A Web- and Mobile App-Based Mental Health Promotion Intervention Comparing Email, Short Message Service, and Videoconferencing Support for a Healthy Cohort: Randomized Comparative Study

A Web-Based Intervention Using "Five Ways to Wellbeing" to Promote Well-Being and Mental Health: Randomized Controlled Trial

BACKGROUND

Compromised well-being and mental health problems pose a significant threat to individuals and societies worldwide. Resource-intensive psychological treatments alone cannot alleviate this burden. There is a need for low-cost, evidence-based interventions aimed at preventing illness and promoting well-being. Five activity domains appear to be linked with well-being promotion across populations: connecting with others, being active, taking notice, learning, and being generous/giving. The activities mentioned are part of the Five Ways to Wellbeing framework and the web-based intervention Five Ways to Wellbeing for All (5waysA).

OBJECTIVE

This randomized controlled trial aims to test the effects of the 5waysA intervention, a web-based, low-cost, well-being-promoting measure targeting the general population. To date, the Five Ways to Wellbeing framework has not been tested in this specific format. The 5waysA intervention comprises 2 webinars and SMS text message reminders delivered over a 10-week period.

METHODS

In 2021, a total of 969 study participants from various regions across Norway were openly recruited through a web page. They were then randomly assigned to either an intervention group or 1 of 2 waiting list control groups, namely, active or passive. Self-reported life satisfaction (Satisfaction With Life Scale [SWLS]), flourishing (Flourishing Scale [FS]), positive emotions, anxiety, and depression symptoms (Hopkins Symptom Checklist-8 [HSCL-8]) were assessed before the intervention, at 4 weeks into the intervention, and 1-2 weeks after the intervention (over 10 weeks). Data analysis was conducted using linear mixed (multilevel) models.

RESULTS

After 10 weeks, 453 participants (171 in the intervention group and 282 in the waiting list control group) were assessed on outcome variables, with a dropout rate of 53.2% (516/969). Results revealed a significantly greater increase in the intervention group compared with the controls for SWLS (b=0.13, 95% CI 0.03-0.23; P=.001), FS (b=0.19, 95% CI 0.08-0.30; P=.001), positive emotions (b=0.43, 95% CI 0.27-0.60; P<.001), and these factors combined into a global well-being measure (b=0.28, CI 0.16-0.39; P<.001). Effect sizes (Cohen d) for the well-being outcomes ranged from 0.30 to 0.49. In addition, a significant decrease in anxiety and depressive symptoms was observed (b=-0.17, 95% CI -0.30 to -0.04; P=.001) with an effect size (Cohen d) of -0.20.

CONCLUSIONS

The findings suggest that the web-based 5waysA intervention could serve as an effective approach for enhancing well-being and mental health within the general population. This study offers individuals, policy makers, and local stakeholders an accessible and potentially cost-effective well-being intervention that could be easily implemented.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04784871; https://clinicaltrials.gov/study/NCT04784871.

Persuasive technologies design for mental and behavioral health platforms: A scoping literature review

This review investigates persuasive design frameworks within eHealth, concentrating on methodologies, their prevalence in mental and behavioral health applications, and identifying current research gaps. An extensive search was conducted across 8 databases, focusing on English publications with full text available. The search prioritized primary research articles, post-2011 applications, and eHealth platforms emphasizing treatment or support. The inclusion process was iterative, involving multiple authors, and relied on detailed criteria to ensure the relevance and contemporaneity of selected works. The final review set comprised 161 articles, providing an overview of persuasive design frameworks in eHealth. The review highlights the state of the art in the domain, emphasizing the utilization and effectiveness of these frameworks in eHealth platforms. This review details the restricted adoption of persuasive design frameworks within the field of eHealth, particularly in the mental and behavioral sectors. Predominant gaps include the scarcity of comparative evaluations, the underrepresentation of tailored interventions, and the unclear influence of persuasive components on user experience. There is a notable requirement for further scrutiny and refinement of persuasive design frameworks. Addressing these concerns promises a more substantial foundation for persuasive design in eHealth, potentially enhancing user commitment and platform efficiency.

The Impact of Intervention Design on User Engagement in Digital Therapeutics Research: Factorial Experiment With a Mixed Methods Study

BACKGROUND

User engagement is crucial for digital therapeutics (DTx) effectiveness; due to variations in the conceptualization of engagement and intervention design, assessment and retention of engagement remain challenging.

OBJECTIVE

We investigated the influence of the perceived acceptability of experimental intervention components and satisfaction with core intervention components in DTx on user engagement, while also identifying potential barriers and facilitators to user engagement.

METHODS

We conducted a mixed methods study with a 2 × 2 factorial design, involving 12 outpatients with atopic dermatitis. Participants were randomized into 4 experimental groups based on push notification ("basic" or "advanced") and human coach ("on" or "off") experimental intervention components. All participants engaged in self-monitoring and learning courses as core intervention components within an app-based intervention over 8 weeks. Data were collected through in-app behavioral data, physician- and self-reported questionnaires, and semistructured interviews assessed at baseline, 4 weeks, and 8 weeks. Descriptive statistics and thematic analysis were used to evaluate user engagement, perceived acceptability of experimental intervention components (ie, push notification and human coach), satisfaction with core intervention components (ie, self-monitoring and learning courses), and intervention effectiveness through clinical outcomes.

RESULTS

The primary outcome indicated that group 4, provided with "advanced-level push notifications" and a "human coach," showed higher completion rates for self-monitoring forms and learning courses compared to the predetermined threshold of clinical significance. Qualitative data analysis revealed three key themes: (1) perceived acceptability of the experimental intervention components, (2) satisfaction with the core intervention components, and (3) suggestions for improvement in the overall intervention program. Regarding clinical outcomes, the Perceived Stress Scale and Dermatology Life Quality Index scores presented the highest improvement in group 4.

CONCLUSIONS

These findings will help refine the intervention and inform the design of a subsequent randomized trial to test its effectiveness. Furthermore, this design may serve as a model for broadly examining and optimizing overall engagement in DTx and for future investigation into the complex relationship between engagement and clinical outcomes.

TRIAL REGISTRATION

Clinical Research Information Service KCT0007675; http://tinyurl.com/2m8rjrmv.

The Degree of Anxiety and Depression in Patients With Cardiovascular Diseases as Assessed Using a Mobile App: Cross-Sectional Study

BACKGROUND

Depression and anxiety are common comorbidities in cardiovascular clinic outpatients. Timely identification and intervention of these mental and psychological disorders can contribute to correct diagnosis, better prognosis, less medical expenses, and improved quality of life. The convenience of online doctor-patient communication platforms has increasingly attracted patients to online consultations. However, online health care and offline health care are very different. Research on how to identify psychological disorders in patients who engage in an online cardiology consultation is lacking.

OBJECTIVE

This study aimed to explore the feasibility of using a self-rating scale to assess mental illness among patients who consult with a cardiologist online and to compare the differences in anxiety and depression between online and offline patients.

METHODS

From June 2022 to July 2022, we conducted follow-up visits with 10,173 patients on the Haodf platform. We conducted detailed consultations with 286 patients who visited the same cardiologist in the outpatient department. We used the self-rated Generalized Anxiety Disorder (GAD-7) and Patient Health Questionnaire (PHQ-9) scales to assess anxiety and depression, respectively. We analyzed the influencing factors related to the degree of coordination of online patients. We also compared the prevalence of anxiety or depression between online and offline patients and analyzed the factors related to anxiety or depression.

RESULTS

Of the 10,173 online consultation patients, only 186 (1.8%) responded effectively. The response rate of the offline consultation patients was 96.5% (276/286). Frequent online communication and watching live video broadcasts were significantly related to effective responses from online patients (P<.001). The prevalence of anxiety (70/160, 43.7% vs 69/266, 25.8%; P<.001) or depression (78/160, 48.7% vs 74/266, 27.7%; P<.001) in online consultation patients was significantly higher than that in offline patients. In bivariate analyses, the factors related to anxiety included female sex, unemployment, no confirmed cardiovascular disease, and the online consultation mode, while smokers and those who underwent COVID-19 quarantine were less likely to present with anxiety. The factors related to depression included female sex, divorced or separated individuals, and the online consultation mode. COVID-19 quarantine was related with a lower likelihood of depression. BMI was negatively correlated with depression. In multiple ordered logistic regression analysis, women were more likely than men to present with anxiety (odds ratio [OR] 2.181, 95% CI 1.365-3.486; P=.001). Women (OR 1.664, 95% CI 1.082-2.559; P=.02) and online patients (OR 2.631, 95% CI 1.305-5.304; P=.007) were more likely to have depression.

CONCLUSIONS

Online patients had more anxiety or depression than offline patients. Anxiety was more prevalent in women, the unemployed, and those without confirmed cardiovascular disease. Women and divorced or separated individuals were more prone to depression. Increasing the frequency of doctor-patient communication and participating in video interactions can help improve patient cooperation.

Types of Digital-Based Nursing Interventions for Reducing Stress and Depression Symptoms on Adolescents During COVID-19 Pandemic: A Scoping Review

COVID-19 pandemic can cause problems in adolescent mental health such as anxiety, depression, and stress. This is because the distance barrier becomes an obstacle in handling mental health in adolescents. The use of technology has the potential to deal with mental health problems. The purpose of this study was to describe the types of digital-based nursing interventions to reduce symptoms of stress and depression on adolescents during the COVID-19 pandemic. This study used the Scoping Review method. Literature from CINAHL, PubMed, and ProQuest databases. The keywords were adolescent, depression, stress, digital, application, and nursing intervention in English. The criteria for articles in this study were full-text articles, the sample of adolescents, digital-based intervention, articles are original research, and time setting 2018-2022. We found 11 articles discussing digital-based nursing interventions to reduce symptoms of stress and depression in adolescents. There are 2 types of intervention, namely mobile-based intervention, and web-based intervention. The two interventions can be combined to become a method of providing digital nursing interventions that are effective and can reach the entire community. Digital-based nursing interventions are carried out by paying attention to physical, psychological, spiritual, and cultural aspects to improve the goals of nursing care so that they can be significant in reducing stress and depression in adolescents during the Covid-19 pandemic. Digital-based nursing interventions consisting of mobile-based intervention and web-based intervention can improve mental health among adolescents by reducing stress, anxiety, depression, and increasing resilience, well-being, and self-efficacy.

Using online social networks to provide a parental health-education intervention for preventing unintentional injuries among children aged 0-3 years: A randomized controlled trial and social network analysis in Shanghai, China

Introduction

Unintentional injury among children represents a major public health problem. Online-social-network-based parental-health-education is a potential way to reduce child unintentional injuries. The study aimed to explore the mechanisms by which online-social-network-based health education may reduce the unintentional injuries among children aged 0-3 years.

Methods

We conducted a participant-blinded, randomized controlled, online-social-network-based health-education intervention study from March 2019 to February 2020 in Shanghai. We established four WeChat groups (two intervention groups and two control groups). For the intervention groups, a doctor's assistant regularly delivered information regarding unintentional injuries among children, and community childcare doctors answered parents' questions concerning their children's health, including unintentional injuries. Meanwhile, the control groups did not receive any information from the assistant. The study selected one intervention group and one control group and compared the ego network and whole network indicators to determine the differences between the intervention and control groups.

Results

In the intervention and control groups, 64.5% and 31.9% of the members, respectively, engaged in communication, and 1,736 and 273 records, respectively, were obtained. Regarding ego network, the doctor showed the largest network in the intervention group, and the size of the intervention group's network was twice that of the control group; the number of ties in the intervention group was nine times that of the control group. Fourteen and four parents in the intervention and control group played an active role, respectively. Regarding centrality, all WeChat groups were loose and multiple centers existed. Regarding subgroup cohesion, the intervention group had 28 cliques with 27 members, and the control group had 4 cliques with 4 members. For structural hole, 23.7% and 7.5% members in the intervention and control group actively participated in interactions, respectively, having strong control and influence over other parents; 69.2% and 59.1% members in the intervention and control group, respectively, had values of < 1.000, showing that they had strong ability to cross-jump structural holes.

Discussion

Online-social-networks-based health education interventions could enhance communication among parents, and between parents and community childcare doctors, and also shorten the social distance between them. Thus, online-social-network-based parental-health-education-intervention can be a feasible and generalizable means of preventing unintentional injuries among children.

Text Messages for Depression, Anxiety and Alcohol Abuse Therapy-Are Construction Guidelines Needed?

Despite the effectiveness of text messaging therapy in improving mental health conditions, limited attention has been paid to how the text messages are constructed. Thus, this study questions whether there is a need to develop a model of text message construction for mental health therapy. In this backdrop, this study reviews how a text message for mental health therapy is constructed, specifically focused on the process and guidelines. This study also aims to identify the research gap regarding the guideline of text message construction for mental health therapy and to identify mental health professionals' practices in text messaging therapy. In addition, the opinions of mental health professionals on the need to develop a text message construction guideline were also gathered. The findings from the literature review confirmed that there are still limited guidelines explaining the process of constructing text messages for mental health therapy. Meanwhile, results from the online survey found that mental health professionals expressed a high need to explore and develop a model of text message construction for mental health therapy. With this research gap addressed, this study proposes further research into the development of a text message construction model for mental health therapy.

Current directions in digital interventions for mood and anxiety disorders

PURPOSE OF REVIEW

The purpose of this review is to provide an overview of the current scope of digital interventions for anxiety and mood disorders, with a focus on smartphone apps, for clinicians and mental healthcare providers.

RECENT FINDINGS

Of the 11 randomized controlled trials analyzed, 7 showed evidence that guided digital interventions (those supported by humans) were effective in improving anxiety and mood symptoms, and 3 showed evidence that unguided (those not supported by humans) interventions were effective. Psychoeducation was the most popular feature of both guided and unguided interventions. Attrition was highest (50%) in the unguided app-based intervention and lowest in the guided interventions. Many studies lacked active control groups and comparison was often made to a nondigital or waitlist control condition.

SUMMARY

Guided digital interventions continue to show promising results and can be used to enhance clinical care with minimal resources although more direct comparisons to existing treatments are necessary to understand their actual efficacy. Unguided self-help apps and chatbots remain promising, but more work is necessary to understand the real-world engagement and efficacy of these interventions.

Efficiency of an Online Health-Promotion Program in Individuals with At-Risk Mental State during the COVID-19 Pandemic

Mental health issues caused by the COVID-19 pandemic greatly impact people’s daily lives. Individuals with an at-risk mental state are more vulnerable to mental health issues, and these may lead to onset of full psychotic illnesses. This study aimed to develop and evaluate an online health-promotion program for physical and mental health of the individuals with at-risk mental state during the COVID-19 pandemic. A single group study with pre- and post-tests was conducted in 39 young adults with at-risk mental state. The participants were provided with the online health-promotion program after completing the pretest. Via social media, the online counseling program released one topic of material (about 15–20 min) every two weeks and provided interactive counseling for specific personal health needs on the platform. Study questionnaires, physiological examination, and blood serum examination were completed at both pre- and post-tests. The participants showed significant improvements in mental risk, anxiety, and physical activity after participating in the program. Furthermore, those who did not complete the program had significantly more severe negative symptoms. These results imply that the online health-promotion program is effective and accessible under certain barriers such as the COVID-19 pandemic, but not for individuals with higher risk of more negative mental health symptoms.

The Influence of Gender and Age on the Outcomes of and Adherence to a Digital Interdisciplinary Mental Health Promotion Intervention in an Australasian Nonclinical Setting: Cohort Study

BACKGROUND

The global prevalence of mental health disorders is at a crisis point, particularly in the wake of COVID-19, prompting calls for the development of digital interdisciplinary mental health promotion interventions (MHPIs) for nonclinical cohorts. However, the influence of gender and age on the outcomes of and adherence to MHPIs is not well understood.

OBJECTIVE

The aim of this study was to determine the influence of gender and age on the outcomes of and adherence to a 10-week digital interdisciplinary MHPI that integrates strategies from positive psychology and lifestyle medicine and utilizes persuasive systems design (PSD) principles in a nonclinical setting.

METHODS

This study involved 488 participants who completed the digital interdisciplinary MHPI. Participants completed a pre and postintervention questionnaire that used: (1) the "mental health" and "vitality" subscales from the Short Form 36 (SF-36) Health Survey; (2) the Depression, Anxiety and Stress Scale (DASS-21); and (3) Satisfaction With Life Scale (SWL). Adherence to the digital interdisciplinary MHPI was measured by the number of educational videos the participants viewed and the extent to which they engaged in experiential challenge activities offered as part of the program.

RESULTS

On average, the participants (N=488; mean age 47.1 years, SD 14.1; 77.5% women) demonstrated statistically significant improvements in all mental health and well-being outcome measures, and a significant gender and age interaction was observed. Women tended to experience greater improvements than men in the mental health and well-being measures, and older men experienced greater improvements than younger men in the mental health and vitality subscales. Multiple analysis of variance results of the adherence measures indicated a significant difference for age but not gender. No statistically significant interaction between gender and age was observed for adherence measures.

CONCLUSIONS

Digital interdisciplinary MHPIs that utilize PSD principles can improve the mental health and well-being of nonclinical cohorts, regardless of gender or age. Hence, there may be a benefit in utilizing PSD principles to develop universal MHPIs such as that employed in this study, which can be used across gender and age groups. Future research should examine which PSD principles optimize universal digital interdisciplinary MHPIs.

TRIAL REGISTRATION

Australian New Zealand Clinical Trials Registry ACTRN12619000993190; http://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=377889 and Australian New Zealand Clinical Trials Registry ACTRN12619001009101; http://www.anzctr.org.au/ACTRN12619001009101.aspx.

The Influence of Human Support on the Effectiveness of Digital Mental Health Promotion Interventions for the General Population

Mental wellbeing amongst the general population is languishing—exacerbated by the Coronavirus Disease 2019 (COVID-19) pandemic. Digital mental health promotion interventions, that improve mental health literacy and encourage adoption of evidence-informed practical strategies are essential. However, attrition and non-adherence are problematic in digital interventions. Human support is often applied as an antidote; yet, there is a paucity of randomized trials that compare different human support conditions amongst general population cohorts. Limited trials generally indicate that human support has little influence on adherence or outcomes in DMHPIs. However, providing participants autonomy to self-select automated support options may enhance motivation and adherence.

The effectiveness of an online interdisciplinary intervention for mental health promotion: a randomized controlled trial

BACKGROUND

There is an urgent need for efficacious interventions to combat the global mental health crisis, and mental health promotion and primary prevention approaches are paramount. The aim of this study is to examine whether an online interdisciplinary intervention that incorporates evidence-based strategies from the disciplines of Lifestyle Medicine and Positive Psychology improves measures of mental health and emotional wellness.

METHODS

A randomized controlled trial with a wait-list control (N = 425, aged 46.97 ± 14.5, 69.9% females) was conducted in Australia and New Zealand. The intervention group participated in a 10-week online interdisciplinary intervention. Primary outcome measures of mental health and emotional wellness were taken at baseline (Week 1), post-intervention (Week 12), and 12 weeks post-intervention (Week 24). The wait-list control completed the same assessments.

RESULTS

General Linear Modelling analyses indicated that the intervention group experienced significantly greater improvements than the wait-list control group over time in all outcome measures: mental health (F(319) = 7.326, p = 0.007) and vitality (F(319) = 9.445, p = 0.002) subscales of the Short Form Survey (SF-36); depression (F(319) = 7.841, p = 0.005), anxiety (F(319) = 4.440, p = 0.36) and stress (F(319) = 12.494, p < 0.001) scales of the Depression, Anxiety and Stress Scale (DASS-21); and life satisfaction (F(319) = 8.731, p = 0.003) as measured by the Satisfaction With Life Scale. Within the intervention group, significant improvements were observed from Week 1 to 12 in all outcome measures: mental health (10%, t(167) = - 6.423), p < 0.001, dz = 0.50), vitality (22%, t(167) = - 7.043, p < 0.001, dz = 0.54), depression (- 41%, t(167) = 6.189, p < 0.001, dz = 0.48), anxiety (- 38%, t(167) = 5.030, p < 0.001, dz = 0.39), stress (- 31%, t(167) = 6.702, p < 0.001, dz = 0.52) and life satisfaction (8%, t(167) = - 6.199, p < 0.001, dz = 0.48). Improvements in the outcome measures remained significant in the intervention group at 12 weeks post-intervention.

CONCLUSION

The online interdisciplinary intervention improved measures of mental health and emotional wellness suggesting that such interventions may be useful for mental health promotion and prevention. Trial registration The Australian New Zealand Clinical Trials Registry. ACTRN12619000993190. Registered on 12 July 2019 (Retrospectively registered). The ANZCTRN is part of the WHO Primary Registries.

Participant Perceptions of Facilitators and Barriers to Adherence in a Digital Mental Health Intervention for a Nonclinical Cohort: Content Analysis

Background

Digital mental health promotion interventions (MHPIs) present a scalable opportunity to attenuate the risk of mental health distress among nonclinical cohorts. However, adherence is frequently suboptimal, and little is known about participants’ perspectives concerning facilitators and barriers to adherence in community-based settings.

Objective

This study aimed to examine participants’ perceptions of facilitators and barriers to adherence in a web- and mobile app–based MHPI for a nonclinical cohort.

Methods

This qualitative study used inductive, reflexive thematic analysis to explore free-text responses in a postintervention evaluation of a 10-week digital MHPI. The intervention was administered using a web and mobile app from September to December 2018. Participants (N=320) were Australian and New Zealand members of a faith-based organization who self-selected into the study, owned a mobile phone with messaging capability, had an email address and internet access, were fluent in English, provided informed consent, and gave permission for their data to be used for research. The postintervention questionnaire elicited participants’ perceptions of facilitators and barriers to adherence during the intervention period.

Results

Key factors that facilitated adherence were engaging content, time availability and management, ease of accessibility, easy or enjoyable practical challenges, high perceived value, and personal motivation to complete the intervention. The primary perceived barrier to adherence was the participants’ lack of time. Other barriers included completing and recording practical activities, length of video content, technical difficulties, and a combination of personal factors.

Conclusions

Time scarcity was the foremost issue for the nonclinical cohort engaged in this digital MHPI. Program developers should streamline digital interventions to minimize the time investment for participants. This may include condensed content, optimization of intuitive web and app design, simplified recording of activities, and greater participant autonomy in choosing optional features. Nonetheless, participants identified a multiplicity of other interindividual factors that facilitated or inhibited adherence.

Addressing the COVID-19 Mental Health Crisis: A Perspective on Using Interdisciplinary Universal Interventions

Mental health is reaching a crisis point due to the ramifications of COVID-19. In an attempt to curb the spread of the virus and circumvent health systems from being overwhelmed, governments have imposed regulations such as lockdown restrictions and home confinement. These restrictions, while effective for infection control, have contributed to poorer lifestyle behaviors. Currently, Positive Psychology and Lifestyle Medicine are two distinct but complimentary disciplines that offer an array of evidence-based approaches for promoting mental health and well-being across a universal population. However, these strategies for improving mental health are typically used in isolation. This perspective calls for a new paradigm shift to create and rollout well-designed interdisciplinary universal multicomponent mental health interventions that integrates the benefits of both disciplines, and uses innovative digital mental health solutions to achieve scalability and accessibility within the limitations and beyond the COVID-19 lockdown and restrictions.

The Influence of Three Modes of Human Support on Attrition and Adherence to a Web- and Mobile App-Based Mental Health Promotion Intervention in a Nonclinical Cohort: Randomized Comparative Study

Background

The escalating prevalence of mental health disorders necessitates a greater focus on web- and mobile app–based mental health promotion initiatives for nonclinical groups. However, knowledge is scant regarding the influence of human support on attrition and adherence and participant preferences for support in nonclinical settings.

Objective

This study aimed to compare the influence of 3 modes of human support on attrition and adherence to a digital mental health intervention for a nonclinical cohort. It evaluated user preferences for support and assessed whether adherence and outcomes were enhanced when participants received their preferred support mode.

Methods

Subjects participated in a 10-week digital mental health promotion intervention and were randomized into 3 comparative groups: standard group with automated emails (S), standard plus personalized SMS (S+pSMS), and standard plus weekly videoconferencing support (S+VCS). Adherence was measured by the number of video lessons viewed, points achieved for weekly experiential challenge activities, and the total number of weeks that participants recorded a score for challenges. In the postquestionnaire, participants ranked their preferred human support mode from 1 to 4 (S, S+pSMS, S+VCS, S+pSMS & VCS combined). Stratified analysis was conducted for those who received their first preference. Preintervention and postintervention questionnaires assessed well-being measures (ie, mental health, vitality, depression, anxiety, stress, life satisfaction, and flourishing).

Results

Interested individuals (N=605) enrolled on a website and were randomized into 3 groups (S, n=201; S+pSMS, n=202; S+VCS, n=201). Prior to completing the prequestionnaire, a total of 24.3% (147/605) dropped out. Dropout attrition between groups was significantly different (P=.009): 21.9% (44/201) withdrew from the S group, 19.3% (39/202) from the S+pSMS group, and 31.6% (64/202) from the S+VCS group. The remaining 75.7% (458/605) registered and completed the prequestionnaire (S, n=157; S+pSMS, n=163; S+VCS, n=138). Of the registered participants, 30.1% (138/458) failed to complete the postquestionnaire (S, n=54; S+pSMS, n=49; S+VCS, n=35), but there were no between-group differences (P=.24). For the 69.9% (320/458; S, n=103; S+pSMS, n=114; S+VCS, n=103) who completed the postquestionnaire, no between-group differences in adherence were observed for mean number of videos watched (P=.42); mean challenge scores recorded (P=.71); or the number of weeks that challenge scores were logged (P=.66). A total of 56 participants (17.5%, 56/320) received their first preference in human support (S, n=22; S+pSMS, n=26; S+VCS, n=8). No differences were observed between those who received their first preference and those who did not with regard to video adherence (P=.91); challenge score adherence (P=.27); or any of the well-being measures including, mental health (P=.86), vitality (P=.98), depression (P=.09), anxiety (P=.64), stress (P=.55), life satisfaction (P=.50), and flourishing (P=.47).

Conclusions

Early dropout attrition may have been influenced by dissatisfaction with the allocated support mode. Human support mode did not impact adherence to the intervention, and receiving the preferred support style did not result in greater adherence or better outcomes.

Trial Registration

Australian New Zealand Clinical Trials Registry (ANZCTR): 12619001009101; http://www.anzctr.org.au/ACTRN12619001009101.aspx

An App for Classifying Personal Mental Illness at Workplace Using Fit Statistics and Convolutional Neural Networks: Survey-Based Quantitative Study

BACKGROUND

Mental illness (MI) is common among those who work in health care settings. Whether MI is related to employees' mental status at work is yet to be determined. An MI app is developed and proposed to help employees assess their mental status in the hope of detecting MI at an earlier stage.

OBJECTIVE

This study aims to build a model using convolutional neural networks (CNNs) and fit statistics based on 2 aspects of measures and outfit mean square errors for the automatic detection and classification of personal MI at the workplace using the emotional labor and mental health (ELMH) questionnaire, so as to equip the staff in assessing and understanding their own mental status with an app on their mobile device.

METHODS

We recruited 352 respiratory therapists (RTs) working in Taiwan medical centers and regional hospitals to fill out the 44-item ELMH questionnaire in March 2019. The exploratory factor analysis (EFA), Rasch analysis, and CNN were used as unsupervised and supervised learnings for (1) dividing RTs into 4 classes (ie, MI, false MI, health, and false health) and (2) building an ELMH predictive model to estimate 108 parameters of the CNN model. We calculated the prediction accuracy rate and created an app for classifying MI for RTs at the workplace as a web-based assessment.

RESULTS

We observed that (1) 8 domains in ELMH were retained by EFA, (2) 4 types of mental health (n=6, 63, 265, and 18 located in 4 quadrants) were classified using the Rasch analysis, (3) the 44-item model yields a higher accuracy rate (0.92), and (4) an MI app available for RTs predicting MI was successfully developed and demonstrated in this study.

CONCLUSIONS

The 44-item model with 108 parameters was estimated by using CNN to improve the accuracy of mental health for RTs. An MI app developed to help RTs self-detect work-related MI at an early stage should be made more available and viable in the future.