Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 May 21;58(8):1415–1424. doi: 10.1002/eat.24468

Integrating Artificial Intelligence and Smartphone Technology to Enhance Personalized Assessment and Treatment for Eating Disorders

Jake Linardon 1,✉, John Torous 2
PMCID: PMC12336790  PMID: 40396625

ABSTRACT

Objective

Smartphone technology presents a promising path toward expanding access to evidence‐based eating disorder assessment and treatment. Despite rapid technological advances, research has yet to harness these systems in ways that make personalized digital health care a clinical reality. In this forum, we review extant research testing smartphone intervention and monitoring tools for eating disorders and explore innovative ways integrating this technology with AI can enhance assessment, symptom detection, and intervention efforts.

Method

We highlight three capabilities of smartphones that hold promise for delivering personalized and maximally effective digital health tools: (1) passive sensing and digital phenotyping; (2) natural language processing of reflections from in‐app homework tasks; and (3) closed‐loop adaptive interventions. We discuss how these capabilities can augment current assessment and treatment efforts and draw on literature from other fields to inform research questions for the eating disorder field.

Results

Evidence from other fields demonstrates the feasibility of constructing data‐driven models from smartphone sensor data and textual input from in‐app CBT activities to predict clinical outcomes. These models may inform closed‐loop interventions, enabling apps to deliver timely, personalized support in response to real‐time changes in a user's needs.

Conclusion

The eating disorder field can draw on lessons from other fields to evaluate smartphone technology that leverages AI to enhance personalization. Realizing the potential of these tools will require addressing challenges related to engagement, trust, data governance, and clinical integration. The testable research questions presented here offer a roadmap to guide future large‐scale, collaborative efforts aimed at transforming eating disorder care.

Keywords: artificial intelligence, digital health, digital phenotyping, eating disorders, machine learning, smartphones

Summary.

  • Smartphones can augment assessment, detection, and treatment for eating disorders, yet this technology remains underutilized in this field and has yet to reach its potential.

  • Sophisticated technological capabilities like digital phenotyping and natural language processing may aid the detection of eating disorders in new ways while also informing the development of closed‐loop, adaptive interventions.

  • The eating disorder field can draw on lessons from other domains of mental health to evaluate smartphone technology that leverages large language models and AI‐driven algorithms to enhance personalization.

Eating disorders run a chronic course, impair functioning, and are associated with high rates of comorbidity and mortality (Solmi et al. 2024). Although considerable progress has been made toward developing valid assessment and diagnostic tools, clear guidelines for clinical care pathways, and evidence‐based interventions, the reality remains that the vast majority affected go undetected and untreated (Weissman and Rosselli 2017). Alongside financial and geographical barriers, a major contributing factor remains the shortage of qualified mental health professionals to deliver appropriate diagnostic assessments, case formulations, or treatment plans, which results in limited access to services, lengthy waitlists, and delays in intervention that can lead to poorer illness prognosis (Kazdin et al. 2017). In response, there have been growing calls for innovative solutions that can expand the availability of services and ensure that all those affected can receive timely, appropriate support (Linardon and Fuller‐Tyszkiewicz 2025).

This forum article discusses the utility of smartphone technology (mobile devices with internet connectivity, app‐based software, and advanced computing capabilities) as one potential solution to expand access to eating disorder assessment, symptom detection, and intervention protocols. We do not argue that smartphone technology should substitute traditional clinical services but rather propose it as a complementary tool that enhances reach, efficiency, and continuity of care in ways that currently remain underutilized in the eating disorders field. We begin by reviewing existing research on the use of smartphone technology in the context of eating disorders. We then highlight key opportunities for these tools to augment assessment, symptom detection, and intervention, with the aim of generating testable research questions and inspiring future work. We conclude by outlining central challenges associated with the use of smartphone tools integrated with AI systems, drawing attention to potential obstacles that may impede progress in harnessing their full potential. We ultimately aim to foster global discussion on how these challenges can be effectively addressed and overcome in future research and implementation efforts.

1. Current Landscape of Research on Smartphone Technology for Eating Disorders

People with eating disorders represent a population that could benefit from smartphone app‐based intervention and monitoring tools, given that the ego‐syntonic nature of these conditions often results in ambivalence to change (Clausen et al. 2013). Apps also enable individuals to engage in treatment at their own pace, fostering a sense of autonomy and control over the therapeutic process (Torous et al. 2021). The scalability, flexibility, and affordability of apps position them as promising tools to overcome barriers to accessing traditional clinical services.

Recent research suggests that there is growing interest in smartphone‐based intervention and monitoring tools among individuals with eating disorders (Torous et al. 2025). With global rates of smartphone ownership now approaching 90% (Pew Research 2024), and access to the internet now more common via smartphones than computers, the scalability of smartphone tools is impressive. Survey studies indicate that some individuals express a preference for digital tools over traditional face‐to‐face options, particularly for managing symptoms and tracking progress outside of formal therapy (Anderson et al. 2024; Linardon et al. 2021). Similar trends have been observed among clinicians, with many reporting openness to incorporating apps as adjuncts to treatment to facilitate between‐session engagement (Liu and Linardon 2025). These findings suggest some degree of enthusiasm and acceptability toward smartphone technology for eating disorder population, highlighting its potential to augment traditional treatment approaches and expand access to additional help options.

In practice, smartphones are already being used across several stages of eating disorder care. As self‐management tools, apps can support individuals in monitoring symptoms, practicing coping strategies, and reinforcing healthy behaviors between or outside of therapy sessions (Linardon et al. 2024a). Smartphones are also being integrated into traditional treatment frameworks; for example, patients may use apps alongside psychotherapy to enhance adherence to treatment plans, complete homework tasks, or log emotions and behaviors in real‐time, allowing for more targeted therapeutic discussions (Liu and Linardon 2025). Aftercare is another promising application, with apps providing continued access to resources and monitoring following discharge from intensive treatment, helping to prevent relapse (Neumayr et al. 2019). Emerging hybrid models of care are beginning to utilize smartphone data to inform clinician action, where real‐time passive and active data streams can alert providers to early signs of symptom deterioration, prompting timely intervention (Macrynikola et al. 2025). These examples illustrate the evolving role of smartphone technology as both a stand‐alone and adjunctive resource across the continuum of eating disorder care.

Monitoring eating disorder symptoms through smartphones has been extensively explored through ecological momentary assessment (EMA) designs. EMA involves repeated sampling of patients' thoughts, feelings, and behaviors in real‐time within their natural environment (Shiffman et al. 2008). Smartphones are particularly suited to this methodology due to their portability, ubiquitous use, and capacity to capture data relatively unobtrusively as individuals move throughout their daily lives. In this way, EMA addresses key limitations associated with traditional retrospective or global summary assessments, which typically lack ecological validity, are subject to recall bias, and fail to capture variability in experiences over time (Moskowitz and Young 2006). There has been considerable research into the use of smartphone‐assisted EMA in eating disorder populations over the past decade, with prior reviews (Schaefer et al. 2020; Smith et al. 2019) suggesting that these studies have been broadly designed to (1) examine functional relationships between risk factors and symptoms, (2) identify novel predictors of symptom onset or escalation, (3) characterize symptom dynamics and variability, and (4) investigate mechanisms that underpin treatment.

Comparatively less research has been conducted on smartphone app‐based interventions for eating disorders, with repeated calls for more rigorous studies designed to evaluate their utility, efficacy, and implementation (Fuller‐Tyszkiewicz et al. 2024). A prior review identified 65 commercially available apps intended to support eating disorder treatment but found that the overall quality of these apps was poor: 92% lacked critical in‐app features, 20% offered crisis management support, and only 7% were supported by research (O'Leary and Torous 2022).

While these findings highlight potential risks associated with current marketplace offerings, several eating disorder apps developed specifically for research purposes are beginning to accrue evidence supporting their efficacy. For instance, incorporating the Noom app (an app that digitizes CBT self‐monitoring journals) into traditional face‐to‐face services has led to greater compliance, more frequent therapeutic skill utilization, and quicker reductions in symptoms (Hildebrandt et al. 2020). The TCApp is another example that has demonstrated some benefits in augmenting standard outpatient treatment (Anastasiadou et al. 2020). Randomized controlled trials (RCTs) testing the Break Binge Eating (Linardon et al. 2022b) and Resilience (Linardon et al. 2024a) apps as stand‐alone, self‐guided, early intervention options also found evidence of short‐ and long‐term symptom and well‐being improvement. A follow‐up trial that incorporated a simplified monitoring version of the Break Binge Eating app alongside a broader, modular‐based CBT web program produced greater retention rates (Linardon et al. 2022a). Preliminary pilot data show that the Recovery Record app may function as an acceptable and efficacious relapse‐prevention tool following the discharge of intensive treatment among patients with anorexia nervosa (Neumayr et al. 2019). Together, these findings highlight the versatility of smartphone apps as promising tools for eating disorder management, demonstrating their potential as stand‐alone interventions, adjunctive supports to traditional treatment, or after‐care programs aimed at preventing relapse.

Smartphone technology holds promise for enhancing the assessment, detection, and intervention of eating disorders, yet research has only begun to tap into its full potential. While existing studies demonstrate that smartphone apps can aid in symptom monitoring and delivery of interventions at scale, their broader capabilities, including the capacity to collect, analyze, and adapt to extensive passive, active, and metadata streams, remain largely underexplored. To realize the full potential of smartphone technology, a fundamental shift is needed in how eating disorder researchers study, conceptualize, and administer these tools, capitalizing on their unique capacity to provide personalized risk assessments, real‐time support, and adaptive interventions during critical moments. In the next section, we discuss key potentials of smartphone technology that are largely unexplored in the eating disorder field, drawing on insights from other areas of mental health to illustrate how harnessing these capabilities could enhance current screening, detection, and intervention efforts.

2. Smartphone Sensors and Digital Phenotyping

One of the unique advantages of smartphones is their ability to continuously capture large amounts of diverse passive data through in‐built sensors, such as accelerometers, Global Positioning System (GPS), and microphones. This capability has led to the emergence of digital phenotyping, which uses sensor data collected via smartphones to derive real‐time behavioral, social, or environmental metrics that can be analyzed to infer mental states (Torous et al. 2016).

Predicting symptom escalation or relapse using objective, unobtrusive data collected passively through smartphone sensors could help augment the clinical care of eating disorders. Unlike traditional assessment methods that rely on retrospective summaries, digital phenotyping collects data continuously in real time, reducing inaccuracies or biases stemming from memory recall, lack of illness insight, or variability in assessor competency (Onnela and Rauch 2016). Similarly, because sensor data can be collected unobtrusively as patients go about their daily lives, it not only minimizes the burden associated with repeated self‐reporting but can also capture subtle yet crucial behavioral changes that individuals may not consciously detect through active data collection methods such as EMA (Torous et al. 2016). Machine learning algorithms can also be generated to analyze temporally dense, multimodal incoming data streams to inform patients of relapse risk and then deliver personalized intervention resources just in time (Torous et al. 2019b). These capabilities may offer new opportunities to enhance clinical care for eating disorders by supporting more proactive, timely, and individualized interventions and assessment protocols.

Efforts to understand the role of sensor‐derived data in predicting eating disorder outcomes do exist, though studies have predominantly used non‐smartphone‐based digital devices. Glucose monitoring devices have been used to examine glycaemic variability, with findings showing that greater fluctuations are associated with increased binge eating (Presseller et al. 2022b). These findings have been used as feasibility data to guide the development of adaptive interventions targeting glucose stabilization (Juarascio et al. 2022). Furthermore, wristband devices have been used to passively track physiological signals, such as heart rate variability and electrodermal activity, with machine learning models based on these inputs predicting eating disorder behaviors with up to 90% accuracy (Ralph‐Nearman et al. 2024). The Binge Eating Genetics Initiative (BEGIN) (Bulik et al. 2020) has paired Apple Watches with Recovery Record to collect passive physiological and behavioral data, with promising pilot results suggesting feasibility for predicting the occurrence of eating disorder symptoms (Butner et al. 2025). Other types of sensor technologies that have been explored in this field include pedometers, armbands, polysomnography, actigraphy monitors, and polygraph systems (Presseller et al. 2022a). However, research exploring these digital devices is generally considered a distinct field from smartphone‐based sensing, owing to important differences in hardware configurations, supporting software infrastructures, user burden, and scalability for real‐world deployment (Linardon 2025; Torous et al. 2021).

Using smartphone sensor data to predict clinical outcomes has been extensively applied in mood, schizophrenia, and anxiety disorders (dos Santos et al. 2024; Torous et al. 2021). Depression symptom severity can be predicted by behavioral features derived from smartphone sensor data, including mobility patterns (step count and GPS‐derived location data; Faurholt‐Jepsen et al. 2022), social interaction (frequency of text messages and calls; Cao et al. 2020), and circadian rhythm disruptions measured from location data (Zhang et al. 2022). Relapse in schizophrenia can also be predicted by changes in social behavior (operationalized as reduced duration and volume of calls and messages; Buck et al. 2019), while negative symptoms and auditory hallucinations can be predicted by mobility patterns (number of location visits) and social behavior (frequency of text messages) the following day (Buck et al. 2022; Depp et al. 2019). Analyzing both passive (GPS, accelerometer, screen time) and active (survey response) smartphone data together can predict with strong accuracy changes of 4 points or greater in GAD‐7 scale scores for general anxiety (the threshold for a minimally important difference) one week in advance (Cohen et al. 2024), while variability in calls, texts, and movement patterns over 2 weeks can predict social anxiety severity and accurately distinguish these symptoms from depression and affect (Jacobson et al. 2020). In non‐clinical samples, elevated stress and loneliness have also been predicted by novel aspects of movement and speech patterns captured through GPS and microphone sensors (DaSilva et al. 2019; Doryab et al. 2019; Melcher et al. 2020). These advances in using digital phenotyping data to assess environmental reactivity, longitudinal symptom changes, and reduce clinical heterogeneity offer clear parallels and relevance for treating eating disorders.

While these findings require further validation through prospective designs with larger and more diverse samples (Currey and Torous 2023), they provide proof‐of‐concept for the potential of passive smartphone data to detect and predict changes in mental health. They may also provide tangible starting points for hypothesis‐driven research on the impact of social patterns, sleep, screen time, and other factors that are clinically relevant but challenging to assess accurately today. Better measurement can also facilitate prevention and early detection of eating disorders. The ability to detect symptoms of depression and stress through sensor‐derived indicators (e.g., mobility or speech patterns) may shed new light on the nature, course, and timing of eating disorders, as these symptoms are also well‐established precipitants of or mechanisms that underpin binge eating and purging (Haedt‐Matt and Keel 2011; Mason et al. 2018; Srivastava et al. 2021).

Other hypothetical behavioral or physiological features derived from smartphone sensors may also be relevant for detecting symptoms, risk factors, or maintenance factors, and merit further exploration as they have not yet been empirically tested. For instance, changes in physical activity patterns captured via accelerometer data or actigraphy could signify compulsive exercise behavior, while anomalies in GPS‐tracked movement (e.g., repetitive gym visits or avoidance of eating‐related locations) may signal compensatory behaviors or restrictive eating. Elevated heart rate variability or disrupted sleep patterns, measured via smartphone‐integrated sensors or connected wearables, could serve as early markers of emotional dysregulation and stress‐related eating behaviors. Alternatively, excessive screen time on social media platforms or calorie‐tracking apps may reflect heightened appearance concerns, while changes in communication patterns (e.g., frequency of calls or messages) could indicate social withdrawal. Beyond these core symptoms, smartphone sensing could also be leveraged to address maintenance factors: for example, fluctuations in app usage following social comparison experiences could help detect body preoccupation, and irregular geolocation patterns related to eating contexts could signal inflexible eating restraint.

The field should capitalize on the growing capabilities of smartphone sensors to predict clinical outcomes. It can also learn from challenges that other fields have experienced and set standards for digital phenotyping data collection and processing. With a unified approach, the eating disorder field could quickly become the leader in the space and be the first to harness smartphone sensing technology to advance care. Detailed tutorials and guidelines on digital phenotyping research have been published (Barnett et al. 2020; Burns et al. 2024) and would be of value for eating disorder researchers exploring this field.

3. Language‐Based Data and Natural Language Processing

Eating disorder apps offer several activities that are grounded in evidence‐based treatment protocols, such as food diaries, thought restructuring records, and exposure experiments. These activities require users to input detailed thoughts, feelings, behaviors, or other contextual information to facilitate therapeutic change. Therefore, apps can capture rich, patient‐generated, novel textual data that could be used through natural language processing (NLP; a set of machine learning techniques used to analyze and interpret human language; Chowdhary and Chowdhary 2020) to predict important clinical outcomes.

Analyzing written text from activities completed in eating disorder apps may generate new insights that support personalized approaches to care. For example, it could help identify whether certain CBT tasks are being completed as intended, ensuring that users are accurately linking their thoughts and emotions in a way that supports rather than hinders therapeutic progress. To illustrate, Furukawa et al. (2023) applied a large language model to analyze thought‐feeling pairs entered into a depression app, finding that when the model's predicted emotions closely matched participants' self‐reported emotions, those entries were associated with greater reductions in negative mood following cognitive restructuring. Integrating such an AI model into a CBT app could prompt users to revise their automatic thoughts when mismatches with emotions are detected, potentially enhancing the effectiveness of the technique. Another example of NLP's capabilities in this context is its potential to predict therapeutic progress by analyzing changes in language patterns within users' written inputs. Mylläri et al. (2022) applied topic modeling to analyze diary entries from internet‐based CBT for anxiety disorders, finding that certain textual themes including worries about others predicted better treatment outcomes, while other textual themes including self‐focused worry and complaints about treatment predicted poorer outcomes. The potential is even broader with the ability of patients to opt to share their Internet search data. While ethical questions remain, NLP may be able to anonymously parse Internet search data to highlight trends of clinical relevance.

These examples highlight the untapped potential of NLP to enhance the clinical utility of written content generated in eating disorder apps. By moving beyond static symptom monitoring to deeper analysis of users' language, NLP can support more adaptive, individualized interventions. In addition to identifying correct skill utilization and predicting treatment outcomes, NLP could also be leveraged to determine the severity of symptoms, detect abrupt shifts in cognition, and capture early warning signs of relapse. These represent important and testable research questions the field could begin to address, and taking advantage of data already collected in prior trials of eating disorder apps could expedite progress in this area.

4. Closed‐Loop App‐Based Interventions

Most eating disorder apps lack personalization, delivering content in a one‐size‐fits‐all manner that does not account for users' ongoing progress or evolving needs (O'Leary and Torous 2022). Yet, people with eating disorders consistently report a desire for digital health tools that are tailored to their unique circumstances (Linardon et al. 2021). While prior efforts to personalize eating disorder apps have relied exclusively on users' self‐reported symptom data (Juarascio et al. 2023; Tregarthen et al. 2019), advances in smartphone technology enable a more sophisticated and unobtrusive approach by incorporating passive sensor data to deliver tailored, closed‐loop interventions that continuously adapt in response to real‐time data (Huckvale et al. 2019).

To illustrate, consider the case of a patient using an eating disorder app with these capabilities over a month, responding to surveys about their thoughts, mood, and behavior, while also passively generating data on screen time, sleep patterns, and location. A sophisticated data‐driven algorithm can establish the baseline of this person using a combination of active and passive inputs. The algorithm may then be able to achieve real‐time assessment of heightened risk of relapse according to anomalies in these data streams. In response, the app can inform the user of this risk and deliver tailored intervention content just‐in‐time to prevent relapse. For example, in one instance an app may be able to infer an urge to binge eat in the context of elevated screen time on Instagram and respond by delivering a body image coping strategy. In another instance, it may infer the same risk from disrupted sleep patterns and close proximity to fast‐food outlets and respond by delivering a problem‐solving exercise.

Incorporating AI‐driven algorithms derived from smartphone sensor data to inform the development of closed‐loop health interventions is feasible (Wang et al. 2023), but it is not without its challenges. One key challenge lies in generating accurate and generalizable algorithms capable of reliably predicting relapse or symptom risk; current machine learning models often suffer from overfitting, lack external validation, and demonstrate variable predictive performance across different populations and contexts (Melcher et al. 2020). Robust and generalizable algorithms typically require data from thousands to tens of thousands of individuals (Barnett et al. 2020), meaning that many existing eating disorder datasets may be insufficiently powered to support the development of clinically useful prediction models. Another challenge is the practical difficulty of implementing just‐in‐time adaptive interventions, which rely on continuous, high‐quality data streams and precise timing of intervention delivery—factors that are hard to achieve outside of controlled research settings. Furthermore, the lack of standardized methods for collecting, analyzing, and reporting smartphone sensor data limits interoperability, transparency, and reproducibility across studies and platforms (Vaidyam et al. 2019; Linardon et al. 2025a). Before these closed‐loop approaches can become a clinical reality in eating disorder care, foundational research is needed on the systematic collection of multimodal app‐based data, the development and validation of robust predictive algorithms, and the integration of these models into user‐centered app designs that are feasible, acceptable, and capable of functioning effectively in real‐world environments.

5. Implementation Challenges

Smartphone technology may augment the assessment and treatment of eating disorders through capabilities such as digital phenotyping, in‐built NLP algorithms, and delivery of closed‐loop interventions (summarized in Table 1). However, despite the clear benefits of scalability, accessibility, and reach, implementing these digital tools in clinical practice remains challenging due to several factors that warrant careful consideration in future research.

TABLE 1.

Summary of capabilities of smartphone technology for eating disorder care.

Capability Data source Opportunity Challenges Research priority
Passive sensing and digital phenotyping Smartphone sensors (e.g., GPS, accelerometer, screen time, keystroke dynamics, microphone, call/text logs) Enable real‐time, unobtrusive monitoring of patterns to detect symptoms/relapse. Offers a low‐burden method to personalize assessment and support, even outside clinical settings. Lack of standardized methods for collecting and reporting passive data; variability across devices and platforms; and concerns around user privacy and data governance Identify which passive data features are most predictive of eating disorder symptoms; validate digital phenotypes across diverse populations; and develop ethical frameworks for secure and transparent data use.
Natural language processing (NLP) Textual data from in‐app exercises, chat interactions, self‐monitoring logs, or social media posts. Analyze language patterns to assess task quality, predict treatment outcomes, classify symptom severity, and detect cognitive or emotional shifts. May support personalized care through deeper insight into user experience and skill use. Variability in user input quality; limited generalizability of existing models; ethical concerns around interpretation of sensitive text; and underuse of NLP in eating disorder research. Develop and validate NLP models tailored to eating disorder contexts; identify linguistic markers of treatment response and relapse risk; and leverage existing app trial data to explore feasibility and effectiveness.
Closed‐loop adaptive interventions Combined active (e.g., EMA surveys, self‐reports) and passive (e.g., sensor data, app usage patterns) inputs. Deliver timely, personalized intervention content that dynamically adapts to fluctuations in user risk or clinical needs. Enhances treatment precision and responsiveness, potentially improving outcomes and engagement. Need for robust, validated algorithms; technical complexity of integrating multimodal data streams; limited evidence in eating disorder populations; barriers to clinical integration and user acceptability. Develop and test predictive models to inform just‐in‐time interventions; evaluate feasibility and efficacy in real‐world settings; and co‐design adaptive systems with end users to ensure acceptability and usability.
Open in a new tab

The first challenge concerns low engagement. In clinical trials, where participants knowingly sign up to receive access to an app, up to 15% never even download the app, while as many as 80% fail to complete all prescribed modules (Linardon et al. 2023). Engagement patterns are even lower in naturalistic settings, including for the most popular marketplace apps like Recovery Record (Lindgreen et al. 2021). Several factors may contribute to poor engagement, including the egosyntonic nature of eating disorder symptoms (e.g., low body weight, dietary restraint, excessive exercise), which can diminish motivation to engage with interventions (Juarascio et al. 2015), alongside skepticism toward AI technologies and fatigue from continuous self‐monitoring (Linardon et al. 2025b). This presents a major barrier to clinical utility, as even the most advanced features like in‐built personalization algorithms are unlikely to produce any benefits if users disengage early. One promising approach to improving uptake and sustained use is co‐design, actively involving people with lived experience of eating disorders in the development process to ensure that interventions are perceived as acceptable, relevant, and empowering (Torous et al. 2018). Engagement is a broader issue faced across fields of mental health, and strategies to address this are being tested with promising pilot results, including digital navigators embedded within hybrid models of care (Perret et al. 2023), gamification principles (Looyestyn et al. 2017), peer support components (Fortuna et al. 2019), and personalized push notifications (Bell et al. 2023). The eating disorder field would also benefit from identifying and testing novel engagement strategies.

A second challenge to widespread implementation of digital health tools concerns trust. Trust from both users and clinicians around mental health apps remains low, particularly in relation to data privacy and security (Torous et al. 2021). Data sharing and privacy are also among the most frequently cited concerns raised among people with eating disorders, contributing to the non‐uptake of digital health devices (Linardon et al. 2021). These concerns are warranted, as fewer than 15% of marketplace eating disorder apps contain a privacy policy that meets the standards of the Health Insurance Portability and Accountability Act (O'Leary and Torous 2022). Additional ethical challenges include difficulties ensuring compliance with international data protection regulations, lack of transparency in AI decision‐making, and risks of algorithmic bias that may lead to misdiagnosis or inappropriate clinical recommendations (Torous et al. 2019a). Limited clinical validation and a lack of long‐term efficacy data may also contribute to clinician skepticism and reluctance to incorporate digital tools into existing treatment frameworks (Torous et al. 2018). Moving the field forward will require greater transparency in data practices, stronger regulatory oversight, and the integration of user‐centered privacy protections into app design. Building trust through clear, accessible privacy policies and ethical data governance may prove critical for improving uptake and engagement.

A third challenge to implementation efforts relates to potential harms associated with digital health devices. While adverse events are rarely and poorly reported in digital health trials (Linardon et al. 2024b), evidence from both research and real‐world settings suggests they are not absent among people with eating disorders (Sharp et al. 2023). The dozens of eating disorder apps available for download in a largely unregulated market have promoted concerns about the quality of content offered, with content syntheses reporting that most of these apps do not follow evidence‐based protocols and, at times, may even offer unsafe information (Fairburn and Rothwell 2015). Greater regulatory oversight is clearly needed to ensure that eating disorder apps adhere to evidence‐based guidelines and safeguard users from potential harm. As an interim step, users and clinicians should be encouraged to consult curated resources such as https://mindapps.org/, which reviews apps based on privacy standards and alignment with clinical evidence (Camacho et al. 2022). In parallel, researchers conducting clinical trials should routinely assess and report adverse events to improve our understanding of their nature, extent, and contributing factors.

A fourth challenge relates to the technical and systemic complexities associated with integrating AI and smartphone technologies into clinical care. Data related to eating disorders can be biased or incomplete, which limits the development of reliable predictive models. For instance, biases commonly emerge when samples overrepresent young, White, female participants (Burnette et al. 2022), while incompleteness is also frequent given intermittent engagement with self‐tracking tools and selective self‐reporting of symptoms (Linardon et al. 2020). Additionally, eating disorder behaviors are dynamic and tend to evolve over time (Smith et al. 2019), meaning that static models trained on past data may quickly become outdated or misaligned with an individual's current symptom profile, thereby reducing predictive accuracy. Beyond technical issues, integration with existing healthcare systems presents additional barriers, including interoperability challenges, disruptions to clinical workflows, and regulatory hurdles (Torous et al. 2025). Addressing these complexities will be critical to realizing the full potential of AI‐enabled smartphone tools for eating disorder care.

A fifth challenge concerns maintaining the human element in mental healthcare. Although digital phenotyping, AI, and smartphone‐based tools offer promising avenues for supporting individuals with eating disorders, there is a risk that overreliance on technology could erode the therapeutic relationship (Blease et al. 2024). The empathic and interpersonal dimensions of psychotherapy are foundational for addressing the emotional distress and complex psychological needs of individuals with eating disorders (Graves et al. 2017), and de‐emphasizing these elements could introduce new risks in mental health. Additionally, promoting self‐sufficiency through AI‐based apps may unintentionally reduce user autonomy or foster social isolation. Algorithmic decision‐making systems may filter intervention options based on pre‐set parameters, narrowing the spectrum of available choices and potentially limiting personalization. Ensuring transparency in how such recommendations are generated will be critical to preserving user agency and trust. Future digital innovations must be designed to complement, rather than replace, human connection, ensuring that technological advances enhance—rather than undermine—therapeutic engagement.

6. Conclusion

Smartphone technology holds considerable potential to enhance the assessment, detection, and treatment of eating disorders through capabilities such as digital phenotyping, NLP, and closed‐loop adaptive interventions. As it stands, the eating disorder field has not kept pace with the current enthusiasm surrounding these tools and the innovations found in other fields of mental health like depression, anxiety, and schizophrenia. Realizing the full potential of smartphone technology will require overcoming key challenges related to engagement, trust, data governance, and clinical integration. The many testable research questions outlined in this forum offer a clear roadmap for advancing the field. Addressing these questions through methodologically rigorous, collaborative efforts will be essential to moving this work forward and addressing the unmet needs of people with eating disorders.

Author Contributions

Jake Linardon: conceptualization, writing – original draft, writing – review and editing. John Torous: writing – original draft, conceptualization, writing – review and editing.

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgments

Open access publishing facilitated by Deakin University, as part of the Wiley ‐ Deakin University agreement via the Council of Australian University Librarians.

Linardon, J. , and Torous J.. 2025. “Integrating Artificial Intelligence and Smartphone Technology to Enhance Personalized Assessment and Treatment for Eating Disorders.” International Journal of Eating Disorders 58, no. 8: 1415–1424. 10.1002/eat.24468.

Action Editor: Ruth Striegel Weissman

Funding: This study was supported by J.L. holds a National Health and Medical Research Council Grant (APP1196948).

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

References

  1. Anastasiadou, D. , Folkvord F., Brugnera A., et al. 2020. “An MHealth Intervention for the Treatment of Patients With an Eating Disorder: A Multicenter Randomized Controlled Trial.” International Journal of Eating Disorders 53, no. 7: 1120–1131. [DOI] [PubMed] [Google Scholar]
  2. Anderson, C. , Fuller‐Tyszkiewicz M., Messer M., et al. 2024. “Target‐user Preferences, Motivations, and Acceptance for a Dialectical Behaviour Therapy Smartphone Application for Eating Disorders.” Eating and Weight Disorders 29: 17. 10.1007/s40519-024-01646-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barnett, I. , Torous J., Reeder H. T., Baker J., and Onnela J.‐P.. 2020. “Determining Sample Size and Length of Follow‐Up for Smartphone‐Based Digital Phenotyping Studies.” Journal of the American Medical Informatics Association 27: 1844–1849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bell, L. , Garnett C., Bao Y., et al. 2023. “How Notifications Affect Engagement With a Behavior Change App: Results From a Micro‐Randomized Trial.” JMIR mHealth and uHealth 11: e38342. 10.2196/38342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Blease, C. , Worthen A., and Torous J.. 2024. “Psychiatrists' Experiences and Opinions of Generative Artificial Intelligence in Mental Healthcare: An Online Mixed Methods Survey.” Psychiatry Research 333: 115724. 10.1016/j.psychres.2024.115724. [DOI] [PubMed] [Google Scholar]
  6. Buck, B. , Munson J., Chander A., et al. 2022. “The Relationship Between Appraisals of Auditory Verbal Hallucinations and Real‐Time Affect and Social Functioning.” Schizophrenia Research 250: 112–119. 10.1016/j.schres.2022.10.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Buck, B. , Scherer E., Brian R., et al. 2019. “Relationships Between Smartphone Social Behavior and Relapse in Schizophrenia: A Preliminary Report.” Schizophrenia Research 208: 167–172. 10.1016/j.schres.2019.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bulik, C. M. , Butner J. E., Tregarthen J., et al. 2020. “The Binge Eating Genetics Initiative (BEGIN): Study Protocol.” BMC Psychiatry 20: 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Burnette, C. B. , Luzier J. L., Weisenmuller C. M., and Boutté R. L.. 2022. “A Systematic Review of Sociodemographic Reporting and Representation in Eating Disorder Psychotherapy Treatment Trials in the United States.” International Journal of Eating Disorders 55: 423–454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Burns, J. , Chen K., Flathers M., et al. 2024. “Transforming Digital Phenotyping Raw Data Into Actionable Biomarkers, Quality Metrics, and Data Visualizations Using Cortex Software Package: Tutorial.” Journal of Medical Internet Research 26: e58502. 10.2196/58502. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Butner, J. E. , Thornton L. M., Kilshaw R. E., et al. 2025. “Distinct Patterns of Dynamical Regulation in Passive Sensor Data Following Binge and Purge Behaviors.” International Journal of Eating Disorders 58, no. 7: 1296–1306. 10.1002/eat.24437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Camacho, E. , Cohen A., and Torous J.. 2022. “Assessment of Mental Health Services Available Through Smartphone Apps.” JAMA Network Open 5, no. 12: e2248784. 10.1001/jamanetworkopen.2022.48784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cao, J. , Truong A. L., Banu S., Shah A. A., Sabharwal A., and Moukaddam N.. 2020. “Tracking and Predicting Depressive Symptoms of Adolescents Using Smartphone‐Based Self‐Reports, Parental Evaluations, and Passive Phone Sensor Data: Development and Usability Study.” JMIR Ment Health 7: e14045. 10.2196/14045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Chowdhary, K. , and Chowdhary K.. 2020. Natural Language Processing: Fundamentals of Artificial Intelligence. Springer. [Google Scholar]
  15. Clausen, L. , Lübeck M., and Jones A.. 2013. “Motivation to Change in the Eating Disorders: A Systematic Review.” International Journal of Eating Disorders 46, no. 8: 755–763. [DOI] [PubMed] [Google Scholar]
  16. Cohen, A. , Naslund J., Lane E., et al. 2024. “Digital Phenotyping Data and Anomaly Detection Methods to Assess Changes in Mood and Anxiety Symptoms Across a Transdiagnostic Clinical Sample.” Acta Psychiatrica Scandinavica 151: 388–400. 10.1111/acps.13712. [DOI] [PubMed] [Google Scholar]
  17. Currey, D. , and Torous J.. 2023. “Digital Phenotyping Data to Predict Symptom Improvement and Mental Health App Personalization in College Students: Prospective Validation of a Predictive Model.” Journal of Medical Internet Research 25: e39258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. DaSilva, A. W. , Huckins J. F., Wang R., Wang W., Wagner D. D., and Campbell A. T.. 2019. “Correlates of Stress in the College Environment Uncovered by the Application of Penalized Generalized Estimating Equations to Mobile Sensing Data.” JMIR mHealth and uHealth 7: e12084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Depp, C. A. , Bashem J., Moore R. C., et al. 2019. “GPS Mobility as a Digital Biomarker of Negative Symptoms in Schizophrenia: A Case Control Study.” NPJ Digital Medicine 2, no. 1: 108. 10.1038/s41746-019-0182-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Doryab, A. , Villalba D. K., Chikersal P., et al. 2019. “Identifying Behavioral Phenotypes of Loneliness and Social Isolation With Passive Sensing: Statistical Analysis, Data Mining and Machine Learning of Smartphone and Fitbit Data.” JMIR mHealth and uHealth 7, no. 7: e13209. 10.2196/13209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. dos Santos, M. P. , Heckler W. F., Bavaresco R. S., and Barbosa J. L. V.. 2024. “Machine Learning Applied to Digital Phenotyping: A Systematic Literature Review and Taxonomy.” Computers in Human Behavior 161: 108422. 10.1016/j.chb.2024.108422. [DOI] [Google Scholar]
  22. Fairburn, C. G. , and Rothwell E. R.. 2015. “Apps and Eating Disorders: A Systematic Clinical Appraisal.” International Journal of Eating Disorders 48: 1038–1046. 10.1002/eat.22398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Faurholt‐Jepsen, M. , Busk J., Rohani D. A., et al. 2022. “Differences in Mobility Patterns According to Machine Learning Models in Patients With Bipolar Disorder and Patients With Unipolar Disorder.” Journal of Affective Disorders 306: 246–253. 10.1016/j.jad.2022.03.054. [DOI] [PubMed] [Google Scholar]
  24. Fortuna, K. L. , Brooks J. M., Umucu E., Walker R., and Chow P. I.. 2019. “Peer Support: A Human Factor to Enhance Engagement in Digital Health Behavior Change Interventions.” Journal of Technology in Behavioral Science 4: 152–161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Fuller‐Tyszkiewicz, M. , Messer M., Krug I., and Linardon J.. 2024. “Digital Health Applications for Eating Disorders Treatment.” Trends in Molecular Medicine 30: 324–326. 10.1016/j.molmed.2023.11.004. [DOI] [PubMed] [Google Scholar]
  26. Furukawa, T. A. , Iwata S., Horikoshi M., et al. 2023. “Harnessing AI to Optimize Thought Records and Facilitate Cognitive Restructuring in Smartphone CBT: An Exploratory Study.” Cognitive Therapy and Research 47, no. 6: 887–893. 10.1007/s10608-023-10411-7. [DOI] [Google Scholar]
  27. Graves, T. A. , Tabri N., Thompson‐Brenner H., et al. 2017. “A Meta‐Analysis of the Relation Between Therapeutic Alliance and Treatment Outcome in Eating Disorders.” International Journal of Eating Disorders 50: 323–340. 10.1002/eat.22672. [DOI] [PubMed] [Google Scholar]
  28. Haedt‐Matt, A. A. , and Keel P. K.. 2011. “Revisiting the Affect Regulation Model of Binge Eating: A Meta‐Analysis of Studies Using Ecological Momentary Assessment.” Psychological Bulletin 137: 660–681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Hildebrandt, T. , Michaeledes A., Mayhew M., et al. 2020. “Randomized Controlled Trial Comparing Health Coach‐Delivered Smartphone‐Guided Self‐Help With Standard Care for Adults With Binge Eating.” American Journal of Psychiatry 177: 134–142. 10.1176/appi.ajp.2019.19020184. [DOI] [PubMed] [Google Scholar]
  30. Huckvale, K. , Venkatesh S., and Christensen H.. 2019. “Toward Clinical Digital Phenotyping: A Timely Opportunity to Consider Purpose, Quality, and Safety.” NPJ Digital Medicine 2: 1–11. 10.1038/s41746-019-0166-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Jacobson, N. C. , Summers B., and Wilhelm S.. 2020. “Digital Biomarkers of Social Anxiety Severity: Digital Phenotyping Using Passive Smartphone Sensors.” Journal of Medical Internet Research 22: e16875. 10.2196/16875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Juarascio, A. S. , Manasse S. M., Goldstein S. P., Forman E. M., and Butryn M. L.. 2015. “Review of Smartphone Applications for the Treatment of Eating Disorders.” European Eating Disorders Review 23: 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Juarascio, A. S. , Presseller E. K., Srivastava P., Manasse S. M., and Forman E. M.. 2023. “A Randomized Controlled Trial of CBT+: A Clinician‐Controlled, Just‐In‐Time, Adjunctive Intervention for Bulimia‐Spectrum Disorders.” Behavior Modification 47: 551–572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Juarascio, A. S. , Srivastava P., Presseller E. K., et al. 2022. “Using Continuous Glucose Monitoring to Detect and Intervene on Dietary Restriction in Individuals With Binge Eating: The SenseSupport Withdrawal Design Study.” JMIR Formative Research 6: e38479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kazdin, A. E. , Fitzsimmons‐Craft E. E., and Wilfley D. E.. 2017. “Addressing Critical Gaps in the Treatment of Eating Disorders.” International Journal of Eating Disorders 50: 170–189. 10.1002/eat.22670. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Linardon, J. 2025. “Navigating the Future of Psychiatry: A Review of Research on Opportunities, Applications, and Challenges of Artificial Intelligence.” Current Treatment Options in Psychiatry 12: 8. 10.1007/s40501-025-00344-1. [DOI] [Google Scholar]
  37. Linardon, J. , Anderson C., McClure Z., et al. 2024a. “A Dialectical Behavior Therapy Skills Training Smartphone App for Recurrent Binge Eating: A Randomized Clinical Trial.” Psychological Medicine 54, no. 16: 1–12. 10.1017/S0033291724002800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Linardon, J. , Chen K., Gajjar S., et al. 2025a. “Smartphone Digital Phenotyping in Mental Health Disorders: A Review of Raw Sensors Utilized, Machine Learning Processing Pipelines, and Derived Behavioral Features.” Psychiatry Research 348: 116483. 10.1016/j.psychres.2025.116483. [DOI] [PubMed] [Google Scholar]
  39. Linardon, J. , and Fuller‐Tyszkiewicz M.. 2025. “Using Artificial Intelligence to Advance Eating Disorder Research, Treatment and Practice.” International Journal of Eating Disorders 58, no. 5: 811–812. 10.1002/eat.24394. [DOI] [PubMed] [Google Scholar]
  40. Linardon, J. , Fuller‐Tyszkiewicz M., Firth J., et al. 2024b. “Systematic Review and Meta‐Analysis of Adverse Events in Clinical Trials of Mental Health Apps.” NPJ Digital Medicine 7: 363. 10.1038/s41746-024-01388-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Linardon, J. , Liu C., Messer M., McClure Z., Anderson C., and Jarman H. K.. 2025b. “Current Practices and Perspectives of Artificial Intelligence in the Clinical Management of Eating Disorders: Insights From Clinicians and Community Participants.” International Journal of Eating Disorders 58: 724–734. 10.1002/eat.24385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Linardon, J. , Messer M., Lee S., and Rosato J.. 2021. “Perspectives of e‐Health Interventions for Treating and Preventing Eating Disorders: Descriptive Study of Perceived Advantages and Barriers, Help‐Seeking Intentions, and Preferred Functionality.” Eating and Weight Disorders 26: 1097–1109. 10.1007/s40519-020-01005-3. [DOI] [PubMed] [Google Scholar]
  43. Linardon, J. , Messer M., Shatte A., et al. 2022a. “Does the Method of Content Delivery Matter? Randomized Controlled Comparison of an Internet‐Based Intervention for Eating Disorder Symptoms With and Without Interactive Functionality.” Behavior Therapy 53, no. 3: 508–520. 10.1016/j.beth.2021.12.001. [DOI] [PubMed] [Google Scholar]
  44. Linardon, J. , Shatte A., McClure Z., and Fuller‐Tyszkiewicz M.. 2023. “A Broad v. Focused Digital Intervention for Recurrent Binge Eating: A Randomized Controlled Non‐Inferiority Trial.” Psychological Medicine 53: 4580–4591. 10.1017/S0033291722001477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Linardon, J. , Shatte A., Messer M., Firth J., and Fuller‐Tyszkiewicz M.. 2020. “E‐Mental Health Interventions for the Treatment and Prevention of Eating Disorders: An Updated Systematic Review and Meta‐Analysis.” Journal of Consulting and Clinical Psychology 88: 994–1007. 10.1037/ccp0000575. [DOI] [PubMed] [Google Scholar]
  46. Linardon, J. , Shatte A., Rosato J., and Fuller‐Tyszkiewicz M.. 2022b. “Efficacy of a Transdiagnostic Cognitive‐Behavioral Intervention for Eating Disorder Psychopathology Delivered Through a Smartphone App: A Randomized Controlled Trial.” Psychological Medicine 52: 1679–1690. 10.1017/S0033291720003426. [DOI] [PubMed] [Google Scholar]
  47. Lindgreen, P. , Lomborg K., and Clausen L.. 2021. “Patient Use of a Self‐Monitoring App During Eating Disorder Treatment: Naturalistic Longitudinal Cohort Study.” Brain and Behavior: A Cognitive Neuroscience Perspective 11: e02039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Liu, C. , and Linardon J.. 2025. “Mental Health Clinicians' Practices and Perspectives of Eating Disorder Apps.” Psychiatry Research 347: 116412. 10.1016/j.psychres.2025.116412. [DOI] [PubMed] [Google Scholar]
  49. Looyestyn, J. , Kernot J., Boshoff K., Ryan J., Edney S., and Maher C.. 2017. “Does Gamification Increase Engagement With Online Programs? A Systematic Review.” PLoS One 12: e0173403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Macrynikola, N. , Chen K., Lane E., et al. 2025. “Testing the Feasibility, Acceptability, and Potential Efficacy of an Innovative Digital Mental Health Care Delivery Model Designed to Increase Access to Care: Open Trial of the Digital Clinic.” JMIR Mental Health 12: e65222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Mason, T. B. , Lavender J. M., Wonderlich S. A., et al. 2018. “Examining a Momentary Mediation Model of Appearance‐Related Stress, Anxiety, and Eating Disorder Behaviors in Adult Anorexia Nervosa.” Eating and Weight Disorders ‐ Studies on Anorexia, Bulimia and Obesity 23: 637–644. 10.1007/s40519-017-0404-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Melcher, J. , Hays R., and Torous J.. 2020. “Digital Phenotyping for Mental Health of College Students: A Clinical Review.” BMJ Ment Health 23: 161–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Moskowitz, D. S. , and Young S. N.. 2006. “Ecological Momentary Assessment: What It Is and Why It Is a Method of the Future in Clinical Psychopharmacology.” Journal of Psychiatry and Neuroscience 31: 13–20. [PMC free article] [PubMed] [Google Scholar]
  54. Mylläri, S. , Saarni S. E., Ritola V., et al. 2022. “Text Topics and Treatment Response in Internet‐Delivered Cognitive Behavioral Therapy for Generalized Anxiety Disorder: Text Mining Study.” Journal of Medical Internet Research 24, no. 11: e38911. 10.2196/38911. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Neumayr, C. , Voderholzer U., Tregarthen J., and Schlegl S.. 2019. “Improving Aftercare With Technology for Anorexia Nervosa After Intensive Inpatient Treatment: A Pilot Randomized Controlled Trial With a Therapist‐Guided Smartphone App.” International Journal of Eating Disorders 52: 1191–1201. 10.1002/eat.23152. [DOI] [PubMed] [Google Scholar]
  56. O'Leary, T. , and Torous J.. 2022. “Smartphone Apps for Eating Disorders: An Overview of the Marketplace and Research Trends.” International Journal of Eating Disorders 55: 625–632. [DOI] [PubMed] [Google Scholar]
  57. Onnela, J.‐P. , and Rauch S. L.. 2016. “Harnessing Smartphone‐Based Digital Phenotyping to Enhance Behavioral and Mental Health.” Neuropsychopharmacology 41: 1691–1696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Perret, S. , Alon N., Carpenter‐Song E., et al. 2023. “Standardising the Role of a Digital Navigator in Behavioural Health: A Systematic Review.” Lancet Digital Health 5, no. 12: e925–e932. 10.1016/S2589-7500(23)00152-8. [DOI] [PubMed] [Google Scholar]
  59. Pew Research Center . 2024. “Americans' Use of Mobile Technology and Home Broadband.” https://www.pewresearch.org/internet/2024/01/31/americans‐use‐of‐mobile‐technology‐and‐home‐broadband/.
  60. Presseller, E. K. , Patarinski A. G. G., Fan S. C., Lampe E. W., and Juarascio A. S.. 2022a. “Sensor Technology in Eating Disorders Research: A Systematic Review.” International Journal of Eating Disorders 55: 573–624. [DOI] [PubMed] [Google Scholar]
  61. Presseller, E. K. , Patarinski A. G. G., Zhang F., et al. 2022b. “Glucose Variability: A Physiological Correlate of Eating Disorder Behaviors Among Individuals With Binge‐Spectrum Eating Disorders.” International Journal of Eating Disorders 55: 1788–1798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Ralph‐Nearman, C. , Sandoval‐Araujo L., Karem A., et al. 2024. “Using Machine Learning With Passive Wearable Sensors to Pilot the Detection of Eating Disorder Behaviors in Everyday Life.” Psychological Medicine 54: 1084–1090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Schaefer, L. M. , Engel S. G., and Wonderlich S. A.. 2020. “Ecological Momentary Assessment in Eating Disorders Research: Recent Findings and Promising New Directions.” Current Opinion in Psychiatry 33: 528–533. 10.1097/yco.0000000000000639. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Sharp, G. , Torous J., and West M. L.. 2023. Ethical Challenges in AI Approaches to Eating Disorders. Vol. 25, e50696. JMIR Publications. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Shiffman, S. , Stone A. A., and Hufford M. R.. 2008. “Ecological Momentary Assessment.” Annual Review of Clinical Psychology 4: 1–32. [DOI] [PubMed] [Google Scholar]
  66. Smith, K. E. , Mason T. B., Juarascio A., et al. 2019. “Moving Beyond Self‐Report Data Collection in the Natural Environment: A Review of the Past and Future Directions for Ambulatory Assessment in Eating Disorders.” International Journal of Eating Disorders 52: 1157–1175. 10.1002/eat.23124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Solmi, M. , Monaco F., Højlund M., et al. 2024. “Outcomes in People With Eating Disorders: A Transdiagnostic and Disorder‐Specific Systematic Review, Meta‐Analysis and Multivariable Meta‐Regression Analysis.” World Psychiatry 23, no. 1: 124–138. 10.1002/wps.21182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Srivastava, P. , Lampe E. W., Michael M. L., Manasse S., and Juarascio A. S.. 2021. “Stress Appraisal Prospectively Predicts Binge Eating Through Increases in Negative Affect.” Eating and Weight Disorders – Studies on Anorexia, Bulimia and Obesity 26: 2413–2420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Torous, J. , Andersson G., Bertagnoli A., et al. 2019a. “Towards a Consensus Around Standards for Smartphone Apps and Digital Mental Health.” World Psychiatry 18, no. 1: 97–98. 10.1002/wps.20592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Torous, J. , Bucci S., Bell I. H., et al. 2021. “The Growing Field of Digital Psychiatry: Current Evidence and the Future of Apps, Social Media, Chatbots, and Virtual Reality.” World Psychiatry 20, no. 3: 318–335. 10.1002/wps.20883. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Torous, J. , Kiang M. V., Lorme J., and Onnela J.‐P.. 2016. “New Tools for New Research in Psychiatry: A Scalable and Customizable Platform to Empower Data Driven Smartphone Research.” JMIR Mental Health 3: e5165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Torous, J. , Linardon J., Goldberg S. B., et al. 2025. “The Evolving Field of Digital Mental Health: Current Evidence and Implementation Issues for Smartphone Apps, Generative Artificial Intelligence, and Virtual Reality.” World Psychiatry 24: 156–174. 10.1002/wps.21299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Torous, J. , Nicholas J., Larsen M. E., Firth J., and Christensen H.. 2018. “Clinical Review of User Engagement With Mental Health Smartphone Apps: Evidence, Theory and Improvements.” Evidence‐Based Mental Health 21: 116–119. 10.1136/eb-2018-102891. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Torous, J. , Wisniewski H., Bird B., et al. 2019b. “Creating a Digital Health Smartphone App and Digital Phenotyping Platform for Mental Health and Diverse Healthcare Needs: An Interdisciplinary and Collaborative Approach.” Journal of Technology in Behavioral Science 4, no. 2: 73–85. 10.1007/s41347-019-00095-w. [DOI] [Google Scholar]
  75. Tregarthen, J. , Kim J. P., Sadeh‐Sharvit S., Neri E., Welch H., and Lock J.. 2019. “Comparing a Tailored Self‐Help Mobile App With a Standard Self‐Monitoring App for the Treatment of Eating Disorder Symptoms: Randomized Controlled Trial.” JMIR Mental Health 6: e14972. 10.2196/14972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Vaidyam, A. , Halamka J., and Torous J.. 2019. “Actionable Digital Phenotyping: A Framework for the Delivery of Just‐In‐Time and Longitudinal Interventions in Clinical Healthcare.” Mhealth 5: 25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Wang, J. , Wu Z., Choi S. W., et al. 2023. “The Dosing of Mobile‐Based Just‐In‐Time Adaptive Self‐Management Prompts for Caregivers: Preliminary Findings From a Pilot Microrandomized Study.” JMIR Formative Research 7: e43099. 10.2196/43099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Weissman, R. S. , and Rosselli F.. 2017. “Reducing the Burden of Suffering From Eating Disorders: Unmet Treatment Needs, Cost of Illness, and the Quest for Cost‐Effectiveness.” Behaviour Research and Therapy 88: 49–64. 10.1016/j.brat.2016.09.006. [DOI] [PubMed] [Google Scholar]
  79. Zhang, Y. , Folarin A. A., Sun S., et al. 2022. “Longitudinal Relationships Between Depressive Symptom Severity and Phone‐Measured Mobility: Dynamic Structural Equation Modeling Study.” JMIR Mental Health 9, no. 3: e34898. 10.2196/34898. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

Articles from The International Journal of Eating Disorders are provided here courtesy of Wiley

RESOURCES