeholder-user interviews. Interviews during research and development phases are especially important. Interviews not only provide data for app development, but they also can identify other stakeholder-users and inform the creation of embodied personas based on interviews and focus groups.
Establish criteria for successful UX evaluation. While some stakeholder-users might benefit from traditional usability testing scenarios (e.g., physicians, patients, caregivers), other people in the app system could contribute to (and benefit from) surveys and interviews. For physicians, researchers will also need to perform a clinical trial of the fully functional app. For hospital administrators, researchers will need to gather the entire corpus of data from all stakeholder-users in a proposal to request full integration in the EHR.
Use multiple methods for app development and UX evaluation. Each stakeholder-user group demands careful consideration of appropriate methods that support the needs of the group but also that support the needs of the system that is populated by the groups. Developers and researchers cannot reach this goal with a single method. From the first step in developing the app to the last step in addressing the UX of hospital administrators, the case study proved that researchers would need a mixed-methods approach to identify the UX needs of a disparate set of stakeholder-users.
Table 1 details the range of stakeholder-users, their embodied UX needs and goals, research questions, and possible methods for answering those questions.
Conclusion
The import of thinking through a method for the development and UX evaluation of an mHealth app at the system level is underscored by Thies, Anderson, and Cramer (2017), who stopped a clinical trial of an app for diabetes management because of several UX issues: There was a poor fit between the app, end users, and recruitment and treatment approaches in our setting. Usability testing might have revealed this prior to launch but was not an option. There was not sufficient time during routine care for clinical staff to familiarize patients with the app or to check clinical data and messages, which are unreimbursed activities. Some patients did not use the app appropriately. The lack of integration with the electronic health record (EHR) was cited as a problem for both patients and staff who also said the app was just one more thing to attend to. (p. 1)
The systemic development and UX evaluation of an mHealth app for older adults developed in the case study emphasize the need for technical communication researchers to consider several stakeholder-users and produce replicable, valid data. The health-care system creates “an interconnected web” that must integrate “contextual awareness” as “an essential element of the test plan” (Albers, 2011, p. 121). Patients, physicians, and health insurance representatives in the case study emphasized the need to consider the embodied contexts of the stakeholder-users as they navigate the socio-political and economic features of a complex health-care system. Understanding these needs through Melonçon’s (2018) concept of performative phenomenology can assist technical communicators to develop and evaluate the UX of all participants using multiple methods “for the study and practice of health care and the experience of health, illness, pain, and disability” (p. 105). In this case study of an mHealth app for older adults, performative phenomenology can assist in defining the relationships and contextualizing the needs not only of the patient but also of the caregivers, health-care providers, IT workers, health insurance representatives, and hospital administrators involved in the process. Through this process of discovery, researchers may also “expose and even shift some power relationships by giving voice to patients or research participants in different and more forceful ways …” (Melonçon, 2018, p. 105). Thus, this project has also sought to broaden and deepen the concept of performative phenomenology by increasing the rhetorical agency of stakeholder-users through an increased understanding of their needs within the complex system of health care.
While this article is limited in its focus on the systemic needs to integrate one mHealth app for older adults into a health-care system, it also serves as a touchstone for development of health-care apps for other populations, which have their own affordances and constraints. Furthermore, much scholarly work remains to be accomplished to increase understanding and application of methodologies and methods necessary in usability testing and UX studies for each of the health-care stakeholder-users introduced here. Hundreds of thousands of mHealth apps are available in the marketplace. Yet few have been evaluated for UX, and fewer still have been successfully integrated into health-care systems.
To be successful, the development and UX evaluation of an mHealth app for older users should not only satisfy the embodied needs of patients in multiple spatial-temporal and discursive contexts, it should also satisfy the needs of the people and organizations who support the patient. Researchers should consider multiple methods that can measure the UX of an mHealth app in several ways among all stakeholder-users to define successful criteria for implementation. These goals can only be accomplished by acknowledging and defining the intricate relationships among the stakeholder-users in a health and medical system. Developers and advocates of mHealth tools should appeal to prospective users by providing qualitative and quantitative evidence not only that their products are usable but also effective. UX studies can act as the force to investigate and explicate the embodied needs and purposes of all stakeholder-users of an mHealth app in providing the patient an exceptional health-care experience.
Acknowledgments
The author would like to thank Hyuntae Na, PhD, William Curry, MD, Collin Swartz, Ronak Patel, Kishan Patel, John Hostetter, Aaron Wasserman, Austin Duncan, Daham Eom, Isaac Khai Ren Chua, and Muhannad Alshihabi for their contributions to this project.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported with funding from the Penn State Harrisburg Office of Research and Outreach.
