Clinical Considerations for Digital Resources in Care for Patients With Suicidal Ideation

A wide variety of apps are currently available for mental health support, but no app is a panacea. Some of the more popular apps have been downloaded more than 100,000 times (12), suggesting that people are currently seeking and interested in trying these digital tools. Yet popularity does not always equate to quality. Although research suggests that these apps often meet minimum usability standards (12), most offer limited content and rely on a single therapeutic strategy (13, 14). The most common app-based program consists of connecting users to a crisis helpline or an emergency contact (12, 14) and providing suicide-related psychoeducation (14). More advanced, but less commonly offered, features include allowing users to build a safety plan (12–15), to find ways to make lethal means less accessible in their homes (12, 13), and to screen for suicidality or common comorbid conditions, such as depression, with validated assessments, such as the Patient Health Questionnaire–9 (13, 15). Table 1 summarizes how apps currently available in the marketplace have translated common approaches to suicide care and how future technological advances hope to improve on these capabilities.

Although these apps strive to offer useful help for suicidal ideation and behavior, it is concerning that many still present risks. Most mental health apps do not completely follow clinical guidelines on suicide prevention (12, 14) and do not offer users innovative therapeutic strategies (13, 15), but they rely instead on education and connection to external care (14, 15). In some cases, there have been concerns for patients’ safety, because apps have been reported to provide invalid helpline phone numbers (14) and potentially harmful content (13). Apps lack robust privacy policies (13, 15, 16) and have inconsistent or missing statements of how they manage users during emergency situations (16). Many apps also lack empirical evidence of their effectiveness (15, 17, 18) and have not been tested with diverse populations. Furthermore, most apps addressing suicidal ideation and behavior are patient facing rather than clinician facing, and thus they are not designed to be integrated into a care plan driven by a clinician (12). Of the current apps available, each should be evaluated for use based on a specific patient’s needs, because there is no one app that can address all needs. This is not to say that impressive apps for suicide prevention do not exist. For example, the SafeUT (Utah) app represents an evidence-based approach utilizing smartphone apps to increase access to mental health care and emergency services for Utah residents (19).

The new apps and technologies being developed for health care are also promising for future care for suicidal ideation and behavior. New features and uses of smartphones to capture real-time symptoms (via ecological momentary assessments) and behaviors (via sensors) provide new potential for the next generation of apps to offer more personalized and responsive support. Historically, research focused on determining patients’ short-term risk of a suicide attempt has been limited by a reliance on self-reports and lack of longitudinal assessments (8, 20). Wearables, smartphones, and smart home devices can mitigate these limitations. These technologies have the potential to measure patients’ real-time data, such as location through GPS, sleep state through accelerometers, and sociability through phone calls and text messages, over extended periods (21). These data can then be used to establish baseline behavioral patterns for each individual and detect when an individual deviates from normal patterns, potentially indicating that a suicide attempt might be proximal (21). However, research on using smartphones to offer such personalized help remains nascent (17). Recent research on ecological momentary assessments has shown promising, but at times mixed, results (20, 22, 23). Determining whether the integration of smartphone sensor data increases a smartphone’s ability to predict a future suicide attempt continues to be a focus of research. Despite promising and continuous advancements of these new data streams and analytical models (24), their predictive sensitivity needs to be refined, and their clinical utility has yet to be established (21). These new features are still in their early stages but offer promise for future apps.

Although the next generation of apps offers appealing features, clinicians need to make decisions on the use of apps based on what is available and known today. Today’s evidence and data on apps suggest that they are not equipped to replace human support (14) or to predict when a suicide attempt might occur. But they can help augment and extend services.

Apps can supplement and promote therapeutic goals at times when health care providers and an individual’s support team are unavailable, such as at night or between office visits. For example, apps can offer scheduled and on-demand distress tolerance exercises, libraries of cognitive-behavioral therapy and dialectical behavior therapy exercises, and even peer support, among other features. A key in using apps in this manner is to stress that they provide adjunctive support and are not to be used in place of any evidence-based interventions or standards of care.

Apps also collect data in real time and thus provide insights on trends, triggers, and recovery over time. Interpreting such longitudinal data can provide a clinician and patient with insight into environmental and behavioral contexts and factors that contribute to the patient’s health. However, it is critical to inform the patient that although data may be captured in real time, it is not viewed or interpreted in real time. This can help minimize false expectations about real-time response and how clinicians use the collected data.

When considering an app to offer adjunctive skills or to support data capture, there are thousands of choices and no need to limit the selection to apps focusing on suicide. For example, tracking one’s mood and journaling are common features of mental health apps that are not represented in many suicide apps (24) but that can be clinically relevant to patients who have suicidal ideation. Using apps in this broader manner opens the entire marketplace for selection, with many more options to support patients.

However, before even considering an app, it is worth exploring how patients can benefit by utilizing the basic features on every smartphone—i.e., native apps. Although mental health apps can offer unique features that are not replaced by apps native to the smartphone, it may be possible to use basic smartphone features to create a “digital toolkit” of resources. For example, patients can create a “hope box” by using the notes app, quickly contact preferred loved ones by “favoriting” their contact information, track physical health metrics through Apple Health or Google Fit, record supportive and positive affirmations in their audio recordings, take photos of favorite events or memories, and set reminders to keep track of medical appointments and other obligations. The benefits of native apps are that they are free and avoid privacy and security concerns that come with downloading an app from a third-party developer. Use of native apps can be especially beneficial for patients with lower baseline digital literacy, because patients do not have to download these apps or create an account to access potentially useful features. The examples presented here are solely examples, and clinicians and patients are encouraged to utilize the full features and functionality of a smartphone when looking to integrate mobile devices into care.

Of course, in many cases it will make sense to download an app to augment care. With an estimated 10,000 mental health apps available today and regulation still evolving, selecting an app can be a challenge. As in selecting a medication or therapy, picking an app should not involve star ratings or other popularity metrics, such as the number of downloads. Rather, matching each person to an app, just as with other therapies, is a better course. Several app evaluation frameworks have been developed that can further assist patients and clinicians during their app selection (25–27), including the American Psychiatric Association’s (APA) app evaluation (28, 29). The APA’s app evaluation is hierarchical; therefore, an app must meet lower-stage criteria before being assessed for higher criteria. For example, if an app does not satisfy privacy concerns, then it should not be assessed for its possible benefit or user engagement. The APA also offers free resources online (29), including, but not limited to, a rapid eight-item screener that can assist in quickly evaluating an app (28).

To further streamline the app selection process, our team created an interactive and searchable database of mental health apps, inspired by the APA model, called MINDApps.org (30). This database allows clinicians and patients to search a curated selection of approximately 650 available apps and customize the search by features that are important to them. Notably, MINDApps does not classify apps with stars or scores or as “good” and “bad” apps, because just as there is no one right treatment for every individual experiencing suicidal ideation and behavior, there is no one right app. Rather, MINDApps empowers users to find an app that meets all criteria that are important to them. Figure 1 illustrates an example of how MINDApps can be used to find a relevant mental health app.

Innovative solutions are needed to address current suicide rates. Apps provide a low-cost, scalable way for clinicians to supplement treatment with on-demand support and to promote therapeutic goals outside the office. Apps can offer basic self-management, connection to emergency contacts and hotlines, and educational content. Current limitations of apps include few innovative features, privacy concerns, and lack of empirical evidence of their effectiveness and safety. When recommending apps, clinicians can integrate app evaluation frameworks and interactive websites to personalize app selection and help ensure that patients choose an app that fits their needs. Future research hopes to integrate ecological momentary assessment to further advance smartphone augmentation of care for patients experiencing suicidal ideation or behavior.

The Sidney R. Baer Jr. Foundation supports the work of the Division of Digital Psychiatry at Beth Israel Deaconess Medical Center in developing and disseminating mental health apps for persons with serious mental illness. The Argosy Foundation provides support for MINDApps.org.