Search

BackgroundStress is highly prevalent and known to be a risk factor for a wide range of physical and mental disorders. The effectiveness of digital stress management interventions has been confirmed; however, research on its economic merits is still limited. ObjectiveThis study aims to assess the cost-effectiveness, cost-utility, and cost-benefit of a universal digital stress management intervention for employees compared with a waitlist control condition within a time horizon of 6 months. MethodsRecruitment was directed at the German working population. A sample of 396 employees was randomly assigned to the intervention group (n=198) or the waitlist control condition (WLC) group (n=198). The digital stress management intervention included 7 sessions plus 1 booster session, which was offered without therapeutic guidance. Health service use, patient and family expenditures, and productivity losses were self-assessed and used for costing from a societal and an employer’s perspective. Costs were related to symptom-free status (PSS-10 [Perceived Stress Scale] score 2 SDs below the study population baseline mean) and quality-adjusted life years (QALYs) gained. The sampling error was handled using nonparametric bootstrapping. ResultsFrom a societal perspective, the digital intervention was likely to be dominant compared with WLC, with a 56% probability of being cost-effective at a willingness-to-pay (WTP) of €0 per symptom-free person gained. At the same WTP threshold, the digital intervention had a probability of 55% being cost-effective per QALY gained relative to the WLC. This probability increased to 80% at a societal WTP of €20,000 per QALY gained. Taking the employer’s perspective, the digital intervention showed a probability of a positive return on investment of 78%. ConclusionsDigital preventive stress management for employees appears to be cost-effective societally and provides a favorable return on investment for employers. Trial RegistrationGerman Clinical Trials Register DRKS00005699; https://drks.de/search/en/trial/DRKS00005699

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.

Automation of incoming invoices processing promises to yield vast efficiency improvements in accounting. Until a universal adoption of fully electronic invoice exchange formats has been achieved, machine learning can help bridge the adoption gaps in electronic invoicing by extracting structured information from unstructured invoice formats. Machine learning especially helps the processing of invoices of suppliers who only send invoices infrequently, as the models are able to capture the semantic and visual cues of invoices and generalize them to previously unknown invoice layouts. Since the population of invoices in many companies is skewed toward a few frequent suppliers and their layouts, this research examines the effects of training data taken from such populations on the predictive quality of different machine-learning approaches for the extraction of information from invoices. Comparing the different approaches, we find that they are affected to varying degrees by skewed layout populations: The accuracy gap between in-sample and out-of-sample layouts is much higher in the Chargrid and random forest models than in the LayoutLM transformer model, which also exhibits the best overall predictive quality. To arrive at this finding, we designed and implemented a research pipeline that pays special attention to the distribution of layouts in the splitting of data and the evaluation of the models.

IntroductionPersonalization is a much-discussed approach to improve adherence and outcomes for Digital Mental Health interventions (DMHIs). Yet, major questions remain open, such as (1) what personalization is, (2) how prevalent it is in practice, and (3) what benefits it truly has.MethodsWe address this gap by performing a systematic literature review identifying all empirical studies on DMHIs targeting depressive symptoms in adults from 2015 to September 2022. The search in Pubmed, SCOPUS and Psycinfo led to the inclusion of 138 articles, describing 94 distinct DMHIs provided to an overall sample of approximately 24,300 individuals.ResultsOur investigation results in the conceptualization of personalization as purposefully designed variation between individuals in an intervention's therapeutic elements or its structure. We propose to further differentiate personalization by what is personalized (i.e., intervention content, content order, level of guidance or communication) and the underlying mechanism [i.e., user choice, provider choice, decision rules, and machine-learning (ML) based approaches]. Applying this concept, we identified personalization in 66% of the interventions for depressive symptoms, with personalized intervention content (32% of interventions) and communication with the user (30%) being particularly popular. Personalization via decision rules (48%) and user choice (36%) were the most used mechanisms, while the utilization of ML was rare (3%). Two-thirds of personalized interventions only tailored one dimension of the intervention.DiscussionWe conclude that future interventions could provide even more personalized experiences and especially benefit from using ML models. Finally, empirical evidence for personalization was scarce and inconclusive, making further evidence for the benefits of personalization highly needed.Systematic Review RegistrationIdentifier: CRD42022357408.

BackgroundEmerging evidence indicates the effectiveness of internet-based mobile-supported stress management interventions (iSMIs) in highly stressed employees. It is yet unclear, however, whether iSMIs are also effective without a preselection process in a universal prevention approach, which more closely resembles routine occupational health care. Moreover, evidence for whom iSMIs might be suitable and for whom not is scarce. ObjectiveThe aim of this study was to evaluate the iSMI GET.ON Stress in a universal prevention approach without baseline inclusion criteria and to examine the moderators of the intervention effects. MethodsA total of 396 employees were randomly assigned to the intervention group or the 6-month waiting list control group. The iSMI consisted of 7 sessions and 1 booster session and offered no therapeutic guidance. Self-report data were assessed at baseline, 7 weeks, and at 6 months following randomization. The primary outcome was perceived stress. Several a priori defined moderators were explored as potential effect modifiers. ResultsParticipants in the intervention group reported significantly lower perceived stress at posttreatment (d=0.71, 95% CI 0.51-0.91) and at 6-month follow-up (d=0.61, 95% CI 0.41-0.81) compared to those in the waiting list control group. Significant differences with medium-to-large effect sizes were found for all mental health and most work-related outcomes. Resilience (at 7 weeks, P=.04; at 6 months, P=.01), agreeableness (at 7 weeks, P=.01), psychological strain (at 6 months, P=.04), and self-regulation (at 6 months, P=.04) moderated the intervention effects. ConclusionsThis study indicates that iSMIs can be effective in a broad range of employees with no need for preselection to achieve substantial effects. The subgroups that might not profit had extreme values on the respective measures and represented only a very small proportion of the investigated sample, thereby indicating the broad applicability of GET.ON Stress. Trial RegistrationGerman Clinical Trials Register DRKS00005699; https://www.drks.de/DRKS00005699

Background: Panic disorder with and without agoraphobia (PD) is a common psychological disorder. Internet-based interventions have the potential to offer highly scalable low-threshold evidence-based care to people suffering from PD. GET.ON Panic is a newly developed internet-based intervention addressing symptoms of PD. In order to transfer the training into the daily life of the individuals, we integrated mobile components in the training and created a so-called hybrid online training. The development and beta-testing of such a training requires a novel interdisciplinary approach between IT specialists and psychologists. From this point of view, we would like to share our experiences in this exploratory paper. Methods: This initial feasibility study (N = 10) offers, on the one hand, a brief overview of the interdisciplinary development phase of the mobile application and on the other hand, provides first insights into the usage, usability and acceptance of this mobile application using qualitative interview data as well quantitative measures of 8 completing participants. For these reasons, we used a pre-posttest design without a control group. Furthermore, we present initial clinical outcomes of the intervention on e.g. panic symptom severity, depressive symptoms as well additional anxiety measures. Finally, we end with implications for further research in the relatively new field of mobile mental health. Results: Overall, usability, user satisfaction, motivational value and technology acceptance of the app were perceived as high. The usage of app components was diverse: The use of interoceptive exposure exercises and daily summaries on anxiety and mood was highest while using in-vivo exposure exercises and monitoring panic symptoms was perceived as difficult. Furthermore, participants showed after the training less clinical symptoms as at baseline-assessment. Discussion: The current feasibility study contributes to an in-depth understanding of the potential of mobile technology in e-mental health. Overall, the GET.ON Panic app appears to be an acceptable and motivational part of a CBT-based hybrid online training for PD that has the potential to promote training success. After some suggested adjustments have been made, the efficacy should be investigated in a randomized controlled trial. Keywords: M-mental health, Panic disorder, Agoraphobia, Feasibility study, Thematic analyses, Cognitive behavior therapy

In this paper, we explore the potential of predicting therapy success for patients in mental health care. Such predictions can eventually improve the process of matching effective therapy types to individuals. In the EU project E-COMPARED, a variety of information is gathered about patients suffering from depression. We use this data, where 276 patients received treatment as usual and 227 received blended treatment, to investigate to what extent we are able to predict therapy success. We utilize different encoding strategies for preprocessing, varying feature selection techniques, and different statistical procedures for this purpose. Significant predictive power is found with average AUC values up to 0.7628 for treatment as usual and 0.7765 for blended treatment. Adding daily assessment data for blended treatment does currently not add predictive accuracy. Cost effectiveness analysis is needed to determine the added potential for real-world applications. Keywords: Prediction, Therapy success, E-health, Depression, Classification

Recent developments in mobile technology, sensor devices, and artificial intelligence have created new opportunities for mental health care research. Enabled by large datasets collected in e-mental health research and practice, clinical researchers and members of the data mining community increasingly join forces to build predictive models for health monitoring, treatment selection, and treatment personalization. This paper aims to bridge the historical and conceptual gaps between the distant research domains involved in this new collaborative research by providing a conceptual model of common research goals. We first provide a brief overview of the data mining field and methods used for predictive modeling. Next, we propose to characterize predictive modeling research in mental health care on three dimensions: 1) time, relative to treatment (i.e., from screening to post-treatment relapse monitoring), 2) types of available data (e.g., questionnaire data, ecological momentary assessments, smartphone sensor data), and 3) type of clinical decision (i.e., whether data are used for screening purposes, treatment selection or treatment personalization). Building on these three dimensions, we introduce a framework that identifies four model types that can be used to classify existing and future research and applications. To illustrate this, we use the framework to classify and discuss published predictive modeling mental health research. Finally, in the discussion, we reflect on the next steps that are required to drive forward this promising new interdisciplinary field.

OBJECTIVE: This study aimed to estimate and evaluate the cost-effectiveness and cost-benefit of a guided internet- and mobile-supported occupational stress-management intervention (iSMI) for employees from the employer’s perspective alongside a randomized controlled trial. METHODS: A sample of 264 employees with elevated symptoms of perceived stress (Perceived Stress Scale, PSS-10 ≥22) was randomly assigned either to the iSMI or a waitlist control (WLC) group with unrestricted access to treatment as usual. The iSMI consisted of seven sessions of problem-solving and emotion-regulation techniques and one booster session. Self-report data on symptoms of perceived stress and economic data were assessed at baseline, and at six months following randomization. A cost-benefit analysis (CBA) and a cost-effectiveness analysis (CEA) with symptom-free status as the main outcome from the employer’s perspective was carried out. Statistical uncertainty was estimated using bootstrapping (N=5000). RESULTS: The CBA yielded a net-benefit of EUR181 [95% confidence interval (CI) -6043–1042] per participant within the first six months following randomization. CEA showed that at a willingness-to-pay ceiling of EUR0, EUR1000, EUR2000 for one additional symptom free employee yielded a 67%, 90%, and 98% probability, respectively, of the intervention being cost-effective compared to the WLC. CONCLUSION: The iSMI was cost-effective when compared to WLC and even lead to cost savings within the first six months after randomization. Offering stress-management interventions can present good value for money in occupational healthcare.

Self-esteem is a crucial factor for an individual’s well-being and mental health. Low self-esteem is associated with depression and anxiety. Data about self-esteem is oftentimes collected in Internet-based interventions through Ecological Momentary Assessments and is usually provided on an ordinal scale. We applied models for ordinal outcomes in order to predict the self-esteem of 130 patients based on diary data of an online depression treatment and thereby illustrated a path of how to analyze EMA data in Internet-based interventions. Specifically, we analyzed the relationship between mood, worries, sleep, enjoyed activities, social contact, and the self-esteem of patients. We explored several ordinal models with varying degrees of heterogeneity and estimated them using Bayesian statistics. Thereby, we demonstrated how accounting for patient-heterogeneity influences the prediction performance of self-esteem. Our results show that models that allow for more heterogeneity performed better regarding various performance measures. We also found that higher mood levels and enjoyed activities are associated with higher self-esteem. Sleep, social contact, and worries were significant predictors for only some individuals. Patient-individual parameters enable us to better understand the relationships between the variables on a patient-individual level. The analysis of relationships between self-esteem and other psychological factors on an individual level can therefore lead to valuable information for therapists and practitioners.