Personalization strategies in digital mental health interventions: a systematic review and conceptual framework for depressive symptoms

Predicting recurrent chat contact in a psychological intervention for the youth using natural language processing

Chat-based counseling hotlines emerged as a promising low-threshold intervention for youth mental health. However, despite the resulting availability of large text corpora, little work has investigated Natural Language Processing (NLP) applications within this setting. Therefore, this preregistered approach (OSF: XA4PN) utilizes a sample of approximately 19,000 children and young adults that received a chat consultation from a 24/7 crisis service in Germany. Around 800,000 messages were used to predict whether chatters would contact the service again, as this would allow the provision of or redirection to additional treatment. We trained an XGBoost Classifier on the words of the anonymized conversations, using repeated cross-validation and bayesian optimization for hyperparameter search. The best model was able to achieve an AUROC score of 0.68 (p < 0.01) on the previously unseen 3942 newest consultations. A shapely-based explainability approach revealed that words indicating younger age or female gender and terms related to self-harm and suicidal thoughts were associated with a higher chance of recontacting. We conclude that NLP-based predictions of recurrent contact are a promising path toward personalized care at chat hotlines.

Evaluation of completely online psychotherapy with app-support versus therapy as usual for clients with depression or anxiety disorder: A retrospective matched cohort study investigating the effectiveness, efficiency, client satisfaction, and costs

INTRODUCTION

Depressive and anxiety disorders are common mental disorders ranking among the leading causes of global disease burden. Not all clients currently benefit from therapy and clients are looking for modern ways of therapy. Online psychotherapy is a promising option for better meeting clients' needs. Recently, a new psychotherapy concept has emerged that combines videoconferencing sessions with support through a mobile application. The latter allows for ecological momentary assessments and interventions, facilitates communication between patients and therapists in between sessions through chat, and allows for incorporating feedback-informed treatment principles.

MATERIAL AND METHODS

The study was a retrospective observational matched cohort study, comparing online psychotherapy with Therapy As Usual (TAU) for clients with depressive or anxiety disorders. Data were obtained via questionnaires, which are part of standard clinical care. Primary outcomes included general mental functioning, and symptoms of depression and anxiety. Secondary outcomes were efficiency, client satisfaction, and therapy costs. Primary endpoints were analyzed using linear mixed models analysis, with an interaction term between time and group. Secondary outcomes were analyzed using linear regression.

RESULTS

Larger improvements were observed in the online compared to the TAU group for general mental functioning and depressive disorder (i.e., General mental functioning: B = -8.50, 95 CI: -15.01 - -1.97, p = 0.011; Depressive disorder: B = - 3.66, 95 % CI: -5.79 - -1.54p < 0.01). No significant differences in change over time between the two groups were observed for anxiety disorder (B = -3.64, 95 % CI: (-13.10 - 5.82) p = 0.447). The total number of sessions was significantly higher in the online psychotherapy group than in TAU (B = 3.71, p < 0.01), although clients were matched on treatment time in weeks. Treatment session duration in minutes was comparable across the groups.

DISCUSSION

Online psychotherapy with app support showed to be a promising alternative to TAU for depressive and anxiety disorders. More research is needed to evaluate the effectiveness, cost-effectiveness and client satisfaction of online psychotherapy compared to TAU, such as randomized controlled trials or studies multiple baseline series designs, and in-depth qualitative research.

Defining and operationalizing personalized psychological treatment - a systematic literature review

This systematic literature review aimed to propose a definition of personalized psychological treatment and to suggest how the definition can be operationalized. PsycINFO, Cochrane Library, MEDLINE and EMBASE were searched up to 11 December 2023 for studies in which a definition of personalized psychological treatment was included or a systematic operationalization of personalized psychological treatment was described. Based on a narrative synthesis of the collected definitions, summary categories were developed that informed the proposed definition. Operationalizations were described according to what aspect of treatment, how and when treatment was personalized. The extent to which the operationalizations deviated from the proposed definition was assessed. Thirty-four studies with definitions and 200 with operationalizations were included. The following definition was proposed: personalized psychological treatment aims to optimize treatment outcome for the individual patient by tailoring treatment to unique or specific needs, preferences or other characteristics and includes a systematic adaptation of treatment or a differentiation between treatment strategies. Based on the operationalizations, timing of personalization, specification of the systematic approach and treatment elements that could be personalized were added to the proposed definition. Evidence-based personalization of psychological treatments can be enhanced by clear operationalization based on a comprehensive definition of personalization.

TONI - One for all? Participatory development of a transtheoretic and transdiagnostic online intervention for blended care

Background

Internet-based interventions offer a way to meet the high demand for psychological support. However, this setting also has disadvantages, such as the lack of personal contact and the limited ability to respond to crises. Blended care combines Internet-based interventions with face-to-face psychotherapy and merges the benefits of both settings. To ensure the uptake of blended care in routine care, Internet-based interventions need to be suitable for different therapeutic approaches and mental disorders.

Objective

This paper describes the participatory development process of the Internet-based intervention "TONI" using a common therapeutic language and content on various transdiagnostic topics to be integrated into routine outpatient psychotherapy.

Methods

To develop this intervention in a participatory manner, we followed the Integrate, Design, Assess, and Share (IDEAS) framework. In a multilevel development process, we used a combination of interviews, focus groups, and proofreading to optimally tailor online modules to routine outpatient psychotherapy. Building on well-established cognitive-behavioral online content, we included expert interviews with psychodynamic (n = 20) and systemic psychotherapists (n = 9) as well as focus groups with psychotherapists of different approaches (n = 10) and persons with lived experience of mental illness (PWLE; n = 10).

Results

We describe the development process of TONI step-by-step, outlining the specific requirements that therapists from different therapeutic approaches as well as PWLE have and how we implemented them in our intervention. This includes the content and specific exercises in the online modules, aspects of data protection, language, design, and usability.

Conclusion

Internet-based interventions that use a common therapeutic language and address therapeutic principles across different approaches have the potential to advance digitalization in psychotherapy. Involving psychotherapists and PWLE in intervention development may positively impact acceptance and usage in practice. This study shows how participatory intervention development involving both psychotherapists and PWLE can be carried out.

Effect of digital based nursing intervention on knowledge of self-care behaviors and self-efficacy of adult clients with diabetes

BACKGROUND

In recent years, there has been growing interest in the use of Digital Based Nursing Intervention to support diabetes management. This study aimed to evaluate the effect of digital based nursing intervention on knowledge of self-care behaviors and self-efficacy of clients with diabetes.

METHODS

Employing a quasi-experimental design, a sample of 120 adult participants diagnosed with type 2 diabetes, aged more than 18 years with focus on older adults was drawn from outpatient clinics at Cairo University Hospital. The intervention was approved and registered by the ethical committee of the faculty of nursing with IRB number: RHDIRB2019041701. The intervention group (n = 60) received a digital-based nursing intervention, while the control group (n = 60) received standard care. Data were collected using adopted standardized tools including the Diabetes Knowledge Test, the Diabetes Self-Efficacy Scale, and the Summary of Diabetes Self-Care Activities. Demographic characteristics were analyzed, and pre- and post-intervention scores were compared using paired t-tests were statistical methods.

RESULTS

The digital-based nursing intervention resulted in significant enhancements in participants with diabetes knowledge and self-efficacy levels. Moreover, the intervention group demonstrated marked improvements in various self-care behaviors encompassing diet, exercise, medication adherence, blood glucose testing, and foot care. While the control group also exhibited some progress, the effects were less pronounced. Regression analyses highlighted age as a consistent factor associated with knowledge, self-efficacy, and specific self-care behaviors.

CONCLUSION

This study underscores the potential of tailored digital nursing interventions to complement traditional care approaches, empowering patients with type 2 diabetes to actively engage in self-management. The findings suggest that digital-based nursing interventions hold promise for enhancing patient knowledge, confidence, and proactive health behaviors. Nevertheless, limitations, including the relatively short intervention duration and a sample from a single clinic, warrant consideration. Future research should address these limitations to bolster the validity and applicability of the study's conclusions.

The impact of using reinforcement learning to personalize communication on medication adherence: findings from the REINFORCE trial

Text messaging can promote healthy behaviors, like adherence to medication, yet its effectiveness remains modest, in part because message content is rarely personalized. Reinforcement learning has been used in consumer technology to personalize content but with limited application in healthcare. We tested a reinforcement learning program that identifies individual responsiveness ("adherence") to text message content and personalizes messaging accordingly. We randomized 60 individuals with diabetes and glycated hemoglobin A1c [HbA1c] ≥ 7.5% to reinforcement learning intervention or control (no messages). Both arms received electronic pill bottles to measure adherence. The intervention improved absolute adjusted adherence by 13.6% (95%CI: 1.7%-27.1%) versus control and was more effective in patients with HbA1c 7.5- < 9.0% (36.6%, 95%CI: 25.1%-48.2%, interaction p < 0.001). We also explored whether individual patient characteristics were associated with differential response to tested behavioral factors and unique clusters of responsiveness. Reinforcement learning may be a promising approach to improve adherence and personalize communication at scale.

The latest developments with internet-based psychological treatments for depression

INTRODUCTION

Internet-based psychological treatments for depression have been around for more than 20 years. There has been a continuous line of research with new research questions being asked and studies conducted.

AREAS COVERED

In this paper, the author reviews studies with a focus on papers published from 2020 and onwards based on a Medline and Scopus search. Internet-based cognitive behavior therapy (ICBT) programs have been developed and tested for adolescents, older adults, immigrant groups and to handle a societal crisis (e.g. COVID-19). ICBT works in regular clinical settings and long-term effects can be obtained. Studies on different treatment orientations and approaches such as acceptance commitment therapy, unified protocol, and tailored treatments have been conducted. Effects on quality-of-life measures, knowledge acquisition and ecological momentary assessment as a research tool have been reported. Factorial design trials and individual patient data meta-analysis are increasingly used in association with internet intervention research. Finally, prediction studies and recent advances in artificial intelligence are mentioned.

EXPERT OPINION

Internet-delivered treatments are effective, in particular if therapist guidance is provided. More target groups have been covered but there are many remaining challenges including how new tools like artificial intelligence will be used when treating depression.

Dataset size versus homogeneity: A machine learning study on pooling intervention data in e-mental health dropout predictions

Objective

This study proposes a way of increasing dataset sizes for machine learning tasks in Internet-based Cognitive Behavioral Therapy through pooling interventions. To this end, it (1) examines similarities in user behavior and symptom data among online interventions for patients with depression, social anxiety, and panic disorder and (2) explores whether these similarities suffice to allow for pooling the data together, resulting in more training data when prediction intervention dropout.

Methods

A total of 6418 routine care patients from the Internet Psychiatry in Stockholm are analyzed using (1) clustering and (2) dropout prediction models. For the latter, prediction models trained on each individual intervention's data are compared to those trained on all three interventions pooled into one dataset. To investigate if results vary with dataset size, the prediction is repeated using small and medium dataset sizes.

Results

The clustering analysis identified three distinct groups that are almost equally spread across interventions and are instead characterized by different activity levels. In eight out of nine settings investigated, pooling the data improves prediction results compared to models trained on a single intervention dataset. It is further confirmed that models trained on small datasets are more likely to overestimate prediction results.

Conclusion

The study reveals similar patterns of patients with depression, social anxiety, and panic disorder regarding online activity and intervention dropout. As such, this work offers pooling different interventions' data as a possible approach to counter the problem of small dataset sizes in psychological research.

E-therapists' views on the acceptability and feasibility of an internet-administered, guided, low-intensity cognitive behavioural therapy intervention for parents of children treated for cancer: A qualitative study

Background

Childhood cancer treatment completion can be a period of vulnerability for parents and is associated with mental health difficulties such as depression and anxiety. We developed an internet-administered, guided, low-intensity cognitive behavioural therapy-based self-help intervention (EJDeR) for parents delivered on the U-CARE-portal (Portal). The acceptability and feasibility of EJDeR and study procedures were examined using a single-arm feasibility trial (ENGAGE). Results indicated that EJDeR and ENGAGE study procedures are acceptable and feasible, however, a need for clinical and technical modifications to EJDeR and refinements to ENGAGE study procedures was identified.

Objectives

This study aimed to explore the acceptability and feasibility of EJDeR and ENGAGE study procedures from the perspective of e-therapists to inform clinical and technical modifications to EJDeR and refinements to study procedures prior to progression to a superiority randomised controlled trial.

Methods

We conducted semi-structured interviews with 10 e-therapists. Data were analysed using manifest content analysis.

Results

We identified three categories relating to the acceptability and feasibility of EJDeR: (a) Support to e-therapists (subcategories: Clinical supervision and Technical difficulties); (b) Guidance to parents (subcategories: Support protocols and Synchronous communication); and (c) Content (subcategories: Relevancy of the intervention and Pacing of the intervention). We identified four categories relating to the acceptability and feasibility of study procedures: (a) Recruitment and training of e-therapists (subcategories: Definition of the role and Training program); (b) Retention of parents (subcategories: Parent suitability and screening and Frequency of weekly Portal assessments); (c) Retention of e-therapists (subcategories: Administrative requirements and Communication with the research team); and (d) The Portal.

Conclusions

EJDeR and study procedures were considered acceptable and feasible, however, clinical and technical modifications and refinements to study procedures were suggested to enhance acceptability and feasibility. Results may also inform implementation considerations for both EJDeR and other similar digital psychological interventions.

Trial registration number

ISRCTN 57233429.

Barriers and facilitators to acceptance and implementation of eMental-health intervention among older adults: A qualitative systematic review

Objective

Electronic mental health interventions are effective but not well promoted currently among older adults. This study sought to systematically review and summarize the barriers and facilitators of accepting and implementing electronic mental health interventions among older adults.

Methods

We comprehensively retrieved six electronic databases from January 2012 to September 2022: PubMed, Web of Science, Embase, Scopus, PsycINFO, and CINAHL. The JBI-QARI was used to assess the quality of the research methodology of each publication. Eligible studies underwent data coding and synthesis aligned to inductive and deductive methods. The Consolidated Framework for Implementation Research 2.0 was used as a deductive framework to guide a more structured analysis.

Results

The systematic review screened 4309 articles, 17 of which were included (eight with mixed methods and nine with qualitative methods). We identified and extracted the barriers and facilitators of accepting and implementing electronic mental health interventions among older adults: (1) innovation: technology challenges, optimized functions, and contents, security and privacy; (2) outer setting: community engagement and partnerships, financing; (3) inner setting: leadership engagement, available resources, incompatibility, intergenerational support, training and guidance; (4) individuals: perceptions, capability, motivation of older adults and healthcare providers; and (5) implementation process: recruit, external assistance, and team.

Conclusion

These findings are critical to optimizing, promoting, and expanding electronic mental health interventions among older adults. The systematic review also provides a reference for better evidence-based implementation strategies in the future.

Exploring Generation Z and Young Millennials' Perspectives of a Spiritual Self-Care App and Their Spiritual Identity (Skylight): Qualitative Semistructured Interview Study

BACKGROUND

Generation Z and young millennials (ages 18-35 years), collectively referred to as GenZennials, are connected to technology and the internet like no other generation before them. This has mental health implications, such as increased rates of anxiety and stress. Recent research has shown that app-based mental health interventions can be useful to address such mental health concerns. However, spirituality is an untapped resource, especially since GenZennials largely identify as spiritual and already integrate spiritual practices into their self-care.

OBJECTIVE

There were four objectives to this study: (1) comprehensively explore reasons why GenZennials use a spiritual self-care app (ie, Skylight; Radiant Foundation), (2) understand how GenZennials identify spiritually, (3) understand the app's relevance to GenZennials, and (4) gather feedback and suggestions to improve the app.

METHODS

Semistructured interviews were conducted with 23 GenZennials (ages 18-35 years; mean 28.7, SD 5.0 years; n=20, 87% female) who used the Skylight app. Interviews were 30 to 60 minutes and conducted on Zoom. Thematic analysis was used to analyze interviews.

RESULTS

Five major categories emerged from the analysis, each encompassing one to several themes: (1) reasons for using the Skylight app, (2) content favorites, (3) defining spiritual identity, (4) relevance to GenZennials, and (5) overall improvement recommendations. Participants used the app for various reasons including to relax, escape, or ground themselves; improve mood; and enhance overall health and wellness. Participants also cited the app's variety of content offerings and its free accessibility as their primary reasons for using it. Most participants identified themselves as solely spiritual (8/23/35%) among the options provided (ie, spiritual or religious or both), and they appreciated the app's inclusive content. Participants felt that the app was relevant to their generation as it offered modern content (eg, spiritual self-care activities and short content). Participants recommended adding more personalization capabilities, content, and representation to the app.

CONCLUSIONS

This is the first study to qualitatively explore GenZennials' perspectives and the use of a spiritual self-care app. Our findings should inform the future creation and improvement of spiritual self-care apps aimed at cultivating GenZennials' spiritual and mental well-being. Future research is warranted to examine the effects of using a spiritual self-care app on GenZennial mental health.

Personalization in digital health interventions for older adults with cancer: A scoping review

INTRODUCTION

Digital health interventions (DHIs) are promising to support older adults with cancer in managing their conditions and improving their health outcomes. However, there is a lack of overall understanding of various DHIs for the aging population with cancer. Specifically, it is unclear how personalization components are included in those DHIs to promote engagement in the interventions among older adults with cancer. This study aimed to provide a comprehensive overview of existing DHIs for older adults with cancer and identify the intervention components, especially personalized features, and effectiveness of these DHIs for improving self-management and psychosocial health.

MATERIALS AND METHODS

A scoping review was conducted following Joanna Briggs Institute guidelines, focusing on older adults diagnosed with cancer who participated in DHIs to improve self-management and psychosocial health. Studies using an experimental design and published from 2000 to January 2023 were retrieved from four databases: PubMed, Embase, CINAHL, and Scopus. After primary data extraction of study characteristics, participants, interventions, and outcomes, DHIs were categorized according to personalized features.

RESULTS

Out of 9,750 articles, 20 were eligible for this scoping review. The main personalized features of DHIs were categorized into four domains: goal setting, adjusting the plan, data-driven approaches, and motivating behavioral changes. Self-management outcomes were focused on physical activity, diet, and symptom management. Quality of life, depression, and anxiety were addressed as psychosocial health-related outcomes. Although no consistent results were reported on the effectiveness, DHIs with a combination of multiple personalized features, more than three domains, were likely to be more effective in improving self-management outcomes.

DISCUSSION

This review enhances the understanding of personalized DHIs for older adults with cancer by identifying intervention components, personalized features, and effectiveness on self-management and psychosocial health. Several gaps were identified, including the absence of targeted studies exclusively focusing on older adults, a relative scarcity of personalized features for improving patient engagement, a lack of understanding of the mechanism of effective personalized features, and the necessity for more experimental studies. Addressing these gaps can contribute to improving health outcomes and the quality of care for older adults with cancer by providing the direction for developing effective DHIs.