mHealth Assessment and Intervention of Depression and Anxiety in Older Adults

Abstract

Mobile technology is increasingly being used to enhance health and wellness, including in the assessment and treatment of psychiatric disorders. Such applications have been referred to collectively as mHealth, and this article provides a comprehensive review and clinical perspective of research regarding mHealth in late-life mood and anxiety disorders. The novel data collection offered by mHealth has contributed to a broader understanding of psychopathology, to an increased diversity of psychological interventions, and to novel methods of assessment that may ultimately provide individually adaptive mental health care for this population. Older adults face challenges (e.g., transportation, mobility) that limit their ability to receive medical and mental health care services, and mHealth may improve the capacity to reach this population. Although several mobile interventions exist for health-related issues in older adults (e.g., balance, diabetes, medication management), mHealth targeting psychiatric disorders is limited and most often focuses on problems related to dementia, cognitive dysfunction, and memory loss. Given that depression and anxiety are two of the most common mental health concerns among this population, mHealth has strong potential for broad public health interventions that may improve effectiveness of mental health care via individualized assessments and treatments.

By 2030, one in five Americans will be over the age of 65.¹ As with younger cohorts, mental health disorders, including depressive and anxiety disorders, affect a sizable portion of older adults (5% and 12%, respectively),² with anxiety disorders remaining common across the lifespan despite decreases with age.³ Generalized anxiety disorder and phobias account for most anxiety disorders in older adults, and these anxiety disorders are often comorbid with depression.^4,5

Depression and anxiety are associated with increased disability, increased health care utilization, and worse quality of life among older adults.^6–8 Mental health disorders also confer risk for several physical health conditions, including cardiovascular disease and type 2 diabetes.^9–12 In addition, depression is a known risk factor for psychological issues of older adulthood, including accelerated cognitive decline and Alzheimer’s disease and related dementias.^13–15 Even after controlling for sociodemographic and disease-related variables, older adults with high levels of depressive symptoms have a 25% higher risk of all-cause mortality.¹⁶ Therefore, treatment of mental health disorders among older adults is critical for both mental and physical well-being.

Effective treatments for depression and anxiety are available, but older adults are less likely to seek mental health care than younger cohorts,¹⁷ probably in part because older adults are less likely to report their symptoms and less knowledgeable regarding mental health disorders and available treatments.¹⁸ Barriers to treatment seeking among older adults include stigma, cost, transportation, and mobility.^19–21 Practically speaking, many older adults lose the ability to drive, leading to less independence and greater reliance on others. This is particularly problematic, given that many older adults live in rural areas away from high-quality treatment providers.²² The reduction or cessation of driving is also associated with increased depressive symptoms, perhaps due to its impact on functioning.^23,24 Thus, for older adults with depression and anxiety, accessibility of treatment may be a major factor in determining who receives effective care.

One way to improve access to mental health care for older adults is through mobile technology, including smartphone applications. mHealth, or the use of mobile technology for medical and mental health care, is increasingly being utilized to improve access and to augment care.^25,26 Smartphone applications (apps) have been touted as having particular utility for mental health disorders in adults in general, as assessment and treatment can be delivered in real-time and to a much wider audience.^27,28 Over the past several years, mental health professionals have begun partnering with industry specialists to adapt a wide range of mobile tools for mental health care. For example, Shen and colleagues²⁹ identified over 200 smartphone applications focused on depression treatment, psychoeducation, assessment, and related topics. Additional reviews found 117 apps offering depression treatment via cognitive-behavioral therapy or behavioral activation³⁰ and 9 randomized, controlled trials utilizing smartphone interventions for anxiety.³¹ Given this trend, there are undoubtedly many more such applications focused around depression and anxiety at the present time.

Despite all this work and interest, little attention has focused specifically on using mHealth to address mental health care among older adults. More attention is warranted because mHealth’s advantages may be particularly suited to improving quality and reach of mental health care among this population. With mobile technology, these individuals can receive care in their homes, thereby reducing concerns related to transportation, mobility, and cost. Although older adults typically have lower rates of technology use than younger individuals,³² smartphone use among individuals 65 and older has more than doubled since 2013.³³ In 2017, 42% of older adults owned a smartphone, and these numbers are likely to continue to rise as the growing population of technology-users ages.³³ Thus, it is important to consider how mobile technology can be leveraged to address the mental health concerns of older adults—by designing interventions to best harness advances in technology, by validating the utility of these interventions, and by identifying issues with implementation.³⁴

In the current review, we examine the rise of mHealth, with a focus on depression and anxiety disorders among older adults. We utilized PsycINFO, PubMed, and Google Scholar to seek empirical literature published prior to 1 November 2019. Search terms included combinations of the following: mobile, tool, device, technology, internet, mHealth, intervention, treatment, therapy, assessment, depression, mood, anxiety, mental health, health, older adults, and late life. Articles were screened for inclusion via search of keywords, examination of abstracts, and more thorough readings as needed. All articles including mHealth assessment or intervention for depression or anxiety among older adults were retained. Additional literature regarding other mHealth applications among older adults and general adult populations is also discussed more briefly.

We begin the review by examining research on acceptability and feasibility of mobile technology among older adults. We then review the uses of mobile technology in research and clinical settings to assess and treat depression and anxiety in this population; notably, research in these areas is limited, and only 13 articles related to these topics were found. We discuss ways in which mobile technology has been used for other health and mental health applications among older adults and the broader adult population, with the goal of identifying and exploring possible areas of growth. We conclude by discussing future directions for improving mental health care in older adults.

MOBILE TECHNOLOGY AMONG OLDER ADULTS

Acceptability, Feasibility, and Engagement

Older adults are often stereotyped as a population subset resistant to using technology.³⁵ In the late 2000s, the portrayal of a divide between people born before versus after 1980 was popular across fields concerned with education.³⁶ This portrayal suggested that those born after 1980 were digital natives whose facility and comfort with digital technology required new methods of teaching. Older adults, as digital immigrants, would be expected to never fully acclimate to new technologies such as smart phones. This portrayal was criticized at the time as being overly general and lacking empirical support.³⁶ Even so, there remains considerable variability in digital health literacy and use among older adults (as there is among and between other subpopulations)—an issue sometimes referred to as the second digital divide.^37,38 In spite of this concern, there are multiple reasons to think that older adults are increasingly comfortable with and able to use mobile technology.

Several studies have provided guidance as to whether mobile technology may be acceptable to older adults in research and treatment settings. Ramsey and colleagues³⁹ utilized smartphone assessments to measure anxiety, mindfulness, and depression for ten days before and after a mindfulness-based stress reduction (MBSR) or health education intervention. Participants in this study were older adults with self-reported cognitive concerns and a clinically diagnosed depressive or anxiety disorder. Notably, the participants in this study were not high-functioning older adults who might be expected to be more comfortable with technology. Average satisfaction was rated as “moderately high” for the daily mobile assessment, and comfort with the assessment was rated as fairly comfortable or very comfortable after 20 days of use. Additionally, 76% of participants completed at least one survey on at least half of the assessment days, and 70% of participants completed at least 30% of total surveys. Participants in this study demonstrated adherence rates comparable to those of younger adults without emotional or cognitive problems. In another study comparing tablet-based research consent versus paper-based consent, older adults reported similar levels of user-friendliness for both methods.⁴⁰ Results also indicated that tablet-based consent took longer than paper-based consent, but comprehension and retention of content was similar. These results suggest that mobile technology is accepted by older adults and feasible in specialized circumstances involving assessment and research.

Additional work reflects older adults’ willingness to use mobile research and intervention. Results from a review of mobile technology among older adults suggest that older adults are generally willing to engage in mobile interventions to monitor and improve health conditions and behaviors.³⁵ For example, more than a decade ago, a survey of health technology utilization in this population revealed that older adults were aware of the benefits of these technologies, with approximately 75% of respondents reporting willingness to use telemedicine to diagnose and monitor health conditions.⁴¹ Additionally, older adults with rheumatic symptoms indicated similar levels of willingness to receive email reminders for appointments and medication adherence as younger adults; however, older adults were less willing than younger adults to receive text message reminders.⁴² Current evidence suggests that mobile technology may be feasibly implemented for health and research purposes among older adults, including some support for adherence and usage among older adults with clinical symptoms of psychopathology.

Mobile Assessment for Depression and Anxiety

Mobile technology has experienced rapid advancement over the past decade, and the National Institute of Mental Health recently reported on opportunities and challenges of developing technology and mHealth for use in clinical research.⁴³ For example, passive sensing of biophysiological and environmental stimuli via mobile phone sensors offers the opportunity to understand pathological and normal developmental processes, including aging. mHealth approaches can also provide longitudinal data with the potential to improve ability to predict and prevent relapse. Iterative evaluations via mobile phone may also promote optimization of adaptive behavioral interventions that may otherwise not be possible among more traditional methods of assessment and in-person treatment. Overall, the possibilities of mHealth for mental health applications are promising, diverse, and many.

One method of assessment that has become increasingly accessible and user-friendly via mobile technology is ecological momentary assessment (EMA), a method by which individuals are repeatedly sampled in real time regarding present-moment symptoms during daily life. Prior to wide availability of smartphones, many EMA studies employed personal digital assistants that were not necessarily connected to the internet.^44–46 The typical study today uses a smartphone or another type of mobile technology.

A researcher or clinician who is accustomed to traditional assessment methods (i.e., interview or paper and pencil) might wonder if EMA via smartphone is worth the trouble. It is first worth noting that the proliferation and widespread adoption of mobile technology, combined with development of specialized EMA applications, has made the difficulties posed by EMA decrease over time. This is fortunate because EMA is thought to decrease recall biases and improve ecological validity over other forms of data collection, such as retrospective self-report.^47,48 More traditional assessment methods take a snapshot of a person’s experience by asking them to remember and sum their experiences over a period of time that is often arbitrary (e.g., “in the last seven days”) or unspecified (e.g., “in general”). EMA typically involves assessment of current experience (e.g., “To what degree do you feel depressed right now?”), which is a far simpler cognitive operation for respondents. Typical use of traditional assessments involves administration of measures two or three times across the course of a study or treatment; notably, this method does not provide much information about changes over time (i.e., whether they are linear, vacillating, and so on).⁴⁹ Traditional methods also frequently provide a minimally reliable assessment of change over time because of the typically poor reliability of difference scores for psychological symptoms.^50,51 By contrast, EMA allows for frequent and consistent data collection that is minimally intrusive and enables researchers and clinicians to gather information about what is true for an individual in daily life.^52–54 Furthermore, EMA has the benefit of including data collection at multiple time points without requiring participants to travel to a designated research laboratory or site. EMA thereby allows for detailed examination of change over time while also limiting burden—a benefit that is particularly relevant to older adults with limited resources and mobility.

Despite the potential benefits of mobile assessment via EMA, we are aware of only two studies that utilized EMA for assessing depression or anxiety among older adults (see Table 1). In one study, EMA was compared against standard paper-and-pencil measures for sensitivity to changes in clinical symptoms.⁵⁷ Participants in this study were randomized to receive either a mindfulness-based stress reduction intervention or a health education control intervention. Results from this study indicated that ratings via EMA (versus paper-and-pencil) evidenced greater improvements in depression, anxiety, and mindfulness in the MBSR intervention than in the control intervention. Paper-and-pencil measures evidenced an improvement only in anxiety during MBSR as compared to the educational control. The study also found higher test-retest reliability and other measures of validity with the EMA method, implying that these improved psychometric properties were responsible for detecting intervention effects using EMA as an outcome. Results from this study provide initial evidence that EMA may increase precision of symptom detection by limiting memory-related issues during retrospective self-reporting. Considering that older adulthood is associated with decreased memory performance even in healthy individuals,⁵⁹ EMA may be a particularly useful method for increasing accuracy of self-reports among this population.

Table 1.

Summary of Studies Using Mobile Technology for Assessing Anxiety or Depression in Older Adults

Study	Sample size	Mode of assessment	Frequency of assessment	Items and rating scale	Additional mobile data
Rabbi et al. (2011)55	n = 8	Waist-worn device	Continuous for 10 days	N/A: passive sensing used to predict mental and physical well-being	Motion and audio data
Sanchez et al. (2015)56	n = 12	Smartphone	Continuous for undefined number of days	N/A: passive sensing used to predict loneliness	GPS, incoming and outgoing calls
Moore et al. (2016)57	n = 67	Smartphone	3 times per day for at least 10 days	All items rated on a 5-point Likert scale ranging from 1 (not at all) to 5 (very much) Example items:  I am preoccupied by the past  I am preoccupied with the future  I feel hopeless  I feel depressed  I feel tense  I feel anxious  I feel nervous	N/A
Kim et al. (2019)58	n = 56	Actiwatch	4 times per day for 2 weeks	Rated mood level on a 10-point Likert scale ranging from 1 (very depressed) to 10 (not depressed)	Physical activity and ambient light exposure measured via Actiwatch

N/A, not applicable.

In another study, a prediction model using machine learning was developed for geriatric depression.⁵⁸ In this study, 47 older adults completed EMA regarding depressed mood four times per day for two weeks, and all participants wore a watch that measured their physical activity and ambient light exposure. Participants also completed self-report questionnaires regarding depressive symptoms at baseline and after the two weeks of EMA. The model evidenced good fit for prediction of depression via the combination of self-report, daily EMA, daily activity level, and light exposure data. These results suggest that EMA in conjunction with other diverse methods of data collection may identify individuals who may be depressed but have not yet been formally diagnosed. Such predictive ability may be tremendously beneficial to older adults with limited access to mental health care.

Despite the advantages of EMA, this method still relies on subjects being responsive and having reasonably accurate insight in reporting their experiences. Not all variables of interest can be easily self-reported. For example, depression and anxiety disorders, while defined primarily symptomatically in the Diagnostic and Statistical Manual of Mental Disorders,⁶⁰ are complex disorders of thought, emotion, behavior, neurobiology, and physiology. In older adults, depression is thought to be as much a cognitive and motoric disorder as an emotional one,⁶¹ and EMA-only approaches are arguably limited in their ability to tap non-emotional components of psychopathology that may improve precision of psychiatric diagnosis.⁶²

However, mobile devices are not limited to EMA; they can also collect vast amounts of data regarding our movement, location, speech patterns, app usage, and so on. As such, researchers may utilize passive sensing in combination with mobile performance tests (e.g., cognitive and motor assessments) and EMA to gather continuous and objective data from individuals.⁶³ For example, Servia-Rodriguez and colleagues⁶⁴ gathered physical activity, sociability, and mobility data via smartphone in over 18,000 Android users; they also assessed mood via EMA. Results indicated that mobile sensing data predicted self-reported mood with about 70% accuracy. Although mobile sensing data are not a plausible replacement for self-report, these data may hold information that people are not directly aware of or able to accurately report. For example, an examination of neural homeostasis found associations between neural changes, EMA, and recurrence of late-life depression, suggesting that monitoring neural shifts could potentially alert clinicians when their patients are at risk for relapse.⁶⁵ This type of non-recall-dependent measurement may be particularly useful among older adults, who often evidence difficulties with recall.⁶⁶

Among older adults, passive sensing has primarily been used to monitor health and safety (e.g., to detect falls). Regarding mental health, we found only two studies that focused on passive sensing for issues of depression and anxiety among older adults (see Table 1). One pilot study examined smartphone sensor data as predictors of loneliness among 12 older adults.⁵⁶ Results indicated that time spent outside the home and total number of outings were the best negative predictors of loneliness. Furthermore, Rabbi and colleagues⁵⁵ found that data collected via mobile sensors correlated highly with measures of depressive symptoms. In this study, passive data were collected via a small wearable sensor rather than directly via smartphone, but modern smartphones have similar capabilities to the wearable sensors used. Overall, utilization of passive sensing technology for health applications in older adults has been limited; however, we expect this area of research to expand as smartphone technology continues to advance.

Mobile Interventions for Depression and Anxiety

As alluded to above, mobile technology has great potential for interventions in older adults. For example, mobile interventions might provide greater reach that could be particularly meaningful to individuals living in rural areas, who often experience limitations in accessibility and availability of mental health resources.^67,68 Yet, specific research regarding mobile interventions for anxiety and depression in older adults is quite limited. In our review, we found no studies of mobile technology interventions specific to anxiety or depression in older adults.

The existing literature regarding other non-traditional delivery methods for older adults provides a clear picture of the potential applications for mHealth, and we report on this literature in the present review (see Table 2). For example, one study examined randomized interventions with telephone-delivered cognitive-behavioral therapy (CBT) or telephone-delivered nondirective supportive therapy (NST) for rural older adults with generalized anxiety disorder.⁷³ Notably, specifics were not provided regarding the types of telephones used by participants, and it is possible that treatment may have actually been delivered via mobile phone for at least some participants. After four months, both treatments resulted in reductions in clinical symptoms; at four-month follow-up, however, telephone-delivered CBT produced significantly greater improvements than NST in symptoms of generalized anxiety disorder, depression, and worry. Both treatments evidenced good adherence, with approximately 75% of CBT participants and 81% of NST participants completing the required number of treatment sessions (i.e., nine and ten sessions, respectively). A follow-up to this study compared ten weeks of yoga versus telephone-based CBT on worry, anxiety, and sleep among older adults.⁷⁸ Participants in this study were randomized to either of the two interventions, or else they were able to select their preferred intervention. Only preliminary results are available, but findings suggest that both interventions are promising, with higher efficacy for CBT (Brenes GA, Danhauer SC & Hargis G, unpublished data). Satisfaction was comparable among those who chose their treatment and those who were randomized. Results from these studies suggest that mHealth may be a viable way to overcome geographical barriers to receiving mental health care.

Table 2.

Summary of Mobile-Based Interventions for Anxiety or Depression in Older Adults

Study	Sample size	Population	Intervention(s)	Results
Spek et al. (2007)69	n = 301	Older adults with subthreshold depression	iCBT vs. group CBT vs. waitlist control	Participants in treatment conditions saw significantly greater improvements in depressive symptoms than those in the control condition No significant difference between the two treatment conditions
Zou et al. (2012)70	n = 22	Older adults with symptoms of GAD	iCBT for anxiety	Participants saw significant improvements in anxiety, stress, and depression symptoms
Dear et al. (2013)71	n = 20	Older adults with symptoms of depression	iCBT for depression	Participants saw significant improvements in depressive symptoms
Mewton et al. (2013)72	n = 225	Older adults prescribed iCBT by their primary care clinicians	iCBT for major depression, generalized anxiety, panic disorder, or social phobia	Participants saw significant reductions in psychological distress and disability
Brenes et al. (2015)73	n = 141	Rural older adults with GAD	Telephone-delivered CBT vs. telephone-delivered NDST	Participants who received telephone-delivered CBT saw greater improvements in anxiety and depressive symptoms than those receiving telephone-delivered NST
Titov et al. (2015)74	n = 54	Older adults with symptoms of depression	iCBT for depression vs. waitlist control	Compared to control group, treatment group had significantly lower levels of depression and anxiety symptoms at posttreatment and follow-ups
Titov et al. (2016)75	n = 433	Older adults with symptoms of anxiety and depression	Clinician-guided CBT vs. initial clinician interview plus self-guided iCBT vs. iCBT alone	Participants in all groups saw significant improvements in depression and anxiety symptoms No significant differences between groups
Jones et al. (2016)76	n = 46	Older adults with GAD	iCBT for anxiety vs. waitlist control	Compared to control group, treatment group had significantly lower levels of anxiety and depression at posttreatment and one-month follow-up
Wahbeh (2018)77	n = 50	Older adults with mild to moderate symptoms of depression	Internet-based mindfulness meditation vs. waitlist control	Participants who received internet-based mindfulness meditation saw greater improvements in depressive symptoms, perceived stress, insomnia symptoms, and pain severity than those in waitlist control condition

CBT, cognitive-behavioral therapy; GAD, generalized anxiety disorder; iCBT, internet-delivered CBT; NST, nondirective supportive therapy

Internet-based interventions have also been examined as a means of extending the reach of mental health care, including among older adults. Such interventions have typically been designed with the assumption that they are being completed on a desktop computer, but most do not try to prevent participants from using the treatment on a mobile device. As such, respondents may increasingly complete such studies using mobile devices regardless of the researchers’ intent. In one example, Wahbeh and colleagues⁷⁷ compared a six-week Internet Mindfulness Meditation Intervention (IMMI) with a waitlist control among depressed older adults. Results from this study indicated that those who received IMMI demonstrated significant improvements in depressive symptoms over the waitlist control both at posttreatment and seven weeks after posttreatment.

Internet-based CBT (iCBT) may also hold promise in extending access to evidence-based mental health care among older adults. Research suggests that iCBT is feasible, acceptable, and cost-effective among older adults with depression and anxiety.^{70–72,74,76} For example, a randomized, controlled trial (RCT) of iCBT among older adults with symptoms of depression found that individuals treated with iCBT reported significantly lower symptoms of depression and anxiety after three months of treatment than participants in the waitlist condition.⁷⁴ In another RCT for subthreshold depression among older adults, iCBT was compared against waitlist and group CBT conditions.⁶⁹ Findings from this study indicated that iCBT was significantly more effective than the waitlist condition and about as effective as group CBT, suggesting that iCBT may be an effective treatment modality among older adults with symptoms of depression. Another study⁷⁵ found that self-guided internet-based treatment may be as effective as clinician-guided treatment for older adults who exhibit mild-to-moderate symptoms of depression. Furthermore, preliminary evidence suggests that self-guided iCBT in older adults is a cost-effective method of treatment.⁷⁹ Together, these results suggest that self-guided iCBT may be a feasible method for extending mental health treatment to older adults. Additional research is needed to investigate whether mobile phone iCBT specifically designed for older adults may result in similar treatment effects for depression and anxiety as well as in increased accessibility to mental health care.

Notably, all of the above interventions share a common trait—that is, all of these interventions may be distributed via mobile format rather than whichever format was originally intended by the researchers. As alluded to above, increasing capabilities of smartphones and their integration into many people’s daily lives means that these interventions may become mHealth interventions in all but name as participants increasingly access the internet and voice telephone on their mobile phones instead of on desktop computers and landlines. At the same time, use of smartphones with older adults involves specific concerns and issues that we review below. We submit that it is time for researchers to plan for older adults to use smartphones whenever it is possible for them to access the intervention in that manner, whether the intent was to create an mHealth intervention or not.

CONSIDERATION OF OTHER MENTAL HEALTH AND HEALTH APPLICATIONS AMONG OLDER ADULTS

In spite of the potential benefits of mHealth, research regarding depression and anxiety specifically among older adults is lacking. Further examination of the literature also suggests a generally limited use of mHealth for other mental health problems among older adults, signaling a distinct area for potential growth. Recent reviews of mobile health technology for mental illness in older adults indicated a primary focus on issues involving dementia or cognitive dysfunction.^80,81 Additionally, one article detailing an iCBT writing therapy among older adults evidenced a decrease in posttraumatic stress disorder symptoms.⁸² Results from these studies reinforce a general trend of good adherence and feasibility; in addition to suggesting the suitability of mHealth in assessing issues related to cognitive dysfunction among older adults, these studies support the notion that mobile interventions may hold benefit in addressing wider and more varied psychological symptoms outside of dementia and cognitive dysfunction.

Among older adults, mobile technology has been utilized to a much greater extent for other health-related issues than it has for mental health. For example, a review of mobile interventions among older adults found 21 articles that focused on issues of older age, including diabetes, chronic obstructive pulmonary disease, Alzheimer’s disease and dementia, osteoarthritis, and fall risk.⁸³ Implementation of mobile technology varied widely between studies, including the use of mobile-phone cameras to document food intake among diabetics,⁸⁴ to send pictures of skin lesions to dermatologists for diagnosis,⁸⁵ and to document activities of daily life.⁸⁶ Mobile phones have also been used to assess fall risk via use of internal accelerometers,⁸⁷ and internet capabilities have been used to monitor and communicate outcomes from an exercise intervention.^88,89 These studies showcase the rich assortment of sensors and methods available to researchers interested in investigating health issues through use of modern mobile technology. These studies also demonstrate that mHealth technology has already been successfully implemented among older adults for problems of health, and it would be prudent to build upon these foundations to expand mental health care for this population.

Our review suggests that mHealth among older adults has thus far focused primarily on chronic medical conditions and issues related to cognitive decline. Although these issues are of major importance to older adults, our findings suggest a significant gap in the literature involving other common problems found among older adults, including issues related to depression and anxiety. On one hand, this gap is unfortunate, given the prevalence of problems with depression and anxiety. However, this gap also represents an opportunity for researchers and clinicians to learn from the existing literature to potentially make rapid progress in developing efficacious mobile assessment and intervention for depression and anxiety.

MOBILE TECHNOLOGY AMONG GENERAL ADULT POPULATIONS

In contrast to the limited mHealth research among older adults, an abundance of research has involved more general adult populations. Although the present review does not aim to thoroughly review this more extensive literature, research in this area may serve as valuable exemplars for further developing assessment and interventions for older adults.

For example, results from a recent meta-analysis suggested that internet and mobile-based interventions (IMIs) were significantly more effective than waitlist conditions in reducing symptoms of depression.⁹⁰ IMIs evidenced large treatment effects for depression at posttreatment and follow-up as well as moderate effects regarding anxiety reduction. Importantly, the examined IMIs evidenced similar levels of effectiveness for depression, suggesting that the specifics of each IMI were not as important as receiving an IMI in general. These findings suggest that IMIs may be effective for general adult populations. Additional, more focused research regarding their effectiveness among older adult populations is warranted.

Computer-based CBT has evidenced effectiveness and feasibility for treating depression,^91,92 and it is likely that a computer-based approach to CBT would share considerable overlap with digital CBT delivered via mobile phones. An RCT comparing mobile phone versus computerized CBT for depression among adults found good satisfaction and adequate adherence among users. Symptoms of depression were significantly reduced at posttreatment and three-month follow-up for both types of treatment, and effects between conditions did not differ.⁹³ It is currently uncertain whether mobile CBT may be adequate as a stand-alone treatment for depression, but other research suggests that mobile CBT may improve treatment engagement by acting as a supportive tool for self-monitoring.⁹⁴ Although current evidence indicates that mobile CBT is effective as an adjunct to more typical forms of in-person treatment, it may also have potential as a sole form of treatment for those who do not otherwise have access or means to receive mental health care. This potential may be particularly applicable to extending care to older adults who face limitations of mobility and access to mental healthcare.

Regarding passive sensing technology, Burns and colleagues⁹⁵ implemented a momentary ecological intervention for depression that utilized at least 38 concurrent phone-sensor values and ecological momentary assessment to compare predicted values of mood, environmental context, cognitive states, and so on versus self-report. This study also incorporated machine-learning models to predict participants’ mood states. Participants reported improved symptoms of depression and anxiety over the course of an eight-week intervention. Such an application may hold significant promise as an intervention for older adults experiencing symptoms of depression and anxiety. For example, older adults experiencing increased difficulties with retrospective self-report related to cognitive decline may benefit from real-time entry, monitoring, and feedback regarding symptoms and mood. Individuals experiencing increased physiological symptoms related to health issues of older age may also benefit from increased monitoring via diverse and generally unobtrusive sensors included in modern mobile devices. Predictive machine-learning models may benefit older adults with limited diversity of environmental and social contexts by delivering interventions that may be optimally suited to the given situational context. These methods may serve to tailor interventions such that they are capable of effectively addressing several of the unique concerns of older adulthood.

Mobile technology has been utilized for anxiety in adults. For example, the Challenger App was designed as a treatment intervention for individuals with social anxiety disorder, either as an adjunct to treatment with a therapist or as a means of exposure for individuals with mild-to-moderate symptoms.⁹⁶ Elements that are unique to this application include gamification of treatment goals, customizability of challenges by the user, and incorporation of anonymous feedback from other users. Although efficacy information is not currently available for this application, it highlights several aspects that may be particularly beneficial for treating social anxiety disorder in older adults. For example, gamification of treatment may encourage greater adherence, and customizability may allow older adults to enter and utilize locations and situations that are unique to their living situations and social interactions. Perhaps most notably, feedback from other users may provide a source of encouragement and positive feedback to older adults who may have fewer social connections and daily interactions with others.

DISCUSSION AND FUTURE DIRECTIONS

The present review suggests that mobile technology has the potential to lessen the barriers of older age by providing care wherever older adults are located. For example, existing research indicates that mobile and internet-based interventions are generally similar in efficacy to the more traditional interventions that they are based on. Further, the format is typically acceptable to those treated for depression and anxiety, including older adults. We recommend that additional research should focus on evaluations of mHealth efficacy among older adult populations; not only is this population historically underserved, but the flexibility of service provision that is possible via mHealth may be particularly suited to benefit a population that frequently struggles with issues of transportation and mobility.^97,98

Among older adults, mHealth has been primarily used for issues related to cognitive decline and health issues such as diabetes and osteoarthritis; it has been used only minimally for treating anxiety and depression in this population. Although considerable research has found mHealth to be effective for anxiety and depression in the general adult population, few of the hundreds of available apps have been formally validated by research efforts. Nevertheless, individuals seeking assistance with their mental health symptoms may increasingly try such apps, and older adults—despite their often limited knowledge of mental health and information technology—may be particularly likely to do so. Clinicians should be aware, however, that most apps available are not evidence based, and even those that may claim (with some grounding) to be based on evidence are unlikely to be thoroughly researched. We obviously need further research and development regarding apps targeting anxiety and depression in later life. At the very least, and in the short term, it would be helpful to determine whether and which apps intended for general populations are similarly effective for older adults. Such research and development efforts hold significant promise for improving quality of life for a great many suffering from symptoms of depression and anxiety.

Through this review, we have identified several promising future directions. For example, mHealth applications have been utilized in general samples to help patients track diet and exercise, find a health care professional, schedule an appointment, fill a prescription, obtain information regarding diagnosis, and be reminded regarding scheduling, medication intake, and other such matters.⁹⁹ All of these uses may hold some benefit for older adults. For example, poor diet and decreased exercise are associated with increased symptoms of major depression.¹⁰⁰ Thus, older adults with emotional disturbances may particularly benefit from tracking and improving their diets and exercise via mobile apps. Apps with integrated capabilities to find mental health care options, schedule an appointment, and fill prescriptions may also reduce barriers to seeking treatment. Older adults underutilize public mental health care,¹⁰¹ and providing information regarding diagnoses and psychoeducation may improve their understanding of mental health symptoms and encourage them to seek and utilize treatment. Because older adulthood is associated with decreases in memory and cognitive functioning, app-based reminders may also improve both attendance at mental health appointments and overall medication adherence.¹⁰²

In addition, the combination of ecological momentary assessment and the ability to adapt interventions for individual users makes smartphones a particularly promising avenue for precision medicine. For example, EMA could be used to precisely and dynamically assess an older adult’s level of depression and anxiety, leading to recommendations for apps, assessment in-person, or other types of interventions. Intervention apps could be designed to maximize patient engagement, including shifting engagement strategies based on user behavior or passive sensing. Whereas traditional treatment methods make it challenging to implement many different interventions reliably, mHealth offers the opportunity for true precision medicine, with each user getting an intervention that is individually tailored to maximize engagement and provide the precise combination of components that best serves their needs.

Such promise comes with significant challenges. Research regarding engagement suggests that apps may be downloaded by thousands of individuals, but often only a fraction of users will remain active beyond a short period of time.^103,104 A recent examination of object-relations theory and engagement in mHealth suggests that mHealth engagement may be increased by including nonspecific factors related to empathy and by improving users’ self-reflective capacities.¹⁰⁴ Given the limited social relationships and decreased cognitive functioning of many older adults, these components may aptly improve engagement among this population.

Implementing mHealth for depression and anxiety among older adults will require further research regarding efficacy and acceptability. Currently, it is not well understood whether mHealth for anxiety and depression may be suitable as stand-alone treatment versus adjunctive treatment for more typical in-person interventions, and for whom each method is likely to be most effective. Furthermore, additional research on symptom severity and aging may provide insight regarding whether mobile interventions may be most suitable for particular subsets of people. Although mobile tools are generally found to be acceptable to older adults, several factors may enhance or inhibit utilization of mobile technology.³⁵ For example, because older adults experience impairments to visual acuity and motor function, efforts should be taken to ensure that mobile applications are designed with font size, color, and user experience in mind. Older adulthood is also associated with hearing loss; thus, utilization of videos and audio should be considered – and possibly minimized or simplified – during development of mobile applications. Further elements that enhance acceptability should be included in apps for older adults, including unobtrusive integration of the app into life, provision of feedback when a task is done successfully, and promotion of a low perceived difficulty of learning how to perform tasks on the app. Apps already developed for younger audiences may require adjustments and testing in order to meet the needs of older adults.

Although clinical evidence in support of most mHealth apps is lacking, technology and mHealth innovations are advancing at a rapid pace.¹⁰⁵ Part of this velocity gap may be related to reluctance to abandon tried-and-true methods of the past in favor of new models and techniques that are unfamiliar and not as extensively tested for effectiveness.¹⁰⁵ Therefore, mHealth may benefit from implementation that bridges the old and the new. For example, conceptualizing mHealth technologies as technology-enabled services that incorporate human support instead of stand-alone treatment products may result in more consistent benefits to mental health.¹⁰⁶

mHealth has often been used to deliver evidence-based treatments such as CBT via a mobile platform, but technology holds unique promise in its ability to deliver fundamentally novel interventions involving virtual reality, artificial intelligence, and so on.¹⁰⁶ Resources are limited, and developments in mental health care may rely upon researchers’ willingness—despite the risks and uncertainties—to develop, adopt, and test new methods of treatment delivery. Perhaps key to both the initial and sustained success of mHealth is its capacity to incorporate feedback regarding user preferences and concerns.¹⁰⁷ No matter how effective a particular app may be in improving psychological symptoms, it will not make much of a difference unless users are willing to adopt and continue using it.

Questions of engagement and implementation deserve careful attention and consideration, but the promise of mobile interventions to provide both broad-based public health interventions and precision medicine implies that these challenges are worth overcoming. Current trends suggest that older adults will increasingly own smartphones, which will make delivery of public health interventions via this channel increasingly effective in terms of reaching a wide audience. Further research is necessary to extend recent advancements in mobile mental health care to the treatment of older adults. It is especially important to capitalize on these opportunities, given that older adults frequently experience limited mobility, a common barrier to seeking mental health treatment. As such, mobile technology has the potential to significantly reduce the impact and burden of mood and anxiety disorders among older adults.