Introduction to Topical Issue on mHealth for Brain Injury Rehabilitation.

Abstract

THE POTENTIAL for mobile technology to change the very nature of healthcare delivery has been noted by the World Health Organization and is reflected in the rapidly growing body of literature on use of mobile health (mHealth) across disciplines and clinical populations, including traumatic brain injury (TBI). Despite its potential, challenges remain for identifying specific evidence-based mHealth technology suitable for persons with TBI. A notable challenge is the rapid pace at which technology develops. By the time a multiyear clinical trial concludes, the technology tested may already be obsolete. Recently published literature reviews of mHealth-based interventions and assessments after TBI reveal a wide variety of outcomes that can be addressed and a similar diversity in platforms and technology used, though most studies included in these reviews describe development, usability, or feasibility of the technology rather than efficacy1–3 (a trend we continue to see in the articles included in this topical issue). Novel technological solutions must be designed with the unique needs of persons with TBI in mind, with attention to both usability and barriers to access. Although there may not yet be evidence to recommend any specific system or device, the existing body of literature does reveal common themes across mHealth research in TBI that provide clinicians with important information to support evidence-based practice in both clinical settings and the community. The purpose of this topical issue was to highlight the current work to advance the evidence for using mHealth technology, specifically in the community, to support individuals with TBI. The 6 articles included herein reflect both that TBI affects individuals across the life span and the myriad ways that various mHealth technologies can be potentially beneficial after TBI. These articles represent diverse technological solutions—including smartphone apps (all articles), wearable sensor technology,4,5 and conversational agents6—to address a variety of clinical applications, such as symptom monitoring,5,7,8 stress and symptom management,4,7,8 and behavior change.6,9 The studies in this issue echo many of the themes identified in recent reviews. Five broader themes across these 6 studies are summarized as follows: Much of the current work in this area aims to establish usability and feasibility, which are appropriate for the state of the science. As noted in previous reviews, mHealth technology development far outpaces the traditional process for establishing scientific evidence. We need to rethink our approach to account for the temporal disconnect between technology development and the scientific process. It may be that the best evidence comes from simply ensuring that the content, or therapeutic ingredients, of mHealth applications is evidence-based (eg, based on sound theory, including content and processes from established behavioral health interventions) and that the technology is usable and acceptable to individuals with TBI and their family members. We can also use more expedient experimental designs for evaluating usability and efficacy of mHealth interventions, such as the single case experiments Rabinowitz et al6 used to investigate RehaBot, an interactive virtual coach (“chatbot”) to promote meaningful activities after TBI. The articles in this topical issue present a variety of mHealth applications for users with TBI that—while feasible and acceptable—may require specialized training, customization, and ongoing support, especially for persons with cognitive impairments following injury. We see in the examples here and elsewhere in the literature that technology and the human-technology interface should be simple to understand and navigate, flexible and customizable, with built-in support for the use of compensatory cognitive strategies. Individuals with TBI want (and often need) reminders sent through push notifications to address limitations in prospective memory, initiation, planning, and follow-through, as noted specifically by participants in the study by Schmidt et al.8 Flexibility and customization, including real-time feedback about symptom levels or symptoms changes and targeted content, are highly desired and make mHealth apps more meaningful. This theme is highlighted in the study by Schmidt et al,8 who report on unanticipated patterns of usage in a gamified mHealth intervention to prompt change in activity level based on postconcussive symptoms in adolescents. Their findings suggest that more sophisticated algorithms are needed that account for both symptom levels and changes in symptoms when making recommendations for changing behaviors. Perhaps, most often noted in both prior studies and those included herein (specifically in Nabasny et al,4 Schmidt et al,8 and Juengst et al9), there is an essential need for more intensive up-front training and support. The study by Juengst et al9 discusses this specifically in the context of substantial differences in use of mHealth apps between groups that underwent behavioral health interventions before versus during the COVID-19 pandemic. Those in the prepandemic group had substantial initial and ongoing support from their health coaches, whereas those in the group during the pandemic did not. Some articles highlight examples of mHealth solutions that have been designed to pair with traditional clinician-delivered interventions in a hybrid model.5,6,9 The mHealth apps in the study by Juengst et al9 were designed to complement and promote ongoing engagement in 2 intensive healthy lifestyle interventions. RehaBot6 is designed to pair with a clinician-delivered behavioral activation treatment of depression providing support and reminders for completing planned activities between sessions with a therapist. Wallace and colleagues5 examined the effects of an mHealth smartwatch/smartphone app system, designed to pair with a conventional clinician-delivered intervention, to prompt service members with mild TBI and posttraumatic stress disorder to practice diaphragmatic breathing with the aid of physiological sensors to guide proper technique. These examples demonstrate how mHealth can extend the reach of clinic-based treatments and provide preliminary evidence that these solutions may promote self-monitoring and practice of therapeutic skills in individuals' everyday environments. This topical issue contains multiple illustrations of the applicability of mobile technology that can be applied to enhance the in situ assessment of physical and neurobehavioral symptoms.4,7,8 For example, Liu and colleagues7 tested an mHealth headache diary for remotely monitoring posttraumatic headaches. The authors were able to evaluate headache triggers, locations, and associated symptoms that may provide valuable clinical information for designing and/or tailoring interventions. Nabasny et al4 examined the feasibility of pairing a commercially available chest strap and mHealth app that captured heart rate variability with self-reported neurobehavioral symptoms over a 2-week period, allowing them to assess whether heart rate variability and neurobehavioral symptoms covaried (ie, whether heart rate variability could be an objective physiological indicator of symptoms that could enhance self-report).4 The rich data that can be collected from ecological momentary assessment (ie, real-time, high-frequency, in situ self-reported symptoms or behaviors) and wearable sensors provide new opportunities for both research and clinical care. Although many technologies described in this issue are promising, adherence to mHealth interventions presents a challenge. Individuals with TBI will only be able to benefit from mHealth interventions to the extent that they engage with them. For example, Schmidt et al8 found individual characteristics (ie, socioeconomic status and the presence of comorbid conditions) were predictive of how often participants engaged with a symptom monitoring app. Liu and colleagues7 also note that identifying patient characteristics associated with app compliance is an important area for future research. Consistent with findings from studies in previous reviews,1,2 many participants never start using mHealth technology (even after consenting to do so), as noted in the studies herein by Juengst et al9 and Nabasny et al.4 More research is needed to characterize and address the factors that limit engagement with mHealth tools to establish how to best serve the diverse population of persons with TBI. Schmidt et al8 again highlight the need for more initial training to promote adherence, including more education about how the mHealth technology will be of optimal benefit to individuals with TBI and their family members. Their study was the only one to include a gamified app to promote adherence; gamification is a well-established approach to maximizing engagement with mHealth apps.10,11 Schmidt et al8 further suggest that avatars to guide and support users while engaging with mHealth technology represent a next step—avatars and technology-enabled interactive agents such as RehaBot (Rabinowitz et al6), for example. Perhaps, most important is that every app, device, or approach is not appropriate for everyone (and that is ok!) and that flexibility and customization are needed not only in the design and content of mHealth technology but also in how—and to whom—it is clinically applied. The articles in this topical issue offer some of the latest advances in the rapidly emerging field of mHealth as applied to brain injury rehabilitation. New technological developments such as cloud computing and advances in machine learning and artificial intelligence are currently paving the way for new possibilities. Exciting as this may be, we caution moving forward without due consideration of the ethical implications of this line of research. First, we must consider what we are tracking or providing to individuals with cognitive disabilities in the absence (or synchronous presence) of a human clinician. Second, we must carefully determine what mHealth technology can replace in current practice versus how it would be best used to complement or supplement current practice. Articles in this issue5,6,9 illustrate models for incorporating mHealth with clinician-delivered interventions. mHealth developers and researchers must consider which therapeutic ingredients are appropriate for mHealth delivery on their own and which should be delivered with the supervision and guidance of a skilled clinician. Third, high technology such as artificial intelligence may not be feasible for low-resource regions or individuals without access to high-speed internet and up-to-date smartphones. Advances in mHealth technology should be made to diminish rather than widen health disparities. As this issue highlights, there is much potential for mHealth technology to improve the effectiveness and reach of brain injury rehabilitation. As this field continues to develop, we must ensure it is done with the same consideration afforded to traditional clinician-delivered care. Amanda R. Rabinowitz, PhD Shannon B. Juengst, PhD, CRC Issue Editors