In the 21st century, anxiety and depression have been among the top 25 causes of global disease burden [1]. The COVID-19 pandemic has only intensified the rising prevalence as well as the personal and societal impacts of these disorders [2,3]. As there are simply not enough mental health professionals to meet this need [4,5], alternative interventions—both preventive and curative—are urgently needed. Targeting physical activity is a clear opportunity. Before the pandemic, more than 1 in 4 adults reported sitting for >8 hours per day [6], and this number rose to >40% during the pandemic [7]. Both of these statistics are likely underestimated [8]. Prolonged sedentary behavior, or extended time spent awake with minimal energy expenditure [9], is associated with more severe anxiety and depression as well as higher odds of developing related disorders [10-14]. In contrast, decades of research have demonstrated that regular physical activity is associated with numerous positive psychological outcomes [15-17]. Cross-sectionally, individuals who engage in regular exercise—a subset of physical activity that involves planned, structured, and repetitive bodily movement intended to improve or maintain fitness [18]—report fewer and less severe symptoms of anxiety and depression [19,20], greater positive affect and well-being [21], less stress [22], and lower rates of anxiety and depressive disorder diagnoses [23-25]. At the individual level, people report feeling better on days when they exercise [26-28].

Prospective data support regular exercise as a potent population-level prevention tool, significantly lowering the risk of developing anxiety and depressive disorders [25,29,30]. Encouragingly, even small amounts of physical activity may have an enormous impact on mental health [31,32]. For example, an estimated 12% of new cases of depression could be prevented if the entire population exercised for just 1 hour per week [33]. In addition, among individuals presenting with diagnosable levels of symptomatology, systematically increasing exercise behavior is therapeutic [33-36].

Exercise appears to enhance emotional flexibility, or a person’s ability to self-regulate under stress [37-40]. Physiologically, individuals who exercise more regularly show faster heart rate recovery following stressors than their peers who exercise less, and individual bouts of exercise can mitigate the hypothalamic-pituitary-adrenal axis, heart rate, and blood pressure reactivity to acute stress [39,41-43]. Physical activity can also increase the production of brain-derived neurotrophic factors, which are neurobiological changes that are understood to increase resilience [44]. These effects are mirrored in reports of exercise bolstering emotional recovery following stressors, enhancing coping self-efficacy, and mitigating the impact of rumination and other emotion regulation deficits on prolonging distress [34,39,42,45,46]. Furthermore, exercise benefits physical health targets that share bidirectional relationships with mental health, such as better sleep and cardiometabolic health [47-51]. Critically, positive treatment effects have been found for directly alleviating anxiety [52-55] and depressive [56-58] disorders as well as related and frequently comorbid conditions such as posttraumatic stress disorder [59] and obsessive-compulsive disorder [60,61]. Importantly, research has replicated the benefits of physical activity (ie, reducing psychiatric symptoms) in samples of people with severe mental illness, such as schizophrenia [62]. Similarly, physical activity and other health-related behaviors (eg, sleep hygiene) are considered to be an integral component of treatment for bipolar disorder [63]. Exercise has also been successfully used to augment the effects of other validated psychosocial treatments such as cognitive behavioral therapy (CBT) [64-67].

Despite the broad knowledge that regular physical activity is physically, cognitively, and emotionally beneficial, only approximately half of Americans meet even the minimum exercise recommendation of 150 minutes per week of moderate-intensity or equivalent physical activity [68]. Individuals with anxiety, depression, or related conditions are even less likely to do so [69,70]. They are also more likely than peers without mental health disorders to exhibit elevated sedentary behavior [71,72]. Thus, although acceptable and efficacious tools exist to help individuals meaningfully change their behavior and improve psychiatric symptoms, there is a large gap between the evidence and real-world implementation. Few clinicians include physical activity as an explicit treatment target or use it as an intervention tool [73-75]. Moreover, the larger barriers to treatment within our health care system remain, including the inaccessibility of treatment due to the acute shortage of qualified clinicians; stigma; and patients’ difficulty with travel, timing, and the cost of regular appointments [76].

With the advent of mobile sensors and phones, experts have quickly noted the ability of technology to expand the reach of evidence-based psychiatric care; overcome the aforementioned barriers by providing treatment flexibly; and begin reducing long-standing disparities in treatment access, response, and dropout [77-79]. This could also be an efficient, scalable method of promoting increased physical activity among adults with or at risk of anxiety and depressive disorders.

Digital solutions show strong early benefits for activity measurement and intervention in nonpsychiatric populations. In fact, leveraging technology to measure and increase physical activity was an official recommendation from the National Heart, Lung, and Blood Institute and National Institute on Aging “Influences on sedentary behavior/Interventions to reduce sedentary behavior” joint workshop [80]. First, measuring behavior via mobile sensor is validated and widely used [81-84]. Similarly, people’s tendency to carry their phones with them throughout the day allows for more accurate monitoring of physical activity and related progress. The ease of use and, therefore, precision of such technologies (wearable and smartphone-based sensors) is an important boon for research and treatment as self-report measures of activity typically result in underreporting [81-84].

Second, delivering treatment in part or fully via mobile phones is effective in increasing physical activity in nonpsychiatric populations [85-87]. This parallels broader findings that apps can effectively promote other health behaviors (eg, improved nutrition, smoking cessation, and medication adherence) [88,89]. Currently, 97% of Americans have a mobile phone, and an estimated 85% have a smartphone [90]. Although these numbers are lower in certain populations, such as those with serious mental illness (wherein an estimated 85% own a mobile phone and 60%-70% own a smartphone), the ubiquity of smartphones allows for the promotion of behavior change in real time and with a wider array of individuals [91-94]. Inactivity frequently occurs out of conscious awareness or choice due to people’s attention being fixated elsewhere (eg, watching television or taking the elevator at work). As such, personal devices can unobtrusively enhance awareness of behavior, which itself can promote increased activity [95]. Furthermore, technologies can deliver notifications in the moment to interrupt passive episodes while also providing tools to increase activity when individuals are most likely to take action [96]. Mobile app–based physical activity interventions can also gamify exercise to enhance enjoyment, which is a key mechanism for long-term engagement in physical activity [97,98]. In-the-moment enjoyment not only promotes regular exercise but is also in itself beneficial for mental health, contributing to the success of broader interventions such as behavioral activation. Overall, digital interventions are promising as they are low risk (ie, typically focus on reducing sitting and increasing light activity), can be deployed without a clinician, and can be used in the context of a patient’s daily life. Ultimately, research conducted thus far in the general population supports mobile technologies as valid, accessible, and effective methods of promoting physical activity and reducing sedentary behavior.

Although it is reasonable to extrapolate that physical activity interventions could be implemented via smartphone in a similarly feasible, acceptable, and effective manner in psychiatric populations or for psychiatric targets, this remains an open question. High-quality trials of in-person exercise programs for mental health often unintentionally include components beyond the activity itself that are potentially therapeutic, such as regular, structured, and supervised sessions [65]. In other words, as part of most exercise programs, participants also receive regular social engagement or support, face-to-face time with a professional, instruction and demonstration of target behaviors, and guidance with behavioral scheduling or activation, all of which may enhance the therapeutic benefits of physical activity. However, the remote and asynchronous nature of technology-based interventions may provide a different experience from that of in-person programs, and thus, the impact may also differ. On the other hand, the aforementioned benefits of digital interventions, such as their ability to increase accessibility, lower logistical barriers to engagement, and enhance self-awareness while also promoting behavior change in real time, may boost response and, thus, lead to comparable—or even stronger—effects than face-to-face trials. As a result, it cannot be assumed that face-to-face physical activity programs or digital programs designed for other populations (eg, medical) will translate when delivered via smartphone or to a new population.

In this study, we explored the potential of physical activity interventions, as delivered (at least in part) via smartphone, to improve mental health in psychiatric populations. As this topic remains relatively new, we also considered available evidence on these tools to address mental health symptoms in nonpsychiatric populations. Specifically, we highlighted in which populations these tools have been tested; what outcomes have been evaluated (eg, acceptability, behavior change, and symptom change); and how strategies and tools have (or have not) been tailored to individuals with depression, anxiety, or related concerns. The goal was to provide a narrative overview and specific recommendations for future research on smartphone-based physical activity interventions for psychological disorders or concerns.

To provide a narrative overview of this emerging research area, we searched for articles that (1) described or reported an intervention intended to increase physical activity or reduce sedentary behavior; (2) included an emotional disorder, concern, or symptom as an outcome measure; (3) described or reported an intervention delivered entirely or in part via a smartphone app; and (4) were published in English and in peer-reviewed journals. Example search terms include “smartphone,” “smartphone application,” “mobile application,” “mobile app,” “digital mental health,” “app-based,” “app-assisted,” “mobile phone,” “ehealth,” “digital,” “mobile,” “exercise,” “physical activity,” “sedentary,” “sedentary behavior,” “physical inactivity,” “depression,” “dysthymia,” “mood,” “anxiety,” “phobia,” “trauma,” “posttraumatic stress,” “obsessive compulsive disorder,” “post-traumatic stress,” “obsessive-compulsive disorder,” “stress,” “emotional disorder,” “emotional problem,” “well-being,” “wellness,” “affective disorder,” “OCD,” “PTSD,” “MDD,” “GAD,” “mental health,” and “mental illness.” Web-based database (PubMed, Google Scholar, and Cochrane) searches and additional manual searches (eg, searching the reference sections of articles identified through database searches) were conducted up to March 2022. Records were initially reviewed by one coauthor; in cases of uncertainty about appropriateness for this review, records were reviewed in full by 2 additional coauthors and discussed until a consensus was reached.

We attempted to extract the following information, where available, from each paper: sample size, inclusion criteria, demographics of the sample, primary aim, trial design (eg, randomized controlled trial), treatment duration, technology used, other interventions used (ie, in addition to physical activity), outcome measures (eg, primary and secondary measures of physical activity), results, treatment components or behavior change strategies, adaptations for mental health, and inclusion of coaching.

In taking a narrative lens, we drew attention to the type of work that has been done and used these exemplars to discuss key directions to build on. Table 1 provides a summary of the included articles.

Overall, a review of the literature revealed that little work has been done to test the impact of smartphone-based physical activity interventions on increasing physical activity or reducing mental health symptoms in psychiatric or at-risk populations. Most related trials with clinical populations have been conducted in the area of medicine, with studies investigating the effects of physical activity—encouraged through smartphone- and wearable-based interventions—on physical health conditions (eg, diabetes [99,116], obesity [86,108], cancer [111,118], cardiovascular issues [100,110], and multiple sclerosis [114]). Physical activity is well established as a means of facilitating rehabilitation following serious illness or injury, as well as mitigating the progression of chronic health conditions [119-121]. For example, adults with Parkinson disease who used a minimally supported, customizable home-based exercise app for 8 weeks doubled their amount of weekly exercise (minutes) while also increasing the intensity of such exercise [106]. Similarly, engagement with smartphone-based physical activity interventions led to increased strenuous exercise among adults with cancer [111] as well as increased step count for those with cardiac issues [100,110] and for youths with multiple sclerosis [114]. These changes are noteworthy as medical illness or disease can serve as a barrier to engaging in health-promoting behaviors [122] despite the knowledge that such behaviors can stabilize or even improve such medical conditions [123].

In contrast, the practice of formally integrating exercise into mental health care is relatively new and not established in current standards of care. This is mirrored by the disproportion of extant research examining digital tools for increasing physical activity in medical versus psychiatric populations. When mental health targets were examined, they were largely included as secondary or exploratory outcomes and frequently framed in relation to coping with the medical concern of interest [99,100,104-108,110,111,113,114]. We only identified 16% (3/19) of the studies that specifically recruited individuals with psychiatric symptoms, and all (3/3, 100%) were focused on individuals with depressive symptoms [103,109,117]. The most common mental health outcomes were depression, anxiety, general quality of life, and emotional well-being [99,100,102-110,112,114-117]. Specifically, subclinical depressive concerns were the most frequently investigated psychiatric target, followed by subclinical anxiety [100-105,112,115-117].

The impact of smartphone-based physical activity interventions is yet to be investigated explicitly for individuals diagnosed with depressive or anxiety disorders, let alone other mental health conditions, including serious mental illness. Furthermore, although wide age brackets were represented across the studies, with average ages ranging from teenagers to older adults, men and non-White individuals were underrepresented. As research progresses in this space, it will be imperative to include the experiences and perspectives of adults with clinical levels of psychiatric concerns as well as diverse backgrounds and identities.

Owing to the limited available data; small samples comprising mostly White Western, educated, industrialized, rich, and Democratic women with subclinical depression or anxiety; and heterogeneity of outcomes measured, it is difficult to conclude whether and to what extent existing smartphone-based physical activity interventions benefit mental health. However, with these caveats in mind, we aimed to synthesize the available evidence in the following sections.

To date, interventions generally have not been tailored to the specific needs or presentations of individuals with mental health concerns. This is unsurprising as most technology-based physical activity trials have not been designed to target mental health. However, some studies have integrated components that specifically address psychological well-being. One adaptation that is low effort but high return is adding psychoeducation about the relationship between physical activity and mental health. For example, a recent trial for adults with mild to moderate depressive symptoms devoted the first in-person group session to discuss the relationship between depression and exercise to complement the personalized exercise program, smartphone app, and wearable device they received [103].

The second adaptation observed in the literature is offering concurrent psychotherapy-based tools. In some cases, psychotherapeutic content was interspersed with the physical activity intervention; for example, in a trial for adults living with HIV and depression in China, both an exercise promotion intervention and cognitive behavioral stress management course were delivered as multimedia messages through the WeChat app [104]. Relatedly, in a study of the eCoFit app for adults with or at risk of type 2 diabetes, short cognitive behavioral tasks (“FitMind Challenges”) were integrated throughout the program [116]. Examples of FitMind Challenges included motivational strategies, relaxation, cognitive restructuring, social support, and problem-solving. In other cases, the approaches were delivered in parallel. In one trial, adults who were overweight and experiencing depression received a 7-step problem-solving therapy via a workbook in addition to live lifestyle coaching, at-home video lessons, the MyFitnessPal app, and a Fitbit for monitoring [108]. In another study, SilverCloud’s guided internet-based CBT program for depression was the primary intervention, with smartwatch-based monitoring of sleep, steps, and mood added to promote greater awareness of the relationship between health behaviors and mood, thereby independently encouraging positive lifestyle changes [109].

A third but largely unexplored avenue is the inclusion of personalized or tailored messaging. This is an opportunity for coaches or other support persons to address barriers that may be specific to the experience of someone with mental health concerns (eg, navigating social anxiety to go to the gym and restructuring depressive thoughts). In one open trial of a physical activity app for youth with multiple sclerosis, coaches were trained in social cognitive theory for behavior change as well as motivational interviewing [114].

Although few digital physical activity interventions have been designed or modified to specifically affect mental health, many have been designed using evidence- and theory-based behavior change strategies that are ripe for implementation in psychiatric contexts. Indeed, the most successful interventions are based on behavioral theory [73,127]—explicitly stated or not—such as the transtheoretical model [128], the theory of planned behavior [129], self-determination theory [130], and social cognitive theory [131]. Social cognitive theory is most often cited given its emphasis on internal, external, and social factors that reinforce learning and contribute to sustained change [131]. Targeting self-efficacy, self-regulation, and social support to engender meaningful, lasting behavior change aligns strongly with principles of psychotherapy as well. Digital physical activity interventions have also experimented with numerous evidence-based behavior change techniques, including goal setting and review, action planning, regular feedback, self-monitoring of behavior, instruction and demonstration of how to perform a new behavior, graded tasks, prompts and cues, and social rewards, to name a few [132-134]. Interventions integrating multiple behavior change strategies are more successful than those that rely on one (eg, self-monitoring or reminders alone [133]). Considering how such strategies could be adapted for individual presentations (eg, those with clinical levels of dysregulation) should be a priority for future iterations of these programs.

Furthermore, technology-driven techniques may uniquely (or at least more strongly than traditional treatments) engage key mechanisms of behavior change. For example, these tools can promote self-awareness. As people tend to keep their devices close to them throughout their daily lives, wearable and mobile platforms can provide objective, continuous monitoring and feedback related to behavioral patterns such as physical activity [102,108]. In addition, these approaches can enhance a person’s likelihood of changing their behavior by lowering the cognitive burden involved in initiating physical activity. Strategies include delivering content more flexibly (eg, when it is most convenient for a participant to engage or in doses of their choosing); modeling target behavior via written, image, or video instructions that can be reviewed on demand or infinite times; or tailoring activity suggestions to a person’s present context (eg, suggesting at-home activities on rainy days). This may be particularly meaningful for psychiatric audiences as depression and anxiety are associated with attention and memory deficits that can interfere with information processing and learning [135-138]. In addition, in-the-moment rewards and other gamification or reinforcement features could be particularly useful early on [139,140] as individuals with depression and anxiety may not experience initial sessions of exercise as intrinsically gratifying or mood boosting as others do; for example, depression is characterized by deficits in reward processing and motivation [141], and anxiety sensitivity and social anxiety can blunt positive responses or promote avoidance [142,143].

Finally, personal devices may allow for more consistent, flexible social support throughout an intervention. Social support is an established, evidence-based behavior change technique that promotes physical activity [126,144-146]. Smartphone-based physical activity interventions provide a range of avenues for social connection, such as texting with a coach [101,107], access to an web or app-based discussion forum [115], and creation of virtual “teams” [102]. Critically, although social media has been frequently incorporated as a means of facilitating connection, participant reactions have been mixed, and it may not be optimal for psychiatric populations [126]. In general, social support appears to boost engagement when it is perceived to facilitate emotional support, provide tips from peers, enhance motivation, foster social comparison or competition [126,147,148]. How to best leverage social support and social media for psychiatric populations requires nuanced future study.

The primary aim of this narrative synthesis was to examine the status of smartphone-based physical activity interventions for mental health and understand how they have and have not been tailored to or evaluated in psychiatric populations. Ultimately, the literature is limited and difficult to synthesize owing to the high heterogeneity across the studies in terms of sample selection; study design; outcomes of feasibility, acceptability, and efficacy; and degree of tailoring. To date, mental health outcomes have typically been secondary or exploratory within trials focused on medical outcomes (eg, diabetes management) and, when included, have had a narrow focus on measures of depression, anxiety, and general well-being in nonclinical populations. As a result, this review relied significantly on research focusing on medical populations to explore how smartphone-based physical activity interventions could be used to impact mental health outcomes and to infer how they may be used in psychiatric populations. Furthermore, although extant studies have included diversity of age, the samples in the included studies comprised mostly White and female individuals, thus reducing the generalizability of the already limited findings.

The feasibility and acceptability of these interventions for subclinical and at-risk populations are encouraging and suggest that digital physical activity programs may be similarly well received among individuals above diagnostic thresholds. The available data on psychiatric outcomes were mixed. However, it is difficult to draw meaningful conclusions given the limited data; high heterogeneity of intervention approach and target behavior; and lack of standardization in measurement and reporting of use, engagement, and behavior change, as well as the elevated risk of floor effects given the subclinical samples. These inconclusive psychiatric outcomes may also be related to a lack of tailoring of smartphone-based physical activity interventions to the specific needs of those presenting with mental health concerns. The existing tailoring included basic psychoeducation about physical activity as a treatment, adding concurrent psychotherapy-based tools, and including personalized or tailored messages. There was no standardization or evidence base for how this tailoring was applied. The upshot is that many of the papers included in this review presented interventions that were already built around established, evidence-based behavior change strategies, which suggests that psychotherapeutic tailoring could be efficiently integrated into existing smartphone-delivered physical activity interventions. In general, effective physical activity interventions use many of the same fundamental behavior change strategies commonly found in psychotherapy, such as education, goal setting, self-monitoring, graded tasks, engaging social support, and motivational interviewing [149,150].

Taken together, the primary barrier to advancing the use of smartphone-based physical activity interventions in mental health care is the absence of evidence. The need for research in this area has been highlighted in other reviews as well [151,152]. To construct a more consistent, evidence-based foundation for intervention development, we outline several avenues for future research.

More research is needed to better understand how existing smartphone interventions can be tailored to fit the needs of psychiatric populations. The following are example adaptations rather than an exhaustive list. One likely critical step is to provide users with psychoeducation early on about the ways in which physical activity can be used to affect psychological health, such as improving mood and reducing anxiety. This should involve making explicit connections between health behaviors (eg, exercise), mental health symptoms, and emotion regulation so that users can better appreciate the bidirectional links between these areas of well-being. Including even a brief text summary of the literature or treatment rationale could likely augment the effects [153,154]. In fact, there is evidence with depression treatment that physical activity interventions lacking such psychoeducation or treatment rationale do not lead to robust clinical changes and can worsen dropout rates [155]. In contrast, attending to the mental and emotional benefits of exercise, particularly the acute or immediate impact on affect or resilience, can further enhance mood and motivation to continue exercising [35,156]. Highlighting these benefits and encouraging users to monitor such positive changes could improve sustained engagement and clinical response. Technology may be particularly helpful for this; apps, for example, can provide in-the-moment reminders through push notifications to attend to one’s affect or visual feedback of a user’s pretest-posttest change in self-reported mood with exercise. Furthermore, digital tools could provide information about the impact that mental health symptoms may have on program engagement. This can help users recognize that it is normal and expected for symptoms such as fatigue, anxiety sensitivity, or low motivation to serve as barriers to physical exercise and can proactively help users engage in related problem-solving.

Another compelling feature to test is the incorporation of modules or content that specifically address mental health concerns or symptoms. For example, technology-based physical activity interventions aimed at improving anxiety symptoms would benefit from including evidence-based skills such as cognitive restructuring and exposure practices. These approaches can be used to identify and challenge maladaptive beliefs that anxiety symptoms such as a racing heart are dangerous (known as anxiety sensitivity [157]), design a more graded exercise plan, and use activity as an interoceptive exposure by allowing patients to experience and tolerate those feared sensations [158]. Furthermore, CBT skills can be incorporated to address exercise-related social anxiety, such as integrating exposures (eg, walking with a friend or going to a gym first at off-peak hours) and challenging associated negative expectations (eg, “I won’t be able to keep up with my friend and they will judge me”). Meanwhile, a digital physical activity intervention aimed at helping people with depression could include skills consistent with behavioral activation, such as tracking the relationship between mood and activities (including physical activity); intentionally adding new behaviors such as gardening, walking, or going to the gym to their weekly schedule; and generating more flexible approaches to regular movement. Furthermore, cognitive skills can be used for participants to identify and evaluate negative thoughts about themselves or the program (eg, “I can never stick to my goal of going for walks, so what’s the point”) in terms of their accuracy or utility. In this vein, equipping coaches with some knowledge of common mental health symptoms to look out for, destigmatize, and address could enhance outcomes.

In addition, coaches are understood to bolster digital interventions in general by providing further psychoeducation or resources, personalizing content or skill use, and answering questions. Given the known relationships between mental illness and both inactivity and chronic medical conditions, even with basic mental health knowledge, coaches could perform these roles better. It is also important to recognize that digital or physical activity interventions may not be the appropriate or most effective level or type of care for all individuals experiencing mental health concerns. As such, tools should include information for users on the signs or symptoms that may indicate that pursuing psychotherapy could be beneficial as well as resources for doing so.

Finally, physical activity promotion tools could be integrated to augment existing treatments. Currently, there are a number of well-established treatments for psychiatric disorders, such as CBT and mindfulness. CBT has extensive research support as the gold-standard treatment for a range of disorders, including depression and anxiety [159]. This treatment integrates both behavioral and cognitive skills, such as tracking and scheduling activities as well as evaluating and challenging maladaptive thoughts, to reduce the severity and impact of symptoms. Mindfulness—or the purposeful, nonjudgmental awareness of the present moment [160]—is increasingly included in “third wave” interventions to reduce psychiatric symptoms [161] as well as increase well-being, such as positive affect and quality of life [162,163]. Recently, a dominant focus of digital mental health innovation has been translating these gold-standard psychotherapies to digital platforms. There is now a strong foundation of evidence supporting the feasibility, acceptability, and efficacy of delivering these treatments through both face-to-face and digital means [164-166].

Thus far, the development of digital interventions for physical activity and for mental health has largely occurred separately. However, their concurrent delivery provides promising initial evidence [111,118]. This parallels in-person trials demonstrating that increasing physical activity strengthens psychotherapy outcomes [167,168]. The next step in this line of research is to more formally integrate the 2 or even develop technologies in which both sets of skills are delivered within a single coherent platform. For example, in a study conducted by Wilczynska et al [116], adults with diabetes used the eCoFit app, which integrated guided workouts, goal setting, and cognitive behavioral skills. Some of the commercially available apps targeting mental health have already begun moving in this direction as well. The mindfulness-based app Headspace has recently incorporated a suite of video- and audio-guided exercises that help users engage in activities such as stretching, dancing, and yoga. Within these integrated platforms, it will be important to explicitly link the mental health and physical activity content rather than presenting them side by side as distinct intervention pathways.

Across intervention types, we must also consider the individual factors that may serve as facilitators of or barriers to engagement and success. This is doubly important for smartphone-based physical activity interventions as there are potential barriers inherent in both smartphone use and physical activity uptake.

The unfortunate reality of scientifically validated digital mental health products is that the vast majority do not move beyond their success in the laboratory [173], and those that do make it to market face fierce competition, flagging engagement rates, and a lack of financial means to scale the project. Thus, in developing and testing these promising digital mental health–focused physical activity tools, investigators should integrate elements essential for successful dissemination. One proposed pathway for improving the dissemination and ultimate success of digital mental health tools is to connect consumers through employers or public and private insurance companies, who have indicated a growing interest in expanding services to cover digital mental health. To illustrate, Blue Shield of California is now offering the mindfulness meditation app Headspace to subscribers [189]; Cigna offers the mental health app Ginger as part of its service package [190]; and Kaiser Permanente supports the use of Ginger, Calm, and MyStrength [191]. Physical activity technologies that track and manage exercise and step count are even more prevalently covered by insurance. Blue Cross Blue Shield, United, and others have partnered with Fitbit to offer low-cost wearable devices and use of their apps to promote health behavior change. Aetna and Cigna offer similar programs and occasional incentives to people who use health-tracking apps and devices. Taken together, the enthusiasm for apps and devices promoting physical health, as well as the recent foray by insurers into the digital mental health space, suggests that smartphone-based physical activity interventions for mental health may be prime for scalable coverage. This also means that academics developing such tools should be mindful when designing research studies to collect outcome data relevant to insurers and other payers, such as outcomes related to health care costs (reduction in insurance claims and physician or therapist visits), disability-adjusted life years (reduction in overall illness burden), adoption and engagement rates, and user data such as acceptability and fidelity (whether people use the tools as intended). In addition, investigators and designers should carefully consider the costs inherent to their interventions, such as relying on “off-the-shelf” versus research-grade devices and other platforms, the extent to which an intervention relies on human support to be administered, and the broader infrastructure required for implementation and sustainment, all of which will alter accessibility and scale. By designing research studies on smartphone-based physical activity interventions with true scalability in mind, researchers will be better poised to expand their intervention beyond academia and better achieve the goal of connecting evidence-based interventions with those who need them.

Physical activity has well-known and broad mental health benefits. However, a minority of at-risk individuals or those with mental disorders meet even the minimum exercise recommendations. Smartphones may bridge this gap given their pervasiveness in daily life, capacity to help concurrently manage multiple dimensions of personal health, and ability to engage key mechanisms of behavior change. Although early data for smartphone-based physical activity interventions reducing psychological symptoms are encouraging, overall, surprisingly little work has been done in this area. Therefore, there is untapped potential for developing and disseminating accessible, beneficial tools that can have a great public health impact.