Return of the JITAI: Applying a Just-in-Time Adaptive Intervention Framework to the Development of m-Health Solutions for Addictive Behaviors

Choosing Health: acceptability and feasibility of a theory-based, online-delivered, tailored weight loss, and weight loss maintenance intervention

Few weight loss and weight loss maintenance interventions are tailored to include factors demonstrated to predict the user's behavior. Establishing the feasibility and acceptability of such interventions is crucial. The aim of this study was to assess the acceptability and feasibility of a theory-based, tailored, online-delivered weight loss and weight loss maintenance intervention (Choosing Health). We conducted a mixed methods process evaluation of the Choosing Health tailored intervention, nested in a randomized controlled trial (N = 288) with an embedded N-of-1 study, investigating participants' and implementers' experiences related to intervention context, implementation, and mechanisms of impact. Measures included: (i) surveys, (ii) data-prompted interviews (DPIs) with study participants, (iii) semi-structured interviews with implementers, and (iv) intervention access and engagement data. Five themes described the acceptability of the intervention to participants: (i) monitoring behavior change and personal progress to better understand the weight management process, (ii) working collaboratively with the intervention implementers to achieve participants' goals, (iii) perceived benefits of non-judgmental and problem-solving tone of the intervention, (iv) changes in personal perception of the weight management process due to intervention tailoring, and (v) insufficient intervention content tailoring. The intervention delivery was feasible, however, emails and text messages differed in terms of accessibility and resources required to deliver the content. The use of Ecological Momentary Assessment as a technique to gather personal data for further tailoring was acceptable, and facilitated behavior change monitoring. Personalization of the intervention content above and beyond domain-specific issues, for example, by addressing participants' social roles may better match their needs. Support from the implementers and feedback on body composition changes may increase participants' engagement.

Applying the Just-In-Time Adaptive Intervention Framework to the Development of Gambling Interventions

Just-In-Time Adaptive Interventions (JITAIs) are emerging "push" mHealth interventions that provide the right type, timing, and amount of support to address the dynamically-changing needs for each individual. Although JITAIs are well-suited to the delivery of interventions for the addictions, few are available to support gambling behaviour change. We therefore developed GamblingLess: In-The-Moment and Gambling Habit Hacker, two smartphone-delivered JITAIs that differ with respect to their target populations, theoretical underpinnings, and decision rules. We aim to describe the decisions, methods, and tools we used to design these two treatments, with a view to providing guidance to addiction researchers who wish to develop JITAIs in the future. Specifically, we describe how we applied a comprehensive, organising scientific framework to define the problem, define just-in-time in the context of the identified problem, and formulate the adaptation strategies. While JITAIs appear to be a promising design in addiction intervention science, we describe several key challenges that arose during development, particularly in relation to applying micro-randomised trials to their evaluation, and offer recommendations for future research. Issues including evaluation considerations, integrating on-demand intervention content, intervention optimisation, combining active and passive assessments, incorporating human facilitation, adding cost-effectiveness evaluations, and redevelopment as transdiagnostic interventions are discussed.

Smartphone Apps for Food Purchase Choices: Scoping Review of Designs, Opportunities, and Challenges

BACKGROUND

Smartphone apps can aid consumers in making healthier and more sustainable food purchases. However, there is still a limited understanding of the different app design approaches and their impact on food purchase choices. An overview of existing food purchase choice apps and an understanding of common challenges can help speed up effective future developments.

OBJECTIVE

We examined the academic literature on food purchase choice apps and provided an overview of the design characteristics, opportunities, and challenges for effective implementation. Thus, we contribute to an understanding of how technologies can effectively improve food purchase choice behavior and provide recommendations for future design efforts.

METHODS

Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines, we considered peer-reviewed literature on food purchase choice apps within IEEE Xplore, PubMed, Scopus, and ScienceDirect. We inductively coded and summarized design characteristics. Opportunities and challenges were addressed from both quantitative and qualitative perspectives. From the quantitative perspective, we coded and summarized outcomes of comparative evaluation trials. From the qualitative perspective, we performed a qualitative content analysis of commonly discussed opportunities and challenges.

RESULTS

We retrieved 55 articles, identified 46 unique apps, and grouped them into 5 distinct app types. Each app type supports a specific purchase choice stage and shares a common functional design. Most apps support the product selection stage (selection apps; 27/46, 59%), commonly by scanning the barcode and displaying a nutritional rating. In total, 73% (8/11) of the evaluation trials reported significant findings and indicated the potential of food purchase choice apps to support behavior change. However, relatively few evaluations covered the selection app type, and these studies showed mixed results. We found a common opportunity in apps contributing to learning (knowledge gain), whereas infrequent engagement presents a common challenge. The latter was associated with perceived burden of use, trust, and performance as well as with learning. In addition, there were technical challenges in establishing comprehensive product information databases or achieving performance accuracy with advanced identification methods such as image recognition.

CONCLUSIONS

Our findings suggest that designs of food purchase choice apps do not encourage repeated use or long-term adoption, compromising the effectiveness of behavior change through nudging. However, we found that smartphone apps can enhance learning, which plays an important role in behavior change. Compared with nudging as a mechanism for behavior change, this mechanism is less dependent on continued use. We argue that designs that optimize for learning within each interaction have a better chance of achieving behavior change. This review concludes with design recommendations, suggesting that food purchase choice app designers anticipate the possibility of early abandonment as part of their design process and design apps that optimize the learning experience.

Just-in-time adaptive ecological momentary assessment (JITA-EMA)

Interest in just-in-time adaptive interventions (JITAI) has rapidly increased in recent years. One core challenge for JITAI is the efficient and precise measurement of tailoring variables that are used to inform the timing of momentary intervention delivery. Ecological momentary assessment (EMA) is often used for this purpose, even though EMA in its traditional form was not designed specifically to facilitate momentary interventions. In this article, we introduce just-in-time adaptive EMA (JITA-EMA) as a strategy to reduce participant response burden and decrease measurement error when EMA is used as a tailoring variable in JITAI. JITA-EMA builds on computerized adaptive testing methods developed for purposes of classification (computerized classification testing, CCT), and applies them to the classification of momentary states within individuals. The goal of JITA-EMA is to administer a small and informative selection of EMA questions needed to accurately classify an individual's current state at each measurement occasion. After illustrating the basic components of JITA-EMA (adaptively choosing the initial and subsequent items to administer, adaptively stopping item administration, accommodating dynamically tailored classification cutoffs), we present two simulation studies that explored the performance of JITA-EMA, using the example of momentary fatigue states. Compared with conventional EMA item selection methods that administered a fixed set of questions at each moment, JITA-EMA yielded more accurate momentary classification with fewer questions administered. Our results suggest that JITA-EMA has the potential to enhance some approaches to mobile health interventions by facilitating efficient and precise identification of momentary states that may inform intervention tailoring.

Challenges and opportunities in mobile e-coaching

Background

Mobile e-health technologies have proven to provide tailored assessment, intervention, and coaching capabilities for various usage scenarios. Thanks to their spread and adoption, smartphones are one of the most important carriers for such applications.

Problem

However, the process of design, realization, evaluation, and implementation of these e-health solutions is wicked and challenging, requiring multiple stakeholders and expertise.

Method

Here, we present a tailorable intervention and interaction e-health solution that allows rapid prototyping, development, and evaluation of e-health interventions at scale. This platform allows researchers and clinicians to develop ecological momentary assessment, just-in-time adaptive interventions, ecological momentary intervention, cohort studies, and e-coaching and personalized interventions quickly, with no-code, and in a scalable way.

Result

The Twente Intervention and Interaction Instrument (TIIM) has been used by over 320 researchers in the last decade. We present the ecosystem and synthesize the main scientific output from clinical and research studies in different fields.

Discussion

The importance of mobile e-coaching for prediction, management, and prevention of adverse health outcomes is increasing. A profound e-health development strategyand strategic, technical, and operational investments are needed to prototype, develop, implement, and evaluate e-health solutions. TIIM ecosystem has proven to support these processes. This paper ends with the main research opportunities in mobile coaching, including intervention mechanisms, fine-grained monitoring, and inclusion of objective biomarker data.

Associations of daily alcohol, cannabis, combustible tobacco, and e-cigarette use with same-day co-use and poly-use of the other substances

BACKGROUND

Young adults frequently use alcohol, cannabis, and tobacco together. Given the increased prevalence of e-cigarette use and recreational cannabis use, we investigated daily patterns of alcohol, cannabis, and tobacco use and distinguished combustible tobacco from e-cigarettes.

METHODS

Young adult college students (N=341) reporting past-month alcohol and cannabis use "at the same time so that their effects overlapped" completed two 28-day bursts of repeated daily surveys. Exposures were day- and person-level use of each substance. Outcomes were (1) same-day co-use of each remaining substance or (2) poly-use of the other substances.

RESULTS

Daily use of alcohol, cannabis, combustible cigarettes, and e-cigarettes increased the odds of same-day co-use of the other substances (except combustible tobacco with e-cigarettes) and each poly-use outcome. The influence of person-level substance use on daily substance use was less consistent. Only e-cigarette use increased the odds of daily alcohol use. Use of either tobacco product but not alcohol increased the odds of daily cannabis use. Person-level alcohol and cannabis use increased the odds of daily use of either tobacco product but use of one tobacco product was not associated with daily use of the other product.

CONCLUSIONS

These findings increase our understanding of emerging daily patterns of alcohol, cannabis, and tobacco co-use, and the impact of different tobacco products. Future work is needed to extend this research into non-college samples and people who use tobacco but do not use alcohol and cannabis simultaneously, and examine daily chronologies of multiple substances that could serve as dynamic markers of risk.

Classification of Lapses in Smokers Attempting to Stop: A Supervised Machine Learning Approach Using Data From a Popular Smoking Cessation Smartphone App

INTRODUCTION

Smoking lapses after the quit date often lead to full relapse. To inform the development of real time, tailored lapse prevention support, we used observational data from a popular smoking cessation app to develop supervised machine learning algorithms to distinguish lapse from non-lapse reports.

AIMS AND METHODS

We used data from app users with ≥20 unprompted data entries, which included information about craving severity, mood, activity, social context, and lapse incidence. A series of group-level supervised machine learning algorithms (eg, Random Forest, XGBoost) were trained and tested. Their ability to classify lapses for out-of-sample (1) observations and (2) individuals were evaluated. Next, a series of individual-level and hybrid algorithms were trained and tested.

RESULTS

Participants (N = 791) provided 37 002 data entries (7.6% lapses). The best-performing group-level algorithm had an area under the receiver operating characteristic curve (AUC) of 0.969 (95% confidence interval [CI] = 0.961 to 0.978). Its ability to classify lapses for out-of-sample individuals ranged from poor to excellent (AUC = 0.482-1.000). Individual-level algorithms could be constructed for 39/791 participants with sufficient data, with a median AUC of 0.938 (range: 0.518-1.000). Hybrid algorithms could be constructed for 184/791 participants and had a median AUC of 0.825 (range: 0.375-1.000).

CONCLUSIONS

Using unprompted app data appeared feasible for constructing a high-performing group-level lapse classification algorithm but its performance was variable when applied to unseen individuals. Algorithms trained on each individual's dataset, in addition to hybrid algorithms trained on the group plus a proportion of each individual's data, had improved performance but could only be constructed for a minority of participants.

IMPLICATIONS

This study used routinely collected data from a popular smartphone app to train and test a series of supervised machine learning algorithms to distinguish lapse from non-lapse events. Although a high-performing group-level algorithm was developed, it had variable performance when applied to new, unseen individuals. Individual-level and hybrid algorithms had somewhat greater performance but could not be constructed for all participants because of the lack of variability in the outcome measure. Triangulation of results with those from a prompted study design is recommended prior to intervention development, with real-world lapse prediction likely requiring a balance between unprompted and prompted app data.

Investigating the Role of Context in the Delivery of Text Messages for Supporting Psychological Wellbeing

Without a nuanced understanding of users' perspectives and contexts, text messaging tools for supporting psychological wellbeing risk delivering interventions that are mismatched to users' dynamic needs. We investigated the contextual factors that influence young adults' day-to-day experiences when interacting with such tools. Through interviews and focus group discussions with 36 participants, we identified that people's daily schedules and affective states were dominant factors that shape their messaging preferences. We developed two messaging dialogues centered around these factors, which we deployed to 42 participants to test and extend our initial understanding of users' needs. Across both studies, participants provided diverse opinions of how they could be best supported by messages, particularly around when to engage users in more passive versus active ways. They also proposed ways of adjusting message length and content during periods of low mood. Our findings provide design implications and opportunities for context-aware mental health management systems.

Effectiveness and Minimum Effective Dose of App-Based Mobile Health Interventions for Anxiety and Depression Symptom Reduction: Systematic Review and Meta-Analysis

BACKGROUND

Mobile health (mHealth) apps offer new opportunities to deliver psychological treatments for mental illness in an accessible, private format. The results of several previous systematic reviews support the use of app-based mHealth interventions for anxiety and depression symptom management. However, it remains unclear how much or how long the minimum treatment "dose" is for an mHealth intervention to be effective. Just-in-time adaptive intervention (JITAI) has been introduced in the mHealth domain to facilitate behavior changes and is positioned to guide the design of mHealth interventions with enhanced adherence and effectiveness.

OBJECTIVE

Inspired by the JITAI framework, we conducted a systematic review and meta-analysis to evaluate the dose effectiveness of app-based mHealth interventions for anxiety and depression symptom reduction.

METHODS

We conducted a literature search on 7 databases (ie, Ovid MEDLINE, Embase, PsycInfo, Scopus, Cochrane Library (eg, CENTRAL), ScienceDirect, and ClinicalTrials, for publications from January 2012 to April 2020. We included randomized controlled trials (RCTs) evaluating app-based mHealth interventions for anxiety and depression. The study selection and data extraction process followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We estimated the pooled effect size using Hedge g and appraised study quality using the revised Cochrane risk-of-bias tool for RCTs.

RESULTS

We included 15 studies involving 2627 participants for 18 app-based mHealth interventions. Participants in the intervention groups showed a significant effect on anxiety (Hedge g=-.10, 95% CI -0.14 to -0.06, I2=0%) but not on depression (Hedge g=-.08, 95% CI -0.23 to 0.07, I2=4%). Interventions of at least 7 weeks' duration had larger effect sizes on anxiety symptom reduction.

CONCLUSIONS

There is inconclusive evidence for clinical use of app-based mHealth interventions for anxiety and depression at the current stage due to the small to nonsignificant effects of the interventions and study quality concerns. The recommended dose of mHealth interventions and the sustainability of intervention effectiveness remain unclear and require further investigation.

Hybrid Experimental Designs for Intervention Development: What, Why, and How

Advances in mobile and wireless technologies offer tremendous opportunities for extending the reach and impact of psychological interventions and for adapting interventions to the unique and changing needs of individuals. However, insufficient engagement remains a critical barrier to the effectiveness of digital interventions. Human delivery of interventions (e.g., by clinical staff) can be more engaging but potentially more expensive and burdensome. Hence, the integration of digital and human-delivered components is critical to building effective and scalable psychological interventions. Existing experimental designs can be used to answer questions either about human-delivered components that are typically sequenced and adapted at relatively slow timescales (e.g., monthly) or about digital components that are typically sequenced and adapted at much faster timescales (e.g., daily). However, these methodologies do not accommodate sequencing and adaptation of components at multiple timescales and hence cannot be used to empirically inform the joint sequencing and adaptation of human-delivered and digital components. Here, we introduce the hybrid experimental design (HED)-a new experimental approach that can be used to answer scientific questions about building psychological interventions in which human-delivered and digital components are integrated and adapted at multiple timescales. We describe the key characteristics of HEDs (i.e., what they are), explain their scientific rationale (i.e., why they are needed), and provide guidelines for their design and corresponding data analysis (i.e., how can data arising from HEDs be used to inform effective and scalable psychological interventions).

GAMBLINGLESS IN-THE-MOMENT: Protocol for a micro-randomised trial of a gambling Just-In-Time Adaptive Intervention (Preprint)

Background

The presence of discrete but fluctuating precipitants, in combination with the dynamic nature of gambling episodes, calls for the development of tailored interventions delivered in real time, such as just-in-time adaptive interventions (JITAIs). JITAIs leverage mobile and wireless technologies to address dynamically changing individual needs by providing the type and amount of support required at the right time and only when needed. They have the added benefit of reaching underserved populations by providing accessible, convenient, and low-burden support. Despite these benefits, few JITAIs targeting gambling behavior are available.

Objective

This study aims to redress this gap in service provision by developing and evaluating a theoretically informed and evidence-based JITAI for people who want to reduce their gambling. Delivered via a smartphone app, GamblingLess: In-The-Moment provides tailored cognitive-behavioral and third-wave interventions targeting cognitive processes explicated by the relapse prevention model (cravings, self-efficacy, and positive outcome expectancies). It aims to reduce gambling symptom severity (distal outcome) through short-term reductions in the likelihood of gambling episodes (primary proximal outcome) by improving craving intensity, self-efficacy, or expectancies (secondary proximal outcomes). The primary aim is to explore the degree to which the delivery of a tailored intervention at a time of cognitive vulnerability reduces the probability of a subsequent gambling episode.

Methods

GamblingLess: In-The-Moment interventions are delivered to gamblers who are in a state of receptivity (available for treatment) and report a state of cognitive vulnerability via ecological momentary assessments 3 times a day. The JITAI will tailor the type, timing, and amount of support for individual needs. Using a microrandomized trial, a form of sequential factorial design, each eligible participant will be randomized to a tailored intervention condition or no intervention control condition at each ecological momentary assessment across a 28-day period. The microrandomized trial will be supplemented by a 6-month within-group follow-up evaluation to explore long-term effects on primary (gambling symptom severity) and secondary (gambling behavior, craving severity, self-efficacy, and expectancies) outcomes and an acceptability evaluation via postintervention surveys, app use and engagement indices, and semistructured interviews. In all, 200 participants will be recruited from Australia and New Zealand.

Results

The project was funded in June 2019, with approval from the Deakin University Human Research Ethics Committee (2020-304). Stakeholder user testing revealed high acceptability scores. The trial began on March 29, 2022, and 84 participants have been recruited (as of June 24, 2022). Results are expected to be published mid-2024.

Conclusions

GamblingLess: In-The-Moment forms part of a suite of theoretically informed and evidence-based web-based and mobile gambling interventions. This trial will provide important empirical data that can be used to facilitate the JITAI’s optimization to make it a more effective, efficient, and scalable tailored intervention.

Trial Registration

Australian New Zealand Clinical Trials Registry (ANZCTR) ACTRN12622000490774; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=380757&isClinicalTrial=False

International Registered Report Identifier (IRRID)

PRR1-10.2196/38958

Gambling Habit Hacker: Protocol for a micro-randomised trial of planning interventions delivered via a Just-In-Time Adaptive Intervention for adult gamblers (Preprint)

Background

People with gambling problems frequently report repeated unsuccessful attempts to change their behavior. Although many behavior change techniques are available to individuals to reduce gambling harm, they can be challenging to implement or maintain. The provision of implementation support tailored for immediate, real-time, individualized circumstances may improve attempts at behavior change.

Objective

We aimed to develop and evaluate a Just-In-Time Adaptive Intervention (JITAI) for individuals who require support to adhere to their gambling limits. JITAI development is based on the principles of the Health Action Process Approach with delivery, in alignment with the principles of self-determination theory. The primary objective was to determine the effect of action- and coping planning compared with no intervention on the goal of subsequently adhering to gambling expenditure limits.

Methods

Gambling Habit Hacker is delivered as a JITAI providing in-the-moment support for adhering to gambling expenditure limits (primary proximal outcome). Delivered via a smartphone app, this JITAI delivers tailored behavior change techniques related to goal setting, action planning, coping planning, and self-monitoring. The Gambling Habit Hacker app will be evaluated using a 28-day microrandomized trial. Up to 200 individuals seeking support for their own gambling from Australia and New Zealand will set a gambling expenditure limit (ie, goal). They will then be asked to complete 3 time-based ecological momentary assessments (EMAs) per day over a 28-day period. EMAs will assess real-time adherence to gambling limits, strength of intention to adhere to goals, goal self-efficacy, urge self-efficacy, and being in high-risk situations. On the basis of the responses to each EMA, participants will be randomized to the control (a set of 25 self-enactable strategies containing names only and no implementation information) or intervention (self-enactable strategy implementation information with facilitated action- and coping planning) conditions. This microrandomized trial will be supplemented with a 6-month within-group follow-up that explores the long-term impact of the app on gambling expenditure (primary distal outcome) and a range of secondary outcomes, as well as an evaluation of the acceptability of the JITAI via postintervention surveys, app use and engagement indices, and semistructured interviews. This trial has been approved by the Deakin University Human Research Ethics Committee (2020-304).

Results

The intervention has been subject to expert user testing, with high acceptability scores. The results will inform a more nuanced version of the Gambling Habit Hacker app for wider use.

Conclusions

Gambling Habit Hacker is part of a suite of interventions for addictive behaviors that deliver implementation support grounded in lived experience. This study may inform the usefulness of delivering implementation intentions in real time and in real-world settings. It potentially offers people with gambling problems new support to set their gambling intentions and adhere to their limits.

Trial Registration

Australian New Zealand Clinical Trials Registry ACTRN12622000497707; www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=383568

International Registered Report Identifier (IRRID)

DERR1-10.2196/38919

Associations among affect, diet, and activity and binge-eating severity using ecological momentary assessment in a non-clinical sample of middle-aged fathers

PURPOSE

Knowledge of within-day factors associated with binge-eating severity among middle-aged fathers is limited. The purpose of the current report was to examine within-day associations of affect, diet, and activity in relation to binge-eating severity using ecological momentary assessment (EMA) in men.

METHODS

Twenty-three middle-aged fathers completed 8 days of EMA and wore accelerometers to objectively assess activity. Generalized estimating equations assessed relationships among affect, diet, and activity and binge-eating severity.

RESULTS

When positive affect was above average, men reported greater binge-eating severity in the next 2 h. Oppositely, when negative affect was above average, men reported less binge-eating severity in the next 2 h. At times when men reported consumption of sweets and fast food, they reported higher binge-eating severity during the same 2-h window. Men with greater average levels of light activity reported less overall binge-eating severity.

CONCLUSIONS

Findings show that affect, unhealthy food intake, and light activity could be targeted among middle-aged fathers to reduce binge-eating severity and prevent eating disorders.

LEVEL OF EVIDENCE

Level III: Evidence obtained from cohort or case-control analytic studies.

Dietary lapses are associated with meaningful elevations in daily caloric intake and added sugar consumption during a lifestyle modification intervention

Objective

Lapses from the dietary prescription in lifestyle modification interventions for overweight/obesity are common and impact weight loss outcomes. While it is expected that lapses influence weight via increased consumption, there are no studies that have evaluated how dietary lapses affect dietary intake during treatment. This study examined the association between daily lapses and daily energy and macronutrient intake during a lifestyle modification intervention.

Methods

This study used an intensive longitudinal design to observe participants throughout a 6-month lifestyle modification intervention. Participants (n = 32) were adults with overweight/obesity (body mass index 25-50 kg/m²) and a diagnosed cardiovascular disease risk factor (e.g., hypertension) with a desire to lose weight. Participants underwent a gold-standard individual in-person lifestyle modification protocol consisting of 3 months of weekly sessions with 3 months of monthly sessions. Each participant's dietary prescription included a calorie target range that was based on their starting weight. Participants completed ecological momentary assessment (EMA; repeated daily smartphone surveys) every other week to self-report on dietary lapses and telephone-based 24-h dietary recalls every 6 weeks.

Results

On days with EMA and recalled intake (n = 210 days), linear mixed models demonstrated significant associations between daily lapse and higher total daily caloric intake (B = 139.20, p < 0.05), more daily grams of added sugar (B = 16.24, p < 0.001), and likelihood of exceeding the daily calorie goal (B = 0.89, p < 0.05). The associations between daily lapse and intake of all other daily macronutrients were non-significant.

Conclusions

This study contributes to literature suggesting that dietary lapses pose a threat to weight loss success. Results indicate that reducing lapse frequency could reduce overall caloric intake and added sugar consumption.

Optimizing a Just-in-Time Adaptive Intervention to Improve Dietary Adherence in Behavioral Obesity Treatment: Protocol for a Microrandomized Trial

BACKGROUND

Behavioral obesity treatment (BOT) is a gold standard approach to weight loss and reduces the risk of cardiovascular disease. However, frequent lapses from the recommended diet stymie weight loss and prevent individuals from actualizing the health benefits of BOT. There is a need for innovative treatment solutions to improve adherence to the prescribed diet in BOT.

OBJECTIVE

The aim of this study is to optimize a smartphone-based just-in-time adaptive intervention (JITAI) that uses daily surveys to assess triggers for dietary lapses and deliver interventions when the risk of lapse is high. A microrandomized trial design will evaluate the efficacy of any interventions (ie, theory-driven or a generic alert to risk) on the proximal outcome of lapses during BOT, compare the effects of theory-driven interventions with generic risk alerts on the proximal outcome of lapse, and examine contextual moderators of interventions.

METHODS

Adults with overweight or obesity and cardiovascular disease risk (n=159) will participate in a 6-month web-based BOT while using the JITAI to prevent dietary lapses. Each time the JITAI detects elevated lapse risk, the participant will be randomized to no intervention, a generic risk alert, or 1 of 4 theory-driven interventions (ie, enhanced education, building self-efficacy, fostering motivation, and improving self-regulation). The primary outcome will be the occurrence of lapse in the 2.5 hours following randomization. Contextual moderators of intervention efficacy will also be explored (eg, location and time of day). The data will inform an optimized JITAI that selects the theory-driven approach most likely to prevent lapses in a given moment.

RESULTS

The recruitment for the microrandomized trial began on April 19, 2021, and is ongoing.

CONCLUSIONS

This study will optimize a JITAI for dietary lapses so that it empirically tailors the provision of evidence-based intervention to the individual and context. The finalized JITAI will be evaluated for efficacy in a future randomized controlled trial of distal health outcomes (eg, weight loss).

TRIAL REGISTRATION

ClinicalTrials.gov NCT04784585; http://clinicaltrials.gov/ct2/show/NCT04784585.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/33568.

The Relationship of Alcohol to ART Adherence Among Black MSM in the U.S.: Is it Any Different Among Black MSM in the South?

Alcohol-using Black MSM (Men who have sex with men) are disproportionately impacted by HIV in the U.S.-particularly in the southern U.S.-despite the availability of antiretroviral therapy (ART). The purpose of this study was to summarize the current evidence on alcohol use and ART adherence among Black MSM in the U.S. and in the South and to identify future research needs. A systematic review was conducted using eight databases to identify relevant peer-reviewed articles published between January 2010 and April 2021. The authors also snowballed remaining studies and hand-searched for additional studies. Including both quantitative and qualitative studies, five published studies examined alcohol and ART adherence among Black MSM in the U.S. The search identified 240 articles, the study team reviewed 114 in full-text and determined that only five met the inclusion criteria. Three of the five included studies identified alcohol use as a barrier to ART adherence. In conclusions, the general lack of literature on HIV disparities among alcohol-using Black MSM in the U.S. (specifically in the South) indicates a critical need for research on this population's unique risks and needs to inform the development of tailored interventions.

Optimizing Adaptive Notifications in Mobile Health Interventions Systems: Reinforcement Learning from a Data-driven Behavioral Simulator

Mobile health (mHealth) intervention systems can employ adaptive strategies to interact with users. Instead of designing such complex strategies manually, reinforcement learning (RL) can be used to adaptively optimize intervention strategies concerning the user’s context. In this paper, we focus on the issue of overwhelming interactions when learning a good adaptive strategy for the user in RL-based mHealth intervention agents. We present a data-driven approach integrating psychological insights and knowledge of historical data. It allows RL agents to optimize the strategy of delivering context-aware notifications from empirical data when counterfactual information (user responses when receiving notifications) is missing. Our approach also considers a constraint on the frequency of notifications, which reduces the interaction burden for users. We evaluated our approach in several simulation scenarios using real large-scale running data. The results indicate that our RL agent can deliver notifications in a manner that realizes a higher behavioral impact than context-blind strategies.

Correlates of engagement within an online HIV prevention intervention for single young men who have sex with men: The myDEx project (Preprint)

Background

Objective

Methods

Results

Conclusions

Trial Registration

Comparing ecological momentary assessment to sensor-based approaches in predicting dietary lapse

Ecological momentary assessment (EMA; brief self-report surveys) of dietary lapse risk factors (e.g., cravings) has shown promise in predicting and preventing dietary lapse (nonadherence to a dietary prescription), which can improve weight loss interventions. Passive sensors also can measure lapse risk factors and may offer advantages over EMA (e.g., objective, automatic, semicontinuous data collection), but currently can measure only a few lapse predictors, a notable limitation. This study preliminarily compared the burden and accuracy of commercially available sensors versus established EMA in lapse prediction. N = 23 adults with overweight/obesity completed a 6-week commercial app-based weight loss program. Participants wore a Fitbit, enabled GPS tracking, completed EMA, and reported on EMA and sensor burden poststudy via a 5-point Likert scale. Sensed risk factors were physical activity and sleep (accelerometer), geolocation (GPS), and time, from which 233 features (measurable characteristics of sensor signals) were extracted. EMA measured 19 risk factors, lapse, and categorized GPS into meaningful geolocations. Two supervised binary classification models (LASSO) were created: the sensor model predicted lapse with 63% sensitivity (true prediction rate of lapse) and 60% specificity (true prediction rate of non-lapse) and EMA model with 59% sensitivity and 72% specificity. EMA model accuracy was higher, but self-reported EMA burden (M = 2.96, SD = 1.02) also was higher (M = 1.50, SD = 0.94). EMA model accuracy was superior, but EMA burden was higher than sensor burden. Findings highlight the promise of sensors in contributing to lapse prediction, and future research may use EMA, sensors, or both depending on prioritization of accuracy versus participant burden.

Examination of the relationship between lapses and weight loss in a smartphone-based just-in time adaptive intervention

We developed a smartphone-based just-in-time adaptive intervention (JITAI), called OnTrack, that provides personalized intervention to prevent dietary lapses (i.e., nonadherence from the behavioral weight loss intervention diet). OnTrack utilizes ecological momentary assessment (EMA; repeated electronic surveys) for self-reporting lapse triggers, predicts lapses using machine learning, and provides brief intervention to prevent lapse. We have established preliminary feasibility, acceptability, and efficacy of OnTrack, but no study has examined our hypothesized mechanism of action: reduced lapse frequency will be associated with greater weight loss while using OnTrack. This secondary analysis investigated the association between lapse frequency and the weekly percentage of weight loss. Post hoc analyses evaluated the moderating effect of OnTrack engagement on this association. Participants (N = 121) with overweight/obesity (MBMI = 34.51; 84.3% female; 69.4% White) used OnTrack with a digital weight loss program for 10 weeks. Engagement with OnTrack (i.e., EMA completed and interventions accessed) was recorded automatically, participants self-reported dietary lapses via EMA, and weighed weekly using Bluetooth scales. Linear mixed models with a random effect of subject and fixed effect of time revealed a nonsignificant association between weekly lapses and the percentage of weight loss. Post hoc analyses revealed a statistically significant moderation effect of OnTrack engagement such that fewer EMA and interventions completed conferred the expected associations between lapses and weight loss. Lapses were not associated with weight loss in this study and one explanation may be the influence of engagement levels on this relationship. Future research should investigate the role of engagement in evaluating JITAIs.

Optimizing Healthcare Through Digital Health and Wellness Solutions to Meet the Needs of Patients With Chronic Disease During the COVID-19 Era

The COVID-19 pandemic exposed and exacerbated longstanding inefficiencies and deficiencies in chronic disease management and treatment in the United States, such as a fragmented healthcare experience and system, narrowly focused services, limited resources beyond office visits, expensive yet low quality care, and poor access to comprehensive prevention and non-pharmacological resources. It is feared that the addition of COVID-19 survivors to the pool of chronic disease patients will burden an already precarious healthcare system struggling to meet the needs of chronic disease patients. Digital health and telemedicine solutions, which exploded during the pandemic, may address many inefficiencies and deficiencies in chronic disease management, such as increasing access to care. However, these solutions are not panaceas as they are replete with several limitations, such as low uptake, poor engagement, and low long-term use. To fully optimize digital health and telemedicine solutions, we argue for the gamification of digital health and telemedicine solutions through a pantheoretical framework-one that uses personalized, contextualized, and behavioral science algorithms, data, evidence, and theories to ground treatments.

The impact of mental and somatic stressors on physical activity and sedentary behaviour in adults with type 2 diabetes mellitus: a diary study

Background

Adopting an active lifestyle is key in the management of type 2 diabetes mellitus (T2DM). Nevertheless, the majority of individuals with T2DM fails to do so. Additionally, individuals with T2DM are likely to experience mental (e.g., stress) and somatic (e.g., pain) stressors. Research investigating the link between these stressors and activity levels within this group is largely lacking. Therefore, current research aimed to investigate how daily fluctuations in mental and somatic stressors predict daily levels of physical activity (PA) and sedentary behaviour among adults with T2DM.

Methods

Individuals with T2DM (N = 54) were instructed to complete a morning diary assessing mental and somatic stressors and to wear an accelerometer for 10 consecutive days. The associations between the mental and somatic stressors and participants’ levels of PA and sedentary behaviour were examined using (generalized) linear mixed effect models.

Results

Valid data were provided by 38 participants. We found no evidence that intra-individual increases in mental and somatic stressors detrimentally affected participants’ activity levels. Similarly, levels of sedentary behaviour nor levels of PA were predicted by inter-individual differences in the mental and somatic stressors.

Identifying behavioral types of dietary lapse from a mobile weight loss program: Preliminary investigation from a secondary data analysis

Success in behavioral weight loss (BWL) programs depends on adherence to the recommended diet to reduce caloric intake. Dietary lapses (i.e., deviations from the BWL diet) occur frequently and can adversely affect weight loss outcomes. Research indicates that lapse behavior is heterogenous; there are many eating behaviors that could constitute a dietary lapse, but they are rarely studied as distinct contributors to weight outcomes. This secondary analysis aims to evaluate six behavioral lapse types during a 10-week mobile BWL program (eating a large portion, eating when not intended, eating an off-plan food, planned lapse, being unaware of caloric content, and endorsing multiple types of lapse). Associations between weekly behavioral lapse type frequency and weekly weight loss were investigated, and predictive contextual characteristics (psychological, behavioral, and environmental triggers for lapse) and individual difference (e.g., age, gender) factors were examined across lapse types. Participants (N = 121) with overweight/obesity (M_BMI = 34.51; 84.3% female; 69.4% White) used a mobile BWL program for 10 weeks, self-weighed weekly using Bluetooth scales, completed daily ecological momentary assessment of lapse behavior and contextual characteristics, and completed a baseline demographics questionnaire. Linear mixed models revealed significant negative associations between unplanned lapses and percent weight loss. Unplanned lapses from eating a large portion, eating when not intended, and having multiple "types" were significantly negatively associated with weekly percent weight loss. A lasso regression showed that behavioral lapse types share many similar stable factors, with other factors being unique to specific lapse types. Results add to the prior literature on lapses and weight loss in BWL and provide preliminary evidence that behavioral lapse types could aid in understanding adherence behavior and developing precision medicine tools to improve dietary adherence.

Reinforcement Learning to Send Reminders at Right Moments in Smartphone Exercise Application: A Feasibility Study

Just-in-time adaptive intervention (JITAI) has gained attention recently and previous studies have indicated that it is an effective strategy in the field of mobile healthcare intervention. Identifying the right moment for the intervention is a crucial component. In this paper the reinforcement learning (RL) technique has been used in a smartphone exercise application to promote physical activity. This RL model determines the 'right' time to deliver a restricted number of notifications adaptively, with respect to users' temporary context information (i.e., time and calendar). A four-week trial study was conducted to examine the feasibility of our model with real target users. JITAI reminders were sent by the RL model in the fourth week of the intervention, while the participants could only access the app's other functionalities during the first 3 weeks. Eleven target users registered for this study, and the data from 7 participants using the application for 4 weeks and receiving the intervening reminders were analyzed. Not only were the reaction behaviors of users after receiving the reminders analyzed from the application data, but the user experience with the reminders was also explored in a questionnaire and exit interviews. The results show that 83.3% reminders sent at adaptive moments were able to elicit user reaction within 50 min, and 66.7% of physical activities in the intervention week were performed within 5 h of the delivery of a reminder. Our findings indicated the usability of the RL model, while the timing of the moments to deliver reminders can be further improved based on lessons learned.

Cognitive and Affective Mediators of Alcohol-Facilitated Intimate Partner Aggression

This multisite study examined whether aggressive cognitions and facial displays of negative affect and anger experienced during provocation mediated the association between alcohol intoxication and intimate partner aggression (IPA). Participants were 249 heavy drinkers (148 men, 101 women) with a recent history of IPA perpetration. Participants were randomly assigned to an Alcohol or No-Alcohol Control beverage condition and completed a shock-based aggression task involving apparent provocation by their intimate partner. During provocation, a hidden camera recorded participants' facial expressions and verbal articulations, which were later coded using the Facial Action Coding System and the Articulated Thoughts in Simulated Situations paradigm. Results indicated that the positive association between alcohol intoxication and partner-directed physical aggression was mediated by participants' aggressive cognitions, but not by negative affect or anger facial expressions. These findings implicate aggressogenic cognitions as a mediating mechanism underlying the association between the acute effects of alcohol and IPA perpetration.

Feasibility and Preliminary Efficacy of a Community-Based Addiction Rehabilitation Electronic System in Substance Use Disorder: Pilot Randomized Controlled Trial

BACKGROUND

Drug use disorder has high potential for relapse and imposes an enormous burden on public health in China. Since the promulgation of the Anti-drug law in 2008, community-based rehabilitation has become the primary approach to treat drug addiction. However, multiple problems occurred in the implementation process, leading to a low detoxification rate in the community. Mobile health (mHealth) serves as a promising tool to improve the effectiveness and efficiency of community-based rehabilitation. Community-based addiction rehabilitation electronic system (CAREs) is an interactive system for drug users and their assigned social workers.

OBJECTIVE

The study aimed to examine the feasibility and preliminary efficacy of CAREs in community-based rehabilitation from the perspective of drug users and social workers in Shanghai, China.

METHODS

In this pilot randomized controlled trial, 40 participants were recruited from the community in Shanghai from January to May 2019. Participants randomized to the intervention group (n=20) received CAREs + community-based rehabilitation, while participants in the control group (n=20) received community-based rehabilitation only for 6 months. CAREs provided education, assessment, and SOS (support) functions for drug users. The assigned social workers provided service and monitored drug use behavior as usual except that the social workers in the intervention group could access the webpage end to obtain drug users' information and fit their routine workflow into CAREs. The primary outcome was the feasibility of CAREs, reflected in the overall proportion and frequency of CAREs features used in both app and webpage end. The secondary outcomes were the effectiveness of CAREs, including the percentage of drug-positive samples, longest period of abstinence, contact times with social workers, and the change of Addiction Severity Index (ASI) from baseline to the 6-month follow-up.

RESULTS

The number of participants logged in to the app ranged from 7 to 20 per week, and CAREs had relatively high levels of continued patient use. Drug users preferred assessment and education features in the app end while their social workers showed high levels of use in urine results record and viewing assessment results on the webpage end. After the 6-month intervention, 3.3% (17/520) of samples in the intervention group and 7.5% (39/520) in the control group were drug-positive (F=4.358, P=.04). No significant differences were noted between the control and intervention groups in terms of longest duration of abstinence, number of contact times and ASI composite scores.

CONCLUSIONS

The study preliminarily demonstrated that with relatively good feasibility and acceptability, CAREs may improve the effectiveness and efficiency of the community-based rehabilitation, which provided instruction for further improvement of the system.

TRIAL REGISTRATION

ClinicalTrials.gov NCT03451344; https://clinicaltrials.gov/ct2/show/NCT03451344.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR2-10.3389/fpsyt.2018.00556.

Making of Mobile SunSmart: Co-designing a Just-in-Time Sun Protection Intervention for Children and Parents

Background

In this study, we describe a participatory design process to develop a technology-based intervention for sun protection for children and their parents. Our methodology embraces and leverages the expert knowledge of the target users, children and their parents, about their sun protection practices to directly influence the design of our mobile just-in-time adaptive intervention (JITAI). The objectives of this paper are to describe our research procedures and summarize primary findings incorporated into developing our JITAI modules.

Methods

We conducted 3 rounds of iterative co-design workshops with design expert KidsTeam UW children (N: 11–12) and subject expert children and their parents from local communities in California (N: 22–48). Iteratively, we thematically coded the qualitative data generated by participants in the co-design sessions to directly inform design specifications.

Results

Three themes emerged: (1) preference for non-linear educational format with less structure,; (2) situations not conducive for prioritizing sun protection; and (3) challenges, barriers, and ambiguity relating to sun protection to protect oneself and one’s family. Based on the design ideas and iterative participant feedback, three categories of modules were developed: personalized and interactive data intake module, narrative-education module with augmented reality experiment, person/real-time tailored JITAI, and assessment modules.

Conclusions

This is one of the first projects that maximally engage children and parents as co-designers to build a technology to improve sun protection with iterative and intentional design principles. Our scalable approach to design a mobile JITAI to improve sun protection will lay the foundation for future public health investigators with similar endeavors.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12529-021-09987-9.

A reinforcement learning based algorithm for personalization of digital, just-in-time, adaptive interventions

Suboptimal health related behaviors and habits; and resulting chronic diseases are responsible for majority of deaths globally. Studies show that providing personalized support to patients yield improved results by preventing and/or timely treatment of these problems. Digital, just-in-time and adaptive interventions are mobile phone-based notifications that are being utilized to support people wherever and whenever necessary in coping with their health problems. In this research, we propose a reinforcement learning-based mechanism to personalize interventions in terms of timing, frequency and preferred type(s). We simultaneously employ two reinforcement learning models, namely intervention-selection and opportune-moment-identification; capturing and exploiting changes in people's long-term and momentary contexts respectively. While the intervention-selection model adapts the intervention delivery with respect to type and frequency, the opportune-moment-identification model tries to find the most opportune moments to deliver interventions throughout a day. We propose two accelerator techniques over the standard reinforcement learning algorithms to boost learning performance. First, we propose a customized version of eligibility traces for rewarding past actions throughout an agent's trajectory. Second, we utilize the transfer learning method to reuse knowledge across multiple learning environments. We validate the proposed approach in a simulated experiment where we simulate four personas differing in their daily activities, preferences on specific intervention types and attitudes towards the targeted behavior. Our experiments show that the proposed approach yields better results compared to the standard reinforcement learning algorithms and successfully capture the simulated variations associated with the personas.

Unplanned vs. planned drinking: Event-level influences of drinking motives and affect

OBJECTIVE

Problematic alcohol involvement in college students remains a public health concern and identifying factors that promote this consequential behavior as it occurs in daily life is critical. Recent work has found that whether a drinking event is unplanned vs. planned has implications for the risk of negative consequences, though less work has identified fine-grained predictors of these two types of drinking occasions.

METHOD

The present study examined drinking motives and positive and negative affect as predictors of unplanned vs. planned drinking in a sample of college students who completed 28 days of ecological momentary assessment (N = 96; 72% White; 52% female). We examined drinking motives reported at two points: (1) in real-time upon initiating drinking and (2) after one day of retrospection (collected at the daily diary report assessing the prior day). Positive and negative affect were both assessed in real-time. Generalized linear mixed-effects models disentangling within- and between-person effects were leveraged.

RESULTS

Drinking "to get high, buzzed, or drunk" - when retrospectively reported for prior-day drinking - exhibited within-person associations with planned drinking, relative to unplanned drinking. This same effect was marginally significant when ascertained in real-time. Individuals with more frequent retrospective endorsement of the motive "to make the day/night more fun" reported more planned drinking. Higher real-time positive affect, but not negative affect, was marginally associated with planned drinking.

CONCLUSIONS

Our findings provide preliminary support that enhancement motives and higher positive affect are related to planned drinking, which may inform the development of momentary interventions.

Digital chemsex support and care: The potential of just-in-time adaptive interventions

Chemsex among gay, bisexual and other men who have sex with men (GBMSM) has received increasing attention as a public health concern in recent years. Chemsex can affect a variety of aspects of the lives of GBMSM and contribute to physical, social and emotional health burden. Starting from a continuum perspective of chemsex, rather than a binary view of problematic vs. non-problematic use, we argue that men engaging in chemsex at different points in their chemsex journey may benefit from tailored and personalized support to cope with the various and evolving challenges and concerns that may be related to their chemsex behavior. To date, interactive digital communication technologies are not much used to provide support and care for GBMSM engaging in chemsex, neither for community-based support and care nor by health services. This suggests potential for missed opportunities, as GBMSM are generally avid users of these technologies for social connections and hookups, including in relation to chemsex. Recent research has provided emerging evidence of the potential effects of so-called just in time adaptive interventions (JITAI) to provide effective support and care for a variety of health issues. JITAI hold much promise for the provision of appropriate, tailored support and care for GBMSM at different points in the chemsex journey. Co-designing JITAI with potential users and other stakeholders (co-design) is key to success. At the Institute for Tropical Medicine, in Antwerp (Belgium), we initiated the Chemified project to develop an innovative digital chemsex support and care tool for GBMSM. This project illustrates how current understanding of chemsex as a journey can be integrated with a JITAI approach and make use of co-design principles to advance the available support and care for GBMSM engaging in chemsex.

Digital Health and Addiction

Digital technologies are rapidly changing how we understand and promote health. A robust and growing line of research has examined how digital health may enhance our understanding and treatment of addiction. This manuscript highlights innovations in the application of digital health approaches to addiction medicine, with a particular emphasis on advances in (1) real-time measurement of drug use events, (2) real-time measurement of the confluence of factors that surround drug use events, and (3) research examining how real-time measurement can inform responsive, in-the-moment interventions to prevent and treat substance use disorder. Although this manuscript focuses on addiction medicine as one exemplar of the striking impact of digital health, science-based digital health offers generalizable solutions to scaling-up unprecedented models of precision healthcare delivery across a broad spectrum of diseases across the globe.

Using Smartphone Survey Data and Machine Learning to Identify Situational and Contextual Risk Factors for HIV Risk Behavior Among Men Who Have Sex with Men Who Are Not on PrEP

"Just-in-time" interventions (JITs) delivered via smartphones have considerable potential for reducing HIV risk behavior by providing pivotal support at key times prior to sex. However, these programs depend on a thorough understanding of when risk behavior is likely to occur to inform the timing of JITs. It is also critical to understand the most important momentary risk factors that may precede HIV risk behavior, so that interventions can be designed to address them. Applying machine learning (ML) methods to ecological momentary assessment data on HIV risk behaviors can help answer both questions. Eighty HIV-negative men who have sex with men (MSM) who were not on PrEP completed a daily diary survey each morning and an experience sampling survey up to six times per day via a smartphone application for 30 days. Random forest models achieved the highest area under the curve (AUC) values for classifying high-risk condomless anal sex (CAS). These models achieved 80% specificity at a sensitivity value of 74%. Unsurprisingly, the most important contextual risk factors that aided in classification were participants' plans and intentions for sex, sexual arousal, and positive affective states. Findings suggest that survey data collected throughout the day can be used to correctly classify about three of every four high-risk CAS events, while incorrectly classifying one of every five non-CAS days as involving high-risk CAS. A unique set of risk factors also often emerge prior to high-risk CAS events that may be useful targets for JITs.

The application of digital health to the assessment and treatment of substance use disorders: The past, current, and future role of the National Drug Abuse Treatment Clinical Trials Network

The application of digital technologies to better assess, understand, and treat substance use disorders (SUDs) is a particularly promising and vibrant area of scientific research. The National Drug Abuse Treatment Clinical Trials Network (CTN), launched in 1999 by the U.S. National Institute on Drug Abuse, has supported a growing line of research that leverages digital technologies to glean new insights into SUDs and provide science-based therapeutic tools to a diverse array of persons with SUDs. This manuscript provides an overview of the breadth and impact of research conducted in the realm of digital health within the CTN. This work has included the CTN's efforts to systematically embed digital screeners for SUDs into general medical settings to impact care models across the nation. This work has also included a pivotal multi-site clinical trial conducted on the CTN platform, whose data led to the very first "prescription digital therapeutic" authorized by the U.S. Food and Drug Administration (FDA) for the treatment of SUDs. Further CTN research includes the study of telehealth to increase capacity for science-based SUD treatment in rural and under-resourced communities. In addition, the CTN has supported an assessment of the feasibility of detecting cocaine-taking behavior via smartwatch sensing. And, the CTN has supported the conduct of clinical trials entirely online (including the recruitment of national and hard-to-reach/under-served participant samples online, with remote intervention delivery and data collection). Further, the CTN is supporting innovative work focused on the use of digital health technologies and data analytics to identify digital biomarkers and understand the clinical trajectories of individuals receiving medications for opioid use disorder (OUD). This manuscript concludes by outlining the many potential future opportunities to leverage the unique national CTN research network to scale-up the science on digital health to examine optimal strategies to increase the reach of science-based SUD service delivery models both within and outside of healthcare.

Refining an algorithm-powered just-in-time adaptive weight control intervention: A randomized controlled trial evaluating model performance and behavioral outcomes

Suboptimal weight losses are partially attributable to lapses from a prescribed diet. We developed an app (OnTrack) that uses ecological momentary assessment to measure dietary lapses and relevant lapse triggers and provides personalized intervention using machine learning. Initially, tension between user burden and complete data was resolved by presenting a subset of lapse trigger questions per ecological momentary assessment survey. However, this produced substantial missing data, which could reduce algorithm performance. We examined the effect of more questions per ecological momentary assessment survey on algorithm performance, app utilization, and behavioral outcomes. Participants with overweight/obesity (n = 121) used a 10-week mobile weight loss program and were randomized to OnTrack-short (i.e. 8 questions/survey) or OnTrack-long (i.e. 17 questions/survey). Additional questions reduced ecological momentary assessment adherence; however, increased data completeness improved algorithm performance. There were no differences in perceived effectiveness, app utilization, or behavioral outcomes. Minimal differences in utilization and perceived effectiveness likely contributed to similar behavioral outcomes across various conditions.

Remotely Delivered Interventions for Obesity Treatment

PURPOSE OF REVIEW

This review synthesizes recent research on remotely delivered interventions for obesity treatment, including summarizing outcomes and challenges to implementing these treatments as well as outlining recommendations for clinical implementation and future research.

RECENT FINDINGS

There are a wide range of technologies used for delivering obesity treatment remotely. Generally, these treatments appear to be acceptable and feasible, though weight loss outcomes are mixed. Engagement in these interventions, particularly in the long term, is a significant challenge. Newer technologies are rapidly developing and enable tailored and adaptable interventions, though research in this area is in its infancy. Further research is required to optimize potential benefits of remotely delivered interventions for obesity.

Disordered eating after bariatric surgery: clinical aspects, impact on outcomes, and intervention strategies

PURPOSE OF REVIEW

Disordered eating behaviors (DEBs) are associated with poor weight outcomes following bariatric surgery. We describe DEBs most relevant to this population, their associations with weight outcomes, and emerging data on interventions for DEBs.

RECENT FINDINGS

Loss of control eating episodes and grazing have been the most well studied DEBs in bariatric samples. Although DEBs often remit after surgery even without targeted intervention, a subgroup of patients have persistent or newly developed DEBs postoperatively. Preoperative DEBs have little effect on weight outcomes, whereas preoperative impulse control-related features commonly associated with DEBs (e.g., inhibitory control) may have stronger predictive value. Postoperatively, DEBs appear to exert robust effects on concurrently measured weight. Postoperative interventions hold promise for optimizing treatment outcomes.

SUMMARY

We recommend the following to improve clinical care and move research forward: a common language for DEB constructs is needed to improve cross-talk among researchers and care providers; diagnostic schemes and assessment tools may require tailoring for the bariatric population; mechanisms underlying improvements in DEBs following surgery should be clarified; ongoing monitoring of DEBs in the postoperative period is warranted; and a stepped-care approach may improve weight outcomes in a cost-effective manner.

Randomized controlled trial of OnTrack, a just-in-time adaptive intervention designed to enhance weight loss

This randomized trial demonstrated qualified support for the ability of a machine learning-powered, smartphone-based just-in-time, adaptive intervention to enhance weight loss over and above a commercial weight loss program.

Potential benefits of integrating ecological momentary assessment data into mHealth care systems

The advancement of wearable/ambulatory technologies has brought a huge change to data collection frameworks in recent decades. Mobile health (mHealth) care platforms, which utilize ambulatory devices to collect naturalistic and often intensively sampled data, produce innovative information of potential clinical relevance. For example, such data can inform clinical study design, recruitment approach, data analysis, and delivery of both "traditional" and novel (e.g., mHealth) interventions. We provide a conceptual overview of how data measured continuously or repeatedly via mobile devices (e.g., smartphone and body sensors) in daily life could be fruitfully used within a mHealth care system. We highlight the potential benefits of integrating ecological momentary assessment (EMA) into mHealth platforms for collecting, processing, and modeling data, and delivering and evaluating novel interventions in everyday life. Although the data obtained from EMA and related approaches may hold great potential benefits for mHealth care system, there are also implementation challenges; we briefly discuss the challenges to integrating EMA into mHealth care system.

Identifying Smoking Environments From Images of Daily Life With Deep Learning

IMPORTANCE

Environments associated with smoking increase a smoker's craving to smoke and may provoke lapses during a quit attempt. Identifying smoking risk environments from images of a smoker's daily life provides a basis for environment-based interventions.

OBJECTIVE

To apply a deep learning approach to the clinically relevant identification of smoking environments among settings that smokers encounter in daily life.

DESIGN, SETTING, AND PARTICIPANTS

In this cross-sectional study, 4902 images of smoking (n = 2457) and nonsmoking (n = 2445) locations were photographed by 169 smokers from Durham, North Carolina, and Pittsburgh, Pennsylvania, areas from 2010 to 2016. These images were used to develop a probabilistic classifier to predict the location type (smoking or nonsmoking location), thus relating objects and settings in daily environments to established smoking patterns. The classifier combines a deep convolutional neural network with an interpretable logistic regression model and was trained and evaluated via nested cross-validation with participant-wise partitions (ie, out-of-sample prediction). To contextualize model performance, images taken by 25 randomly selected participants were also classified by smoking cessation experts. As secondary validation, craving levels reported by participants when viewing unfamiliar environments were compared with the model's predictions. Data analysis was performed from September 2017 to May 2018.

MAIN OUTCOMES AND MEASURES

Classifier performance (accuracy and area under the receiver operating characteristic curve [AUC]), comparison with 4 smoking cessation experts, contribution of objects and settings to smoking environment status (standardized model coefficients), and correlation with participant-reported craving.

RESULTS

Of 169 participants, 106 (62.7%) were from Durham (53 [50.0%] female; mean [SD] age, 41.4 [12.0] years) and 63 (37.3%) were from Pittsburgh (31 [51.7%] female; mean [SD] age, 35.2 [13.8] years). A total of 4902 images were available for analysis, including 3386 from Durham (mean [SD], 31.9 [1.3] images per participant) and 1516 from Pittsburgh (mean [SD], 24.1 [0.5] images per participant). Images were evenly split between the 2 classes, with 2457 smoking images (50.1%) and 2445 nonsmoking images (49.9%). The final model discriminated smoking vs nonsmoking environments with a mean (SD) AUC of 0.840 (0.024) (accuracy [SD], 76.5% [1.6%]). A model trained only with images from Durham participants effectively classified images from Pittsburgh participants (AUC, 0.757; accuracy, 69.2%), and a model trained only with images from Pittsburgh participants effectively classified images from Durham participants (AUC, 0.821; accuracy, 75.0%), suggesting good generalizability between geographic areas. Only 1 expert's performance was a statistically significant improvement compared with the classifier (α = .05). Median self-reported craving was significantly correlated with model-predicted smoking environment status (ρ = 0.894; P = .003).

CONCLUSIONS AND RELEVANCE

In this study, features of daily environments predicted smoking vs nonsmoking status consistently across participants. The findings suggest that a deep learning approach can identify environments associated with smoking, can predict the probability that any image of daily life represents a smoking environment, and can potentially trigger environment-based interventions. This work demonstrates a framework for predicting how daily environments may influence target behaviors or symptoms that may have broad applications in mental and physical health.

Strengthening and Expanding Child Services in Low Resource Communities: The Role of Task-Shifting and Just-in-Time Training

In the United States, the demand for child mental health services is increasing, while the supply is limited by workforce shortages. These shortages are unlikely to be corrected without significant structural changes in how mental health services are provided. One strategy for bridging this gap is task-shifting, defined as a process by which services that are typically delivered by professionals are moved to individuals with less extensive qualifications or training. Although task-shifting can increase the size of the workforce, there are challenges related to training new workers. In this paper, we propose Just-In-Time Training (JITT) as one strategy for improving task-shifting efforts. We define JITT as on-demand training experiences that only include what is necessary, when it is necessary, to promote competent service delivery. We offer a proof of concept from our own work shifting counseling and academic support tasks from school mental health professionals to pre-baccalaureate mentors, citing lessons learned during our iterative process of JITT development. We conclude with a series of key considerations for scaling up the pairing of task-shifting and JITT, including expanding the science of JITT and anticipating how task-shifting and JITT would work within the context of dynamic mental health service systems.

An expandable approach for design and personalization of digital, just-in-time adaptive interventions

Objective

We aim to deliver a framework with 2 main objectives: 1) facilitating the design of theory-driven, adaptive, digital interventions addressing chronic illnesses or health problems and 2) producing personalized intervention delivery strategies to support self-management by optimizing various intervention components tailored to people's individual needs, momentary contexts, and psychosocial variables.

Materials and Methods

We propose a template-based digital intervention design mechanism enabling the configuration of evidence-based, just-in-time, adaptive intervention components. The design mechanism incorporates a rule definition language enabling experts to specify triggering conditions for interventions based on momentary and historical contextual/personal data. The framework continuously monitors and processes personal data space and evaluates intervention-triggering conditions. We benefit from reinforcement learning methods to develop personalized intervention delivery strategies with respect to timing, frequency, and type (content) of interventions. To validate the personalization algorithm, we lay out a simulation testbed with 2 personas, differing in their various simulated real-life conditions.

Results

We evaluate the design mechanism by presenting example intervention definitions based on behavior change taxonomies and clinical guidelines. Furthermore, we provide intervention definitions for a real-world care program targeting diabetes patients. Finally, we validate the personalized delivery mechanism through a set of hypotheses, asserting certain ways of adaptation in the delivery strategy, according to the differences in simulation related to personal preferences, traits, and lifestyle patterns.

Conclusion

While the design mechanism is sufficiently expandable to meet the theoretical and clinical intervention design requirements, the personalization algorithm is capable of adapting intervention delivery strategies for simulated real-life conditions.

OnTrack: development and feasibility of a smartphone app designed to predict and prevent dietary lapses

Given that the overarching goal of weight loss programs is to remain adherent to a dietary prescription, specific moments of nonadherence known as "dietary lapses" can threaten weight control via the excess energy intake they represent and by provoking future lapses. Just-in-time adaptive interventions could be particularly useful in preventing dietary lapses because they use real-time data to generate interventions that are tailored and delivered at a moment computed to be of high risk for a lapse. To this end, we developed a smartphone application (app) called OnTrack that utilizes machine learning to predict dietary lapses and deliver a targeted intervention designed to prevent the lapse from occurring. This study evaluated the feasibility, acceptability, and preliminary effectiveness of OnTrack among weight loss program participants. An open trial was conducted to investigate subjective satisfaction, objective usage, algorithm performance, and changes in lapse frequency and weight loss among individuals (N = 43; 86% female; body mass index = 35.6 kg/m2) attempting to follow a structured online weight management plan for 8 weeks. Participants were adherent with app prompts to submit data, engaged with interventions, and reported high levels of satisfaction. Over the course of the study, participants averaged a 3.13% weight loss and experienced a reduction in unplanned lapses. OnTrack, the first Just-in-time adaptive intervention for dietary lapses was shown to be feasible and acceptable, and OnTrack users experienced weight loss and lapse reduction over the study period. These data provide the basis for further development and evaluation.

Personalized models of physical activity responses to text message micro-interventions: A proof-of-concept application of control systems engineering methods

OBJECTIVES

The conceptual models underlying physical activity interventions have been based largely on differences between more and less active people. Yet physical activity is a dynamic behavior, and such models are not sensitive to factors that regulate behavior at a momentary level or how people respond to individual attempts at intervening. We demonstrate how a control systems engineering approach can be applied to develop personalized models of behavioral responses to an intensive text message-based intervention.

DESIGN & METHOD

To establish proof-of-concept for this approach, 10 adults wore activity monitors for 16 weeks and received five text messages daily at random times. Message content was randomly selected from three types of messages designed to target (1) social-cognitive processes associated with increasing physical activity, (2) social-cognitive processes associated with reducing sedentary behavior, or (3) general facts unrelated to either physical activity or sedentary behavior. A dynamical systems model was estimated for each participant to examine the magnitude and timing of responses to each type of text message.

RESULTS

Models revealed heterogeneous responses to different message types that varied between people and between weekdays and weekends.

CONCLUSIONS

This proof-of-concept demonstration suggests that parameters from this model can be used to develop personalized algorithms for intervention delivery. More generally, these results demonstrate the potential utility of control systems engineering models for optimizing physical activity interventions.

Adaptive Intervention Designs to Promote Behavioral Change in Adults: What Is the Evidence?

PURPOSE OF REVIEW

Adaptive behavioral interventions tailor the type or dose of intervention strategies to individuals over time to improve saliency and intervention efficacy. This review describes the unique characteristics of adaptive intervention designs, summarizes recent diabetes-related prevention studies, which used adaptive designs, and offers recommendations for future research.

RECENT FINDINGS

Eight adaptive intervention studies were reported since 2013 to reduce sedentary behavior or improve weight management in overweight or obese adults. Primarily, feasibility studies were conducted. Preliminary results suggest that just-in-time adaptive interventions can reduce sedentary behavior or increase minutes of physical activity through repeated prompts. A stepped-down weight management intervention did not increase weight loss compared to a fixed intervention. Other adaptive interventions to promote weight management are underway and require further evaluation. Additional research is needed to target a broader range of health-related behaviors, identify optimal decision points and dose for intervention, develop effective engagement strategies, and evaluate outcomes using randomized trials.

Digital Interventions for Mental Disorders: Key Features, Efficacy, and Potential for Artificial Intelligence Applications

Mental disorders are highly prevalent and often remain untreated. Many limitations of conventional face-to-face psychological interventions could potentially be overcome through Internet-based and mobile-based interventions (IMIs). This chapter introduces core features of IMIs, describes areas of application, presents evidence on the efficacy of IMIs as well as potential effect mechanisms, and delineates how Artificial Intelligence combined with IMIs may improve current practices in the prevention and treatment of mental disorders in adults. Meta-analyses of randomized controlled trials clearly show that therapist-guided IMIs can be highly effective for a broad range of mental health problems. Whether the effects of unguided IMIs are also clinically relevant, particularly under routine care conditions, is less clear. First studies on IMIs for the prevention of mental disorders have shown promising results. Despite limitations and challenges, IMIs are increasingly implemented into routine care worldwide. IMIs are also well suited for applications of Artificial Intelligence and Machine Learning, which provides ample opportunities to improve the identification and treatment of mental disorders. Together with methodological innovations, these approaches may also deepen our understanding of how psychological interventions work, and why. Ethical and professional restraints as well as potential contraindications of IMIs, however, should also be considered. In sum, IMIs have a high potential for improving the prevention and treatment of mental health disorders across various indications, settings, and populations. Therefore, implementing IMIs into routine care as both adjunct and alternative to face-to-face treatment is highly desirable. Technological advancements may further enhance the variability and flexibility of IMIs, and thus even further increase their impact in people's lives in the future.

Heart rate variability and emotion regulation among individuals with obesity and loss of control eating

Autonomic nervous system functioning, measured with heart rate variability (HRV), is associated with emotion regulation and likely contributes to binge eating. This study examined the link between HRV and binge eating severity and analyzed changes in HRV as a marker of emotion regulation in individuals with binge eating. Participants (n = 28) with obesity and loss of control eating reported overeating and loss of control episodes and completed an HRV assessment at rest and during a mental stressor. At rest, lower time-domain HRV was linked to more severe loss of control (SDNN B = -0.18, p = 0.03). Frequency-domain HRV was associated with more severe overeating (LFn B = 14.92, p = 0.03; HFn B = -14.81, p = 0.04). Frequency-domain HRV differed between resting and stressed conditions (p's < 0.001). Findings contribute to understanding emotion regulation in binge eating and guide future research and novel intervention development.

Longitudinal Validity and Reliability of Brief Smartphone Self-Monitoring of Diet, Stress, and Physical Activity in a Diverse Sample of Mothers

Background

Multiple strategies can be used when self-monitoring diet, physical activity, and perceived stress, but no gold standards are available. Although self-monitoring is a core element of self-management and behavior change, the success of mHealth behavioral tools depends on their validity and reliability, which lack evidence. African American and Latina mothers in the United States are high-priority populations for apps that can be used for self-monitoring of diet, physical activity, and stress because the body mass index (BMI) of mothers typically increases for several years after childbirth and the risks of obesity and its’ sequelae diseases are elevated among minority populations.

Objective

To examine the intermethod reliability and concurrent validity of smartphone-based self-monitoring via ecological momentary assessments (EMAs) and use of daily diaries for diet, stress, and physical activity compared with brief recall measures, anthropometric biomeasures, and bloodspot biomarkers.

Methods

A purposive sample (n=42) of primarily African American (16/42, 39%) and Latina (18/42, 44%) mothers was assigned Android smartphones for using Ohmage apps to self-monitor diet, perceived stress, and physical activity over 6 months. Participants were assessed at 3- and 6-month follow-ups. Recall measures included brief food frequency screeners, physical activity assessments adapted from the National Health and Nutrition Examination Survey, and the nine-item psychological stress measure. Anthropometric biomeasures included BMI, body fat, waist circumference, and blood pressure. Bloodspot assays for Epstein–Barr virus and C-reactive protein were used as systemic load and stress biomarkers. EMAs and daily diary questions assessed perceived quality and quantity of meals, perceived stress levels, and moderate, vigorous, and light physical activity. Units of analysis were follow-up assessments (n=29 to n=45 depending on the domain) of the participants (n=29 with sufficient data for analyses). Correlations, R² statistics, and multivariate linear regressions were used to assess the strength of associations between variables.

Results

Almost all participants (39/42, 93%) completed the study. Intermethod reliability between smartphone-based EMAs and diary reports and their corresponding recall reports was highest for stress and diet; correlations ranged from .27 to .52 (P<.05). However, it was unexpectedly low for physical activity; no significant associations were observed. Concurrent validity was demonstrated for diet EMAs and diary reports on systolic blood pressure (r=−.32), C-reactive protein level (r=−.34), and moderate and vigorous physical activity recalls (r=.35 to.48), suggesting a covariation between healthy diet and physical activity behaviors. EMAs and diary reports on stress were not associated with Epstein–Barr virus and C-reactive protein level. Diary reports on moderate and vigorous physical activity were negatively associated with BMI and body fat (r=−.35 to −.44, P<.05).

Conclusions

Brief smartphone-based EMA use may be valid and reliable for long-term self-monitoring of diet, stress, and physical activity. Lack of intermethod reliability for physical activity measures is consistent with prior research, warranting more research on the efficacy of smartphone-based self-monitoring of self-management and behavior change support.

Application of Machine Learning to Predict Dietary Lapses During Weight Loss

BACKGROUND

Individuals who adhere to dietary guidelines provided during weight loss interventions tend to be more successful with weight control. Any deviation from dietary guidelines can be referred to as a "lapse." There is a growing body of research showing that lapses are predictable using a variety of physiological, environmental, and psychological indicators. With recent technological advancements, it may be possible to assess these triggers and predict dietary lapses in real time. The current study sought to use machine learning techniques to predict lapses and evaluate the utility of combining both group- and individual-level data to enhance lapse prediction.

METHODS

The current study trained and tested a machine learning algorithm capable of predicting dietary lapses from a behavioral weight loss program among adults with overweight/obesity (n = 12). Participants were asked to follow a weight control diet for 6 weeks and complete ecological momentary assessment (EMA; repeated brief surveys delivered via smartphone) regarding dietary lapses and relevant triggers.

RESULTS

WEKA decision trees were used to predict lapses with an accuracy of 0.72 for the group of participants. However, generalization of the group algorithm to each individual was poor, and as such, group- and individual-level data were combined to improve prediction. The findings suggest that 4 weeks of individual data collection is recommended to attain optimal model performance.

CONCLUSIONS

The predictive algorithm could be utilized to provide in-the-moment interventions to prevent dietary lapses and therefore enhance weight losses. Furthermore, methods in the current study could be translated to other types of health behavior lapses.