Digital health data-driven approaches to understand human behavior

Augmenting clinicians' analytical workflow through task-based integration of data visualizations and algorithmic insights: a user-centered design study

OBJECTIVES

To understand healthcare providers' experiences of using GlucoGuide, a mockup tool that integrates visual data analysis with algorithmic insights to support clinicians' use of patientgenerated data from Type 1 diabetes devices.

MATERIALS AND METHODS

This qualitative study was conducted in three phases. In Phase 1, 11 clinicians reviewed data using commercial diabetes platforms in a think-aloud data walkthrough activity followed by semistructured interviews. In Phase 2, GlucoGuide was developed. In Phase 3, the same clinicians reviewed data using GlucoGuide in a think-aloud activity followed by semistructured interviews. Inductive thematic analysis was used to analyze transcripts of Phase 1 and Phase 3 think-aloud activity and interview.

RESULTS

3 high level tasks, 8 sub-tasks, and 4 challenges were identified in Phase 1. In Phase 2, 3 requirements for GlucoGuide were identified. Phase 3 results suggested that clinicians found GlucoGuide easier to use and experienced a lower cognitive burden as compared to the commercial diabetes data reports that were used in Phase 1. Additionally, GlucoGuide addressed the challenges experienced in Phase 1.

DISCUSSION

The study suggests that the knowledge of analytical tasks and task-specific visualization strategies in implementing features of data interfaces can result in tools that lower the perceived burden of engaging with data. Additionally, supporting clinicians in contextualizing algorithmic insights by visual analysis of relevant data can positively influence clinicians' willingness to leverage algorithmic support.

CONCLUSION

Task-aligned tools that combine multiple data-driven approaches, such as visualization strategies and algorithmic insights, can improve clinicians' experience in reviewing device data.

Prescription Digital Therapeutics: An Emerging Treatment Option for Negative Symptoms in Schizophrenia

Digital therapeutics-web-based programs, smartphone applications, and wearable devices designed to prevent, treat, or manage clinical conditions through software-driven, evidence-based intervention-can provide accessible alternatives and/or may supplement standard care for patients with serious mental illnesses, including schizophrenia. In this article, we provide a targeted summary of the rapidly growing field of digital therapeutics for schizophrenia and related serious mental illnesses. First, we define digital therapeutics. Then, we provide a brief summary of the emerging evidence of the efficacy of digital therapeutics for improving clinical outcomes, focusing on potential mechanisms of action for addressing some of the most challenging problems, including negative symptoms of psychosis. Our focus on these promising targets for digital therapeutics, including the latest in prescription models in the commercial space, highlights future directions for research and practice in this exciting field.

TanhReLU -based convolutional neural networks for MDD classification

Major Depression Disorder (MDD), a complex mental health disorder, poses significant challenges in accurate diagnosis. In addressing the issue of gradient vanishing in the classification of MDD using current data-driven electroencephalogram (EEG) data, this study introduces a TanhReLU-based Convolutional Neural Network (CNN). By integrating the TanhReLU activation function, which combines the characteristics of the hyperbolic tangent (Tanh) and rectified linear unit (ReLU) activations, the model aims to improve performance in identifying patterns associated with MDD while alleviating the issue of model overfitting and gradient vanishing. Experimental results demonstrate promising outcomes in the task of MDD classification upon the publicly available EEG data, suggesting potential clinical applications.

Examining health apps and wearable use in improving physical and mental well-being across U.S., China, and Singapore

Health apps and wearables are touted to improve physical health and mental well-being. However, it is unclear from existing research the extent to which these health technologies are efficacious in improving physical and mental well-being at a population level, particularly for the underserved groups from the perspective of health equity and social determinants. Also, it is unclear if the relationship between health apps and wearables use and physical and mental well-being differs across individualistic, collectivistic, and a mix of individual-collectivistic cultures. A large-scale online survey was conducted in the U.S. (individualist culture), China (collectivist culture), and Singapore (mix of individual-collectivist culture) using quota sampling after obtaining ethical approval from the Institutional Review Board (IRB-2021-262) of Nanyang Technological University (NTU), Singapore. There was a total of 1004 respondents from the U.S., 1072 from China, and 1017 from Singapore. Data were analyzed using multiple regression and negative binomial regression. The study found that income consistently had the strongest relationship with physical and mental well-being measures in all three countries, while the use of health apps and wearables only had a moderate association with psychological well-being only in the US. Health apps and wearables were associated with the number of times people spent exercising and some mental health outcomes in China and Singapore, but they were only positively associated with psychological well-being in the US. The study emphasizes the importance of considering the social determinants, social-cultural context of the population, and the facilitating conditions for the effective use of digital health technologies. The study suggests that the combined use of both health apps and wearables is most strongly associated with better physical and mental health, though this association is less pronounced when individuals use only apps or wearables.

Tracking Chinese Online Activity and Interest in Osteoporosis Using the Baidu Index

Introduction China's most widely used online search engine, Baidu (Baidu, Inc., Beijing, China), has developed a data collection and analysis tool called the Baidu Index for tracking Internet search trends. The purpose of this study was to examine the utility of the Baidu Index in tracking online osteoporosis information-seeking behavior and comprehending the traits and concerns of the Chinese population. Methods We used the search term "osteoporosis" for the Baidu Index for the years 2018-2022. The geographic and demographic distributions, search volumes, and demand maps were recorded. Results The popularity of the search term "osteoporosis" has increased over time. The search was mostly conducted among women aged 20-39 in northern China. The demand map revealed that the most significant concerns are related to the diagnosis, treatment, and etiology of osteoporosis. Conclusion The Baidu Index is a valuable tool for tracking online health information-seeking behavior among Chinese netizens. Online search trend data appears to reflect the geographic and demographic aspects of osteoporosis to a certain extent.

Personalized mood prediction from patterns of behavior collected with smartphones

Over the last ten years, there has been considerable progress in using digital behavioral phenotypes, captured passively and continuously from smartphones and wearable devices, to infer depressive mood. However, most digital phenotype studies suffer from poor replicability, often fail to detect clinically relevant events, and use measures of depression that are not validated or suitable for collecting large and longitudinal data. Here, we report high-quality longitudinal validated assessments of depressive mood from computerized adaptive testing paired with continuous digital assessments of behavior from smartphone sensors for up to 40 weeks on 183 individuals experiencing mild to severe symptoms of depression. We apply a combination of cubic spline interpolation and idiographic models to generate individualized predictions of future mood from the digital behavioral phenotypes, achieving high prediction accuracy of depression severity up to three weeks in advance (R2 ≥ 80%) and a 65.7% reduction in the prediction error over a baseline model which predicts future mood based on past depression severity alone. Finally, our study verified the feasibility of obtaining high-quality longitudinal assessments of mood from a clinical population and predicting symptom severity weeks in advance using passively collected digital behavioral data. Our results indicate the possibility of expanding the repertoire of patient-specific behavioral measures to enable future psychiatric research.

The Role of a Smart Health Ecosystem in Transforming the Management of Chronic Health Conditions

The effective management of chronic conditions requires an approach that promotes a shift in care from the clinic to the home, improves the efficiency of health care systems, and benefits all users irrespective of their needs and preferences. Digital health can provide a solution to this challenge, and in this paper, we provide our vision for a smart health ecosystem. A smart health ecosystem leverages the interoperability of digital health technologies and advancements in big data and artificial intelligence for data collection and analysis and the provision of support. We envisage that this approach will allow a comprehensive picture of health, personalization, and tailoring of behavioral and clinical support; drive theoretical advancements; and empower people to manage their own health with support from health care professionals. We illustrate the concept with 2 use cases and discuss topics for further consideration and research, concluding with a message to encourage people with chronic conditions, their caregivers, health care professionals, policy and decision makers, and technology experts to join their efforts and work toward adopting a smart health ecosystem.

The RODI mHealth app Insight: Machine-Learning-Driven Identification of Digital Indicators for Neurodegenerative Disorder Detection

Neurocognitive Disorders (NCDs) pose a significant global health concern, and early detection is crucial for optimizing therapeutic outcomes. In parallel, mobile health apps (mHealth apps) have emerged as a promising avenue for assisting individuals with cognitive deficits. Under this perspective, we pioneered the development of the RODI mHealth app, a unique method for detecting aligned with the criteria for NCDs using a series of brief tasks. Utilizing the RODI app, we conducted a study from July to October 2022 involving 182 individuals with NCDs and healthy participants. The study aimed to assess performance differences between healthy older adults and NCD patients, identify significant performance disparities during the initial administration of the RODI app, and determine critical features for outcome prediction. Subsequently, the results underwent machine learning processes to unveil underlying patterns associated with NCDs. We prioritize the tasks within RODI based on their alignment with the criteria for NCDs, thus acting as key digital indicators for the disorder. We achieve this by employing an ensemble strategy that leverages the feature importance mechanism from three contemporary classification algorithms. Our analysis revealed that tasks related to visual working memory were the most significant in distinguishing between healthy individuals and those with an NCD. On the other hand, processes involving mental calculations, executive working memory, and recall were less influential in the detection process. Our study serves as a blueprint for future mHealth apps, offering a guide for enhancing the detection of digital indicators for disorders and related conditions.

Digital Health Technology for Real-World Clinical Outcome Measurement Using Patient-Generated Data: Systematic Scoping Review

BACKGROUND

Digital health technologies (DHTs) play an ever-expanding role in health care management and delivery. Beyond their use as interventions, DHTs also serve as a vehicle for real-world data collection to characterize patients, their care journeys, and their responses to other clinical interventions. There is a need to comprehensively map the evidence-across all conditions and technology types-on DHT measurement of patient outcomes in the real world.

OBJECTIVE

We aimed to investigate the use of DHTs to measure real-world clinical outcomes using patient-generated data.

METHODS

We conducted this systematic scoping review in accordance with the Joanna Briggs Institute methodology. Detailed eligibility criteria documented in a preregistered protocol informed a search strategy for the following databases: MEDLINE (Ovid), CINAHL, Cochrane (CENTRAL), Embase, PsycINFO, ClinicalTrials.gov, and the EU Clinical Trials Register. We considered studies published between 2000 and 2022 wherein digital health data were collected, passively or actively, from patients with any specified health condition outside of clinical visits. Categories for key concepts, such as DHT type and analytical applications, were established where needed. Following screening and full-text review, data were extracted and analyzed using predefined fields, and findings were reported in accordance with established guidelines.

RESULTS

The search strategy identified 11,015 publications, with 7308 records after duplicates and reviews were removed. After screening and full-text review, 510 studies were included for extraction. These studies encompassed 169 different conditions in over 20 therapeutic areas and 44 countries. The DHTs used for mental health and addictions research (111/510, 21.8%) were the most prevalent. The most common type of DHT, mobile apps, was observed in approximately half of the studies (250/510, 49%). Most studies used only 1 DHT (346/510, 67.8%); however, the majority of technologies used were able to collect more than 1 type of data, with the most common being physiological data (189/510, 37.1%), clinical symptoms data (188/510, 36.9%), and behavioral data (171/510, 33.5%). Overall, there has been real growth in the depth and breadth of evidence, number of DHT types, and use of artificial intelligence and advanced analytics over time.

CONCLUSIONS

This scoping review offers a comprehensive view of the variety of types of technology, data, collection methods, analytical approaches, and therapeutic applications within this growing body of evidence. To unlock the full potential of DHT for measuring health outcomes and capturing digital biomarkers, there is a need for more rigorous research that goes beyond technology validation to demonstrate whether robust real-world data can be reliably captured from patients in their daily life and whether its capture improves patient outcomes. This study provides a valuable repository of DHT studies to inform subsequent research by health care providers, policy makers, and the life sciences industry.

TRIAL REGISTRATION

Open Science Framework 5TMKY; https://osf.io/5tmky/.

Identifying factors associated with user retention and outcomes of a digital intervention for substance use disorder: a retrospective analysis of real-world data

Objectives

Successful delivery of digital health interventions is affected by multiple real-world factors. These factors may be identified in routinely collected, ecologically valid data from these interventions. We propose ideas for exploring these data, focusing on interventions targeting complex, comorbid conditions.

Materials and Methods

This study retrospectively explores pre-post data collected between 2016 and 2019 from users of digital cognitive behavioral therapy (CBT)-containing psychoeducation and practical exercises-for substance use disorder (SUD) at UK addiction services. To identify factors associated with heterogenous user responses to the technology, we employed multivariable and multivariate regressions and random forest models of user-reported questionnaire data.

Results

The dataset contained information from 14 078 individuals of which 12 529 reported complete data at baseline and 2925 did so again after engagement with the CBT. Ninety-three percent screened positive for dependence on 1 of 43 substances at baseline, and 73% screened positive for anxiety or depression. Despite pre-post improvements independent of user sociodemographics, women reported more frequent and persistent symptoms of SUD, anxiety, and depression. Retention-minimum 2 use events recorded-was associated more with deployment environment than user characteristics. Prediction accuracy of post-engagement outcomes was acceptable (Area Under Curve [AUC]: 0.74-0.79), depending non-trivially on user characteristics.

Discussion

Traditionally, performance of digital health interventions is determined in controlled trials. Our analysis showcases multivariate models with which real-world data from these interventions can be explored and sources of user heterogeneity in retention and symptom reduction uncovered.

Conclusion

Real-world data from digital health interventions contain information on natural user-technology interactions which could enrich results from controlled trials.

Impact of Digital Globalization on Health Behavior Through Integration of Ideological and Political Education in Schools Leading to the Spirit for Fighting Against (COVID)-19

Objectives: In this study, we examined the impact of digital globalization on health behavior among students in Chinese schools, particularly in relation to the fight against COVID-19. Despite China's well-established system and positive health behavior towards the pandemic, students' health behavior is lacking. The study focuses on the role of ideological and political education in addressing this issue. Methods: Data were collected from Chinese schools with the help of a survey questionnaire by using area cluster sampling. Data analysis was carried out by employing Smart PLS. Results: We found that digital globalization has a positive effect on health behavior. Digital globalization also has a positive effect on global knowledge about COVID-19 and ideological and political education leading to health behavior. Conclusion: We identified that the influential role of digital globalization can change health behavior. Digital globalization led to global knowledge about the COVID-19 and further caused an influence health behavior among schools that led to improved health behavior of students. The outcomes of the study have valuable importance for the management of schools to decrease the effect of COVID-19 by developing positive health behavior.

Discerning conversational context in online health communities for personalized digital behavior change solutions using Pragmatics to Reveal Intent in Social Media (PRISM) framework

BACKGROUND

Online health communities (OHCs) have emerged as prominent platforms for behavior modification, and the digitization of online peer interactions has afforded researchers with unique opportunities to model multilevel mechanisms that drive behavior change. Existing studies, however, have been limited by a lack of methods that allow the capture of conversational context and socio-behavioral dynamics at scale, as manifested in these digital platforms.

OBJECTIVE

We develop, evaluate, and apply a novel methodological framework, Pragmatics to Reveal Intent in Social Media (PRISM), to facilitate granular characterization of peer interactions by combining multidimensional facets of human communication.

METHODS

We developed and applied PRISM to analyze peer interactions (N = 2.23 million) in QuitNet, an OHC for tobacco cessation. First, we generated a labeled set of peer interactions (n = 2,005) through manual annotation along three dimensions: communication themes (CTs), behavior change techniques (BCTs), and speech acts (SAs). Second, we used deep learning models to apply our qualitative codes at scale. Third, we applied our validated model to perform a retrospective analysis. Finally, using social network analysis (SNA), we portrayed large-scale patterns and relationships among the aforementioned communication dimensions embedded in peer interactions in QuitNet.

RESULTS

Qualitative analysis showed that the themes of social support and behavioral progress were common. The most used BCTs were feedback and monitoring and comparison of behavior, and users most commonly expressed their intentions using SAs-expressive and emotion. With additional in-domain pre-training, bidirectional encoder representations from Transformers (BERT) outperformed other deep learning models on the classification tasks. Content-specific SNA revealed that users' engagement or abstinence status is associated with the prevalence of various categories of BCTs and SAs, which also was evident from the visualization of network structures.

CONCLUSIONS

Our study describes the interplay of multilevel characteristics of online communication and their association with individual health behaviors.

Midwifery learning and forecasting: Predicting content demand with user-generated logs

Every day, 800 women and 6700 newborns die from complications related to pregnancy or childbirth. A well-trained midwife can prevent most of these maternal and newborn deaths. Data science models together with logs generated by users of online learning applications for midwives can help improve their learning competencies. In this work, we evaluate various forecasting methods to determine the future interest of users for the different types of content available in the Safe Delivery App, a digital training tool for skilled birth attendants, broken down by profession and region. This first attempt at health content demand forecasting for midwifery learning shows that DeepAR can accurately anticipate content demand in operational settings, and could therefore be used to offer users personalized content and to provide an adaptive learning journey.

Mood and implicit confidence independently fluctuate at different time scales

Mood is an important ingredient of decision-making. Human beings are immersed into a sea of ​​emotions where episodes of high mood alternate with episodes of low mood. While changes in mood are well characterized, little is known about how these fluctuations interact with metacognition, and in particular with confidence about our decisions. We evaluated how implicit measurements of confidence are related with mood states of human participants through two online longitudinal experiments involving mood self-reports and visual discrimination decision-making tasks. Implicit confidence was assessed on each session by monitoring the proportion of opt-out trials when an opt-out option was available, as well as the median reaction time on standard correct trials as a secondary proxy of confidence. We first report a strong coupling between mood, stress, food enjoyment, and quality of sleep reported by participants in the same session. Second, we confirmed that the proportion of opt-out responses as well as reaction times in non-opt-out trials provided reliable indices of confidence in each session. We introduce a normative measure of overconfidence based on the pattern of opt-out selection and the signal-detection-theory framework. Finally and crucially, we found that mood, sleep quality, food enjoyment, and stress level are not consistently coupled with these implicit confidence markers, but rather they fluctuate at different time scales: mood-related states display faster fluctuations (over one day or half-a-day) than confidence level (two-and-a-half days). Therefore, our findings suggest that spontaneous fluctuations of mood and confidence in decision making are independent in the healthy adult population.

Who uses mHealth apps? Identifying user archetypes of mHealth apps

Objective

This study aims to explore the user archetypes of health apps based on average usage and psychometrics.

Methods

The study utilized a dataset collected through a dedicated smartphone application and contained usage data, i.e. the timestamps of each app session from October 2020 to April 2021. The dataset had 129 participants for mental health apps usage and 224 participants for physical health apps usage. Average daily launches, extraversion, neuroticism, and satisfaction with life were the determinants of the mental health apps clusters, whereas average daily launches, conscientiousness, neuroticism, and satisfaction with life were for physical health apps.

Results

Two clusters of mental health apps users were identified using k-prototypes clustering: help-seeking and maintenance users and three clusters of physical health apps users were identified: happy conscious occasional, happy neurotic occasional, and unhappy neurotic frequent users.

Conclusion

The findings from this study helped to understand the users of health apps based on the frequency of usage, personality, and satisfaction with life. Further, with these findings, apps can be tailored to optimize user experience and satisfaction which may help to increase user retention. Policymakers may also benefit from these findings since understanding the populations' needs may help to better invest in effective health technology.

Linguistic predictors from Facebook postings of substance use disorder treatment retention versus discontinuation

Background: Early indicators of who will remain in - or leave - treatment for substance use disorder (SUD) can drive targeted interventions to support long-term recovery.Objectives: To conduct a comprehensive study of linguistic markers of SUD treatment outcomes, the current study integrated features produced by machine learning models known to have social-psychology relevance.Methods: We extracted and analyzed linguistic features from participants' Facebook posts (N = 206, 39.32% female; 55,415 postings) over the two years before they entered a SUD treatment program. Exploratory features produced by both Linguistic Inquiry and Word Count (LIWC) and Latent Dirichlet Allocation (LDA) topic modeling and the features from theoretical domains of religiosity, affect, and temporal orientation via established AI-based linguistic models were utilized.Results: Patients who stayed in the SUD treatment for over 90 days used more words associated with religion, positive emotions, family, affiliations, and the present, and used more first-person singular pronouns (Cohen's d values: [-0.39, -0.57]). Patients who discontinued their treatment before 90 days discussed more diverse topics, focused on the past, and used more articles (Cohen's d values: [0.44, 0.57]). All ps < .05 with Benjamini-Hochberg False Discovery Rate correction.Conclusions: We confirmed the literature on protective and risk social-psychological factors linking to SUD treatment in language analysis, showing that Facebook language before treatment entry could be used to identify the markers of SUD treatment outcomes. This reflects the importance of taking these linguistic features and markers into consideration when designing and recommending SUD treatment plans.

Bibliometric Analysis of Health Technology Research: 1990~2020

This paper aims to summarize the publishing trends, current status, research topics, and frontier evolution trends of health technology between 1990 and 2020 through various bibliometric analysis methods. In total, 6663 articles retrieved from the Web of Science core database were analyzed by Vosviewer and CiteSpace software. This paper found that: (1) The number of publications in the field of health technology increased exponentially; (2) there is no stable core group of authors in this research field, and the influence of the publishing institutions and journals in China is insufficient compared with those in Europe and the United States; (3) there are 21 core research topics in the field of health technology research, and these research topics can be divided into four classes: hot spots, potential hot spots, margin topics, and mature topics. C21 (COVID-19 prevention) and C10 (digital health technology) are currently two emerging research topics. (4) The number of research frontiers has increased in the past five years (2016-2020), and the research directions have become more diverse; rehabilitation, pregnancy, e-health, m-health, machine learning, and patient engagement are the six latest research frontiers.

Design and Methods of a Prospective Smartphone App-Based Study for Digital Phenotyping of Mood and Anxiety Symptoms Mixed With Centralized and Decentralized Research Form: The Search Your Mind (S.Y.M., ) Project

In this study, the Search Your Mind (S.Y.M., ) project aimed to collect prospective digital phenotypic data centered on mood and anxiety symptoms across psychiatric disorders through a smartphone application (app) platform while using both centralized and decentralized research designs: the centralized research design is a hybrid of a general prospective observational study and a digital platform-based study, and it includes face-to-face research such as informed written consent, clinical evaluation, and blood sampling. It also includes digital phenotypic assessment through an application-based platform using wearable devices. Meanwhile, the decentralized research design is a non-face-to-face study in which anonymous participants agree to electronic informed consent forms on the app. It also exclusively uses an application-based platform to acquire individualized digital phenotypic data. We expect to collect clinical, biological, and digital phenotypic data centered on mood and anxiety symptoms, and we propose a possible model of centralized and decentralized research design.

A Machine Learning Approach for Detecting Digital Behavioral Patterns of Depression Using Nonintrusive Smartphone Data (Complementary Path to Patient Health Questionnaire-9 Assessment): Prospective Observational Study

BACKGROUND

Depression is a major global cause of morbidity, an economic burden, and the greatest health challenge leading to chronic disability. Mobile monitoring of mental conditions has long been a sought-after metric to overcome the problems associated with the screening, diagnosis, and monitoring of depression and its heterogeneous presentation. The widespread availability of smartphones has made it possible to use their data to generate digital behavioral models that can be used for both clinical and remote screening and monitoring purposes. This study is novel as it adds to the field by conducting a trial using private and nonintrusive sensors that can help detect and monitor depression in a continuous, passive manner.

OBJECTIVE

This study demonstrates a novel mental behavioral profiling metric (the Mental Health Similarity Score), derived from analyzing passively monitored, private, and nonintrusive smartphone use data, to identify and track depressive behavior and its progression.

METHODS

Smartphone data sets and self-reported Patient Health Questionnaire-9 (PHQ-9) depression assessments were collected from 558 smartphone users on the Android operating system in an observational study over an average of 10.7 (SD 23.7) days. We quantified 37 digital behavioral markers from the passive smartphone data set and explored the relationship between the digital behavioral markers and depression using correlation coefficients and random forest models. We leveraged 4 supervised machine learning classification algorithms to predict depression and its severity using PHQ-9 scores as the ground truth. We also quantified an additional 3 digital markers from gyroscope sensors and explored their feasibility in improving the model's accuracy in detecting depression.

RESULTS

The PHQ-9 2-class model (none vs severe) achieved the following metrics: precision of 85% to 89%, recall of 85% to 89%, F1 of 87%, and accuracy of 87%. The PHQ-9 3-class model (none vs mild vs severe) achieved the following metrics: precision of 74% to 86%, recall of 76% to 83%, F1 of 75% to 84%, and accuracy of 78%. A significant positive Pearson correlation was found between PHQ-9 questions 2, 6, and 9 within the severely depressed users and the mental behavioral profiling metric (r=0.73). The PHQ-9 question-specific model achieved the following metrics: precision of 76% to 80%, recall of 75% to 81%, F1 of 78% to 89%, and accuracy of 78%. When a gyroscope sensor was added as a feature, the Pearson correlation among questions 2, 6, and 9 decreased from 0.73 to 0.46. The PHQ-9 2-class model+gyro features achieved the following metrics: precision of 74% to 78%, recall of 67% to 83%, F1 of 72% to 78%, and accuracy of 76%.

CONCLUSIONS

Our results demonstrate that the Mental Health Similarity Score can be used to identify and track depressive behavior and its progression with high accuracy.

The Feasibility and Utility of Harnessing Digital Health to Understand Clinical Trajectories in Medication Treatment for Opioid Use Disorder: D-TECT Study Design and Methodological Considerations

Introduction

Across the U.S., the prevalence of opioid use disorder (OUD) and the rates of opioid overdoses have risen precipitously in recent years. Several effective medications for OUD (MOUD) exist and have been shown to be life-saving. A large volume of research has identified a confluence of factors that predict attrition and continued substance use during substance use disorder treatment. However, much of this literature has examined a small set of potential moderators or mediators of outcomes in MOUD treatment and may lead to over-simplified accounts of treatment non-adherence. Digital health methodologies offer great promise for capturing intensive, longitudinal ecologically-valid data from individuals in MOUD treatment to extend our understanding of factors that impact treatment engagement and outcomes.

Methods

This paper describes the protocol (including the study design and methodological considerations) from a novel study supported by the National Drug Abuse Treatment Clinical Trials Network at the National Institute on Drug Abuse (NIDA). This study (D-TECT) primarily seeks to evaluate the feasibility of collecting ecological momentary assessment (EMA), smartphone and smartwatch sensor data, and social media data among patients in outpatient MOUD treatment. It secondarily seeks to examine the utility of EMA, digital sensing, and social media data (separately and compared to one another) in predicting MOUD treatment retention, opioid use events, and medication adherence [as captured in electronic health records (EHR) and EMA data]. To our knowledge, this is the first project to include all three sources of digitally derived data (EMA, digital sensing, and social media) in understanding the clinical trajectories of patients in MOUD treatment. These multiple data streams will allow us to understand the relative and combined utility of collecting digital data from these diverse data sources. The inclusion of EHR data allows us to focus on the utility of digital health data in predicting objectively measured clinical outcomes.

Discussion

Results may be useful in elucidating novel relations between digital data sources and OUD treatment outcomes. It may also inform approaches to enhancing outcomes measurement in clinical trials by allowing for the assessment of dynamic interactions between individuals' daily lives and their MOUD treatment response.

Clinical Trial Registration

Identifier: NCT04535583.

Intensive Longitudinal Data Collection Using Microinteraction Ecological Momentary Assessment: Pilot and Preliminary Results

Background

Ecological momentary assessment (EMA) uses mobile technology to enable in situ self-report data collection on behaviors and states. In a typical EMA study, participants are prompted several times a day to answer sets of multiple-choice questions. Although the repeated nature of EMA reduces recall bias, it may induce participation burden. There is a need to explore complementary approaches to collecting in situ self-report data that are less burdensome yet provide comprehensive information on an individual’s behaviors and states. A new approach, microinteraction EMA (μEMA), restricts EMA items to single, cognitively simple questions answered on a smartwatch with single-tap assessments using a quick, glanceable microinteraction. However, the viability of using μEMA to capture behaviors and states in a large-scale longitudinal study has not yet been demonstrated.

Objective

This paper describes the μEMA protocol currently used in the Temporal Influences on Movement & Exercise (TIME) Study conducted with young adults, the interface of the μEMA app used to gather self-report responses on a smartwatch, qualitative feedback from participants after a pilot study of the μEMA app, changes made to the main TIME Study μEMA protocol and app based on the pilot feedback, and preliminary μEMA results from a subset of active participants in the TIME Study.

Methods

The TIME Study involves data collection on behaviors and states from 246 individuals; measurements include passive sensing from a smartwatch and smartphone and intensive smartphone-based hourly EMA, with 4-day EMA bursts every 2 weeks. Every day, participants also answer a nightly EMA survey. On non–EMA burst days, participants answer μEMA questions on the smartwatch, assessing momentary states such as physical activity, sedentary behavior, and affect. At the end of the study, participants describe their experience with EMA and μEMA in a semistructured interview. A pilot study was used to test and refine the μEMA protocol before the main study.

Results

Changes made to the μEMA study protocol based on pilot feedback included adjusting the single-question selection method and smartwatch vibrotactile prompting. We also added sensor-triggered questions for physical activity and sedentary behavior. As of June 2021, a total of 81 participants had completed at least 6 months of data collection in the main study. For 662,397 μEMA questions delivered, the compliance rate was 67.6% (SD 24.4%) and the completion rate was 79% (SD 22.2%).

Conclusions

The TIME Study provides opportunities to explore a novel approach for collecting temporally dense intensive longitudinal self-report data in a sustainable manner. Data suggest that μEMA may be valuable for understanding behaviors and states at the individual level, thus possibly supporting future longitudinal interventions that require within-day, temporally dense self-report data as people go about their lives.

Feasibility and acceptability of a digital health intervention to promote engagement in and adherence to medication for opioid use disorder

BACKGROUND

Buprenorphine-naloxone is an evidence-based treatment for opioid use disorder (OUD). Despite its efficacy, nearly half of patients discontinue treatment prematurely. Novel intervention strategies that may be delivered outside of traditional treatment settings are needed to support buprenorphine uptake and maintenance. The goal of this study was to elucidate key elements surrounding the acceptability/feasibility and structure of an interactive computer- and text message-delivered personalized feedback intervention for adults initiating outpatient buprenorphine treatment.

METHODS

Twenty-four adults engaged in treatment at two outpatient addiction treatment centers completed semistructured interviews exploring preferences around digital health interventions. Trained interviewers conducted interviews, the study audio-recorded them, and a professional agency transcribed them verbatim. The research team iteratively developed a coding structure using thematic and content analysis and entered it into a framework matrix. The team double coded each transcript.

RESULTS

The sample was balanced by gender, primary type of opioid use (prescription pills; heroin/fentanyl), and phase of recovery [early (≤8 weeks of treatment) vs. late (>8 weeks of treatment)]. The study reached saturation after 24 interviews (mean age = 38.9; 70.8% white; 8.3% Hispanic/Latino). (1) Acceptability/feasibility themes: A computer- and text message-based intervention that incorporates a motivational- and distress tolerance-based framework is highly acceptable. Presentation of material, including the length of the intervention, is effective in facilitating learning. The center should offer the intervention to individuals entering treatment and they should have the flexibility to complete the intervention at the center or in private from their own home. The use of technology for intervention delivery helps to overcome fears of judgment stemming from stigmatizing experiences. (2) Structural themes: The text message intervention should deliver both predetermined (automatic) and on demand messages. Two to three messages per day (morning and early evening), with the option to elicit additional messages as needed, would be ideal. The messages must be personalized. Incorporating multimedia such as emojis, gifs, and links to videos will increase interactivity.

CONCLUSIONS

Overall, adults engaged in outpatient buprenorphine treatment were receptive to an interactive computer- and text messaged-delivered personalized feedback intervention to support recovery. Incorporating thematic results on suggested structural changes may increase the usability of this intervention to improve treatment outcomes by reducing illicit opioid use, increasing adherence/retention, and preventing future overdose and other complications of illicit opioid use.

Research on the Application of Intelligent Choreography for Musical Theater Based on Mixture Density Network Algorithm

Musical choreography is usually completed by professional choreographers, which is very professional and time-consuming. In order to realize the intelligent choreography of musical, based on the mixed density network (MDN), this paper generates the dance matching with the target music through three steps: motion generation, motion screening, and feature matching. The choreography results in this paper have a high degree of matching with music, which makes it possible for the development of motion capture technology and artificial intelligence and computer automatic choreography based on music. In the process of motion generation, the average value of Gaussian model output by MDN is used as the bone position and the consistency of motion is measured according to the change rate of joint velocity in adjacent frames in the process of motion selection. Compared with the existing studies, the dance generated in this paper has improved in motion coherence and realism. In this paper, a multilevel music and action feature matching algorithm combining global feature matching and local feature matching is proposed. The algorithm improves the unity and coherence of music and action. The algorithm proposed in this paper improves the consistency and novelty of movement, the compatibility with music, and the controllability of dance characteristics. Therefore, the algorithm in this paper technically changes the way of artistic creation and provides the possibility for the development of motion capture technology and artificial intelligence.

Pragmatics to Reveal Intent in Social Media Peer Interactions: Mixed Methods Study

Background

Online health communities (OHCs) have emerged as the leading venues for behavior change and health-related information seeking. The soul and success of these digital platforms lie in their ability to foster social togetherness and a sense of community by providing personalized support. However, we have a minimal understanding of how conversational posts in these settings lead to collaborative societies and ultimately result in positive health changes through social influence.

Objective

Our objective is to develop a content-specific and intent-sensitive methodological framework for analyzing peer interactions in OHCs.

Methods

We developed and applied a mixed-methods approach to understand the manifestation of expressions in peer interactions in OHCs. We applied our approach to describe online social dialogue in the context of two online communities, QuitNet (QN) and the American Diabetes Association (ADA) support community. A total of 3011 randomly selected peer interactions (n=2005 from QN, n=1006 from ADA) were analyzed. Specifically, we conducted thematic analysis to characterize communication content and linguistic expressions (speech acts) embedded within the two data sets. We also developed an empirical user persona based on their engagement levels and behavior profiles. Further, we examined the association between speech acts and communication themes across observed tiers of user engagement and self-reported behavior profiles using the chi-square test or the Fisher test.

Results

Although social support, the most prevalent communication theme in both communities, was expressed in several subtle manners, the prevalence of emotions was higher in the tobacco cessation community and assertions were higher in the diabetes self-management (DSM) community. Specific communication theme-speech act relationships were revealed, such as the social support theme was significantly associated (P<.05) with 9 speech acts from a total of 10 speech acts (ie, assertion, commissive, declarative, desire, directive, expressive, question, stance, and statement) within the QN community. Only four speech acts (ie, commissive, emotion, expressive, and stance) were significantly associated (P<.05) with the social support theme in the ADA community. The speech acts were also significantly associated with the users’ abstinence status within the QN community and with the users’ lifestyle status within the ADA community (P<.05).

Conclusions

Such an overlay of communication intent implicit in online peer interactions alongside content-specific theory-linked characterizations of social media discourse can inform the development of effective digital health technologies in the field of health promotion and behavior change. Our analysis revealed a rich gradient of expressions across a standardized thematic vocabulary, with a distinct variation in emotional and informational needs, depending on the behavioral and disease management profiles within and across the communities. This signifies the need and opportunities for coupling pragmatic messaging in digital therapeutics and care management pathways for personalized support.

Intelligent health system for the investigation of consenting COVID-19 patients and precision medicine

Advancing frontiers of clinical research, we discuss the need for intelligent health systems to support a deeper investigation of COVID-19. We hypothesize that the convergence of the healthcare data and staggering developments in artificial intelligence have the potential to elevate the recovery process with diagnostic and predictive analysis to identify major causes of mortality, modifiable risk factors and actionable information that supports the early detection and prevention of COVID-19. However, current constraints include the recruitment of COVID-19 patients for research; translational integration of electronic health records and diversified public datasets; and the development of artificial intelligence systems for data-intensive computational modeling to assist clinical decision making. We propose a novel nexus of machine learning algorithms to examine COVID-19 data granularity from population studies to subgroups stratification and ensure best modeling strategies within the data continuum.

A Machine Learning Approach to Passively Informed Prediction of Mental Health Risk in People with Diabetes: Retrospective Case-Control Analysis

Background

Proactive detection of mental health needs among people with diabetes mellitus could facilitate early intervention, improve overall health and quality of life, and reduce individual and societal health and economic burdens. Passive sensing and ecological momentary assessment are relatively newer methods that may be leveraged for such proactive detection.

Objective

The primary aim of this study was to conceptualize, develop, and evaluate a novel machine learning approach for predicting mental health risk in people with diabetes mellitus.

Methods

A retrospective study was designed to develop and evaluate a machine learning model, utilizing data collected from 142,432 individuals with diabetes enrolled in the Livongo for Diabetes program. First, participants’ mental health statuses were verified using prescription and medical and pharmacy claims data. Next, four categories of passive sensing signals were extracted from the participants’ behavior in the program, including demographics and glucometer, coaching, and event data. Data sets were then assembled to create participant-period instances, and descriptive analyses were conducted to understand the correlation between mental health status and passive sensing signals. Passive sensing signals were then entered into the model to train and test its performance. The model was evaluated based on seven measures: sensitivity, specificity, precision, area under the curve, F₁ score, accuracy, and confusion matrix. SHapley Additive exPlanations (SHAP) values were computed to determine the importance of individual signals.

Results

In the training (and validation) and three subsequent test sets, the model achieved a confidence score greater than 0.5 for sensitivity, specificity, area under the curve, and accuracy. Signals identified as important by SHAP values included demographics such as race and gender, participant’s emotional state during blood glucose checks, time of day of blood glucose checks, blood glucose values, and interaction with the Livongo mobile app and web platform.

Conclusions

Results of this study demonstrate the utility of a passively informed mental health risk algorithm and invite further exploration to identify additional signals and determine when and where such algorithms should be deployed.

Beyond abstinence and relapse II: momentary relationships between stress, craving, and lapse within clusters of patients with similar patterns of drug use

RATIONALE

Given that many patients being treated for opioid-use disorder continue to use drugs, identifying clusters of patients who share similar patterns of use might provide insight into the disorder, the processes that affect it, and ways that treatment can be personalized.

OBJECTIVES AND METHODS

We applied hierarchical clustering to identify patterns of opioid and cocaine use in 309 participants being treated with methadone or buprenorphine (in a buprenorphine-naloxone formulation) for up to 16 weeks. A smartphone app was used to assess stress and craving at three random times per day over the course of the study.

RESULTS

Five basic patterns of use were identified: frequent opioid use, frequent cocaine use, frequent dual use (opioids and cocaine), sporadic use, and infrequent use. These patterns were differentially associated with medication (methadone vs. buprenorphine), race, age, drug-use history, drug-related problems prior to the study, stress-coping strategies, specific triggers of use events, and levels of cue exposure, craving, and negative mood. Craving tended to increase before use in all except those who used sporadically. Craving was sharply higher during the 90 min following moderate-to-severe stress in those with frequent use, but only moderately higher in those with infrequent or sporadic use.

CONCLUSIONS

People who share similar patterns of drug-use during treatment also tend to share similarities with respect to psychological processes that surround instances of use, such as stress-induced craving. Cluster analysis combined with smartphone-based experience sampling provides an effective strategy for studying how drug use is related to personal and environmental factors.

Leveraging applied behavior analysis research and practice in the service of public health

Human behavior plays a central role in all domains of public health. Applied behavior analysis (ABA) research and practice can contribute to public health solutions that directly address human behavior. In this paper, we describe the field of public health, identify points of interaction between public health and ABA, summarize what ABA research has already contributed, and provide several recommendations for how ABA research and practice could continue to promote public health outcomes. A clearer focus on behavior and widespread adoption of research designs and interventions informed by the ABA literature could lead to better public health outcomes. Reciprocally, better integration of public health goals and strategies into ABA research, harnessing of technology, and more collaboration would help diversify and disseminate our applied science and could yield more effective and scalable interventions to prevent and treat public health problems.

HOPES: An Integrative Digital Phenotyping Platform for Data Collection, Monitoring, and Machine Learning

The collection of data from a personal digital device to characterize current health conditions and behaviors that determine how an individual's health will evolve has been called digital phenotyping. In this paper, we describe the development of and early experiences with a comprehensive digital phenotyping platform: Health Outcomes through Positive Engagement and Self-Empowerment (HOPES). HOPES is based on the open-source Beiwe platform but adds a wider range of data collection, including the integration of wearable devices and further sensor collection from smartphones. Requirements were partly derived from a concurrent clinical trial for schizophrenia that required the development of significant capabilities in HOPES for security, privacy, ease of use, and scalability, based on a careful combination of public cloud and on-premises operation. We describe new data pipelines to clean, process, present, and analyze data. This includes a set of dashboards customized to the needs of research study operations and clinical care. A test use case for HOPES was described by analyzing the digital behavior of 22 participants during the SARS-CoV-2 pandemic.

Pain Sensitivity in Schizophrenia Spectrum Disorders: A Narrative Review of Recent Work

Many patients with schizophrenia seem relatively immune to physical pain while others complain of constant pain. This may result from disturbances or alterations of the sensory threshold for pain in populations with psychosis, a possibility for which there is some preliminary evidence. The inconsistency in pain perception may, in part, be explained by the treatments patients receive, but treatment-naïve patients also exhibit differences in response to pain. This suggests that decreased pain sensitivity may represent a specific psychosis endophenotype. Thus far, few experimental studies have investigated sensory thresholds, pain modalities, or other factors contributing to the perception or expression of physical pain in psychosis. A digital search for information on this topic was conducted in PubMed and Google Scholar. The result is a non-systematic, narrative review focusing on recent clinical and experimental findings of pain sensitivity in patients with psychosis. Importantly, physical and mental pain are closely connected constructs that may be difficult to differentiate. Our hope is that the review provides some clarity to the field in the specific context of schizophrenia.

Digital Phenotyping and Patient-Generated Health Data for Outcome Measurement in Surgical Care: A Scoping Review

Digital phenotyping-the moment-by-moment quantification of human phenotypes in situ using data related to activity, behavior, and communications, from personal digital devices, such as smart phones and wearables-has been gaining interest. Personalized health information captured within free-living settings using such technologies may better enable the application of patient-generated health data (PGHD) to provide patient-centered care. The primary objective of this scoping review is to characterize the application of digital phenotyping and digitally captured active and passive PGHD for outcome measurement in surgical care. Secondarily, we synthesize the body of evidence to define specific areas for further work. We performed a systematic search of four bibliographic databases using terms related to "digital phenotyping and PGHD," "outcome measurement," and "surgical care" with no date limits. We registered the study (Open Science Framework), followed strict inclusion/exclusion criteria, performed screening, extraction, and synthesis of results in line with the PRISMA Extension for Scoping Reviews. A total of 224 studies were included. Published studies have accelerated in the last 5 years, originating in 29 countries (mostly from the USA, n = 74, 33%), featuring original prospective work (n = 149, 66%). Studies spanned 14 specialties, most commonly orthopedic surgery (n = 129, 58%), and had a postoperative focus (n = 210, 94%). Most of the work involved research-grade wearables (n = 130, 58%), prioritizing the capture of activity (n = 165, 74%) and biometric data (n = 100, 45%), with a view to providing a tracking/monitoring function (n = 115, 51%) for the management of surgical patients. Opportunities exist for further work across surgical specialties involving smartphones, communications data, comparison with patient-reported outcome measures (PROMs), applications focusing on prediction of outcomes, monitoring, risk profiling, shared decision making, and surgical optimization. The rapidly evolving state of the art in digital phenotyping and capture of PGHD offers exciting prospects for outcome measurement in surgical care pending further work and consideration related to clinical care, technology, and implementation.