Development of a Control-Oriented Model of Social Cognitive Theory for Optimized mHealth Behavioral Interventions

The iterative development and refinement of health psychology theories through formal, dynamical systems modelling: a scoping review and initial expert-derived 'best practice' recommendations

This scoping review aimed to synthesise methodological steps taken by researchers in the development of formal, dynamical systems models of health psychology theories. We searched MEDLINE, PsycINFO, the ACM Digital Library and IEEE Xplore in July 2023. We included studies of any design providing that they reported on the development or refinement of a formal, dynamical systems model unfolding at the within-person level, with no restrictions on population or setting. A narrative synthesis with frequency analyses was conducted. A total of 17 modelling projects reported across 29 studies were included. Formal modelling efforts have largely been concentrated to a small number of interdisciplinary teams in the United States (79.3%). The models aimed to better understand dynamic processes (69.0%) or inform the development of adaptive interventions (31.0%). Models typically aimed to formalise the Social Cognitive Theory (31.0%) or the Self-Regulation Theory (17.2%) and varied in complexity (range: 3-30 model components). Only 3.4% of studies reported involving stakeholders in the modelling process and 10.3% drew on Open Science practices. We conclude by proposing an initial set of expert-derived 'best practice' recommendations. Formal, dynamical systems modelling is poised to help health psychologists develop and refine theories, ultimately leading to more potent interventions.

Modeling engagement with a digital behavior change intervention (HeartSteps II): An exploratory system identification approach

OBJECTIVE

Digital behavior change interventions (DBCIs) are feasibly effective tools for addressing physical activity. However, in-depth understanding of participants' long-term engagement with DBCIs remains sparse. Since the effectiveness of DBCIs to impact behavior change depends, in part, upon participant engagement, there is a need to better understand engagement as a dynamic process in response to an individual's ever-changing biological, psychological, social, and environmental context.

METHODS

The year-long micro-randomized trial (MRT) HeartSteps II provides an unprecedented opportunity to investigate DBCI engagement among ethnically diverse participants. We combined data streams from wearable sensors (Fitbit Versa, i.e., walking behavior), the HeartSteps II app (i.e. page views), and ecological momentary assessments (EMAs, i.e. perceived intrinsic and extrinsic motivation) to build the idiographic models. A system identification approach and a fluid analogy model were used to conduct autoregressive with exogenous input (ARX) analyses that tested hypothesized relationships between these variables inspired by Self-Determination Theory (SDT) with DBCI engagement through time.

RESULTS

Data from 11 HeartSteps II participants was used to test aspects of the hypothesized SDT dynamic model. The average age was 46.33 (SD=7.4) years, and the average steps per day at baseline was 5,507 steps (SD=6,239). The hypothesized 5-input SDT-inspired ARX model for app engagement resulted in a 31.75 % weighted RMSEA (31.50 % on validation and 31.91 % on estimation), indicating that the model predicted app page views almost 32 % better relative to the mean of the data. Among Hispanic/Latino participants, the average overall model fit across inventories of the SDT fluid analogy was 34.22 % (SD=10.53) compared to 22.39 % (SD=6.36) among non-Hispanic/Latino Whites, a difference of 11.83 %. Across individuals, the number of daily notification prompts received by the participant was positively associated with increased app page views. The weekend/weekday indicator and perceived daily busyness were also found to be key predictors of the number of daily application page views.

CONCLUSIONS

This novel approach has significant implications for both personalized and adaptive DBCIs by identifying factors that foster or undermine engagement in an individual's respective context. Once identified, these factors can be tailored to promote engagement and support sustained behavior change over time.

Predicting Goal Attainment in Process-Oriented Behavioral Interventions Using a Data-Driven System Identification Approach

Behavioral interventions (such as those developed to increase physical activity, achieve smoking cessation, or weight loss) can be represented as dynamic process systems incorporating a multitude of factors, ranging from cognitive (internal) to environmental (external) influences. This facilitates the application of system identification and control engineering methods to address questions such as: what drives individuals to improve health behaviors (such as engaging in physical activity)? In this paper, the goal is to efficiently estimate personalized, dynamic models which in turn will lead to control systems that can optimize this behavior. This problem is examined in system identification applied to the Just Walk study that aimed to increase walking behavior in sedentary adults. The paper presents a Discrete Simultaneous Perturbation Stochastic Approximation (DSPSA)-based modeling of the Goal Attainment construct estimated using AutoRegressive with eXogenous inputs (ARX) models. Feature selection of participants and ARX order selection is achieved through the DSPSA algorithm, which efficiently handles computationally expensive calculations. DSPSA can search over large sets of features as well as regressor structures in an informed, principled manner to model behavioral data within reasonable computational time. DSPSA estimation highlights the large individual variability in motivating factors among participants in Just Walk, thus emphasizing the importance of a personalized approach for optimized behavioral interventions.

An Ecological Mobile Momentary Intervention to Support Dynamic Goal Pursuit: Feasibility and Acceptability Study

BACKGROUND

Individuals can experience difficulties pursuing their goals amid multiple competing priorities in their environment. Effective goal dynamics require flexible and generalizable pursuit skills. Supporting successful goal pursuit requires a perpetually adapting intervention responsive to internal states.

OBJECTIVE

The purpose of this study was to (1) develop a flexible intervention that can adapt to an individual's changing short to medium-term goals and be applied to their daily life and (2) examine the feasibility and acceptability of the just-in-time adaptive intervention for goal pursuit.

METHODS

This study involved 3 iterations to test and systematically enhance all aspects of the intervention. During the pilot phase, 73 participants engaged in an ecological momentary assessment (EMA) over 1 month. After week 1, they attended an intervention training session and received just-in-time intervention prompts during the following 3 weeks. The training employed the Capability, Opportunity, Motivation, and Behavior (COM-B) framework for goal setting, along with mental contrasting with implementation intentions (MCII). Subsequent prompts, triggered by variability in goal pursuit, guided the participants to engage in MCII in relation to their current goal. We evaluated feasibility and acceptability, efficacy, and individual change processes by combining intensive (single-case experimental design) and extensive methods.

RESULTS

The results suggest that the digital intervention was feasible and acceptable to participants. Compliance with the intervention was high (n=63, 86%). The participants endorsed high acceptability ratings relating to both the study procedures and the intervention. All participants (N=73, 100%) demonstrated significant improvements in goal pursuit with an average difference of 0.495 units in the outcome (P<.001). The results of the dynamic network modeling suggest that self-monitoring behavior (EMA) and implementing the MCII strategy may aid in goal reprioritization, where goal pursuit itself is a driver of further goal pursuit.

CONCLUSIONS

This pilot study demonstrated the feasibility and acceptability of a just-in-time adaptive intervention among a nonclinical adult sample. This intervention used self-monitoring of behavior, the COM-B framework, and MCII strategies to improve dynamic goal pursuit. It was delivered via an Ecological Momentary Intervention (EMI) procedure. Future research should consider the utility of this approach as an additional intervention element within psychological interventions to improve goal pursuit. Sustaining goal pursuit throughout interventions is central to their effectiveness and warrants further evaluation.

The Digital Therapeutics Real-World Evidence Framework: An Approach for Guiding Evidence-Based Digital Therapeutics Design, Development, Testing, and Monitoring

Digital therapeutics (DTx) are a promising way to provide safe, effective, accessible, sustainable, scalable, and equitable approaches to advance individual and population health. However, developing and deploying DTx is inherently complex in that DTx includes multiple interacting components, such as tools to support activities like medication adherence, health behavior goal-setting or self-monitoring, and algorithms that adapt the provision of these according to individual needs that may change over time. While myriad frameworks exist for different phases of DTx development, no single framework exists to guide evidence production for DTx across its full life cycle, from initial DTx development to long-term use. To fill this gap, we propose the DTx real-world evidence (RWE) framework as a pragmatic, iterative, milestone-driven approach for developing DTx. The DTx RWE framework is derived from the 4-phase development model used for behavioral interventions, but it includes key adaptations that are specific to the unique characteristics of DTx. To ensure the highest level of fidelity to the needs of users, the framework also incorporates real-world data (RWD) across the entire life cycle of DTx development and use. The DTx RWE framework is intended for any group interested in developing and deploying DTx in real-world contexts, including those in industry, health care, public health, and academia. Moreover, entities that fund research that supports the development of DTx and agencies that regulate DTx might find the DTx RWE framework useful as they endeavor to improve how DTxcan advance individual and population health.

Effects of physical health beliefs on college students' physical exercise behavior intention: mediating effects of exercise imagery

OBJECTIVE

This study explores the relationship between physical health beliefs and physical exercise behavior intention of college students and constructs a mediation model through the mediation role of exercise imagery.

METHODS

Using the stratified cluster sampling method, 1356 college students were measured in group psychology by using the Physical Health Beliefs Scale, Exercise Imagery Inventory, and Physical Exercise Behavior Intention Scale. The statistical software Mplus 8.1, SPSS 22.0 and SPSS PROCESS 3.3 were used for statistical processing. The common method deviation test was carried out by Harman single-factor control method. Finally, the bootstrap sampling test method and process plug-in were used to test the significance of intermediary effect.

RESULTS

(1) physical health beliefs have a significant predictive effect on physical exercise behavior intention (β = 0.32, p < 0.001); (2) exercise imagery (β = 0.13, p < 0.001) mediate the relationship between physical health beliefs and physical exercise behavior intention (physical health beliefs → exercise imagery → physical exercise behavior intention (95% Cl: 0.14, 0.32)).

CONCLUSION

physical health beliefs can directly improve the physical exercise behavior intention of college students, which can also affect college students' physical exercise behavior intention indirectly through exercise imagery. The findings suggest that exercise imagery are important variables that mediate the effect of the college students' physical health beliefs on their physical exercise behavior intention.

Advancing Understanding of Just-in-Time States for Supporting Physical Activity (Project JustWalk JITAI): Protocol for a System ID Study of Just-in-Time Adaptive Interventions

BACKGROUND

Just-in-time adaptive interventions (JITAIs) are designed to provide support when individuals are receptive and can respond beneficially to the prompt. The notion of a just-in-time (JIT) state is critical for JITAIs. To date, JIT states have been formulated either in a largely data-driven way or based on theory alone. There is a need for an approach that enables rigorous theory testing and optimization of the JIT state concept.

OBJECTIVE

The purpose of this system ID experiment was to investigate JIT states empirically and enable the empirical optimization of a JITAI intended to increase physical activity (steps/d).

METHODS

We recruited physically inactive English-speaking adults aged ≥25 years who owned smartphones. Participants wore a Fitbit Versa 3 and used the study app for 270 days. The JustWalk JITAI project uses system ID methods to study JIT states. Specifically, provision of support systematically varied across different theoretically plausible operationalizations of JIT states to enable a more rigorous and systematic study of the concept. We experimentally varied 2 intervention components: notifications delivered up to 4 times per day designed to increase a person's steps within the next 3 hours and suggested daily step goals. Notifications to walk were experimentally provided across varied operationalizations of JIT states accounting for need (ie, whether daily step goals were previously met or not), opportunity (ie, whether the next 3 h were a time window during which a person had previously walked), and receptivity (ie, a person previously walked after receiving notifications). Suggested daily step goals varied systematically within a range related to a person's baseline level of steps per day (eg, 4000) until they met clinically meaningful targets (eg, averaging 8000 steps/d as the lower threshold across a cycle). A series of system ID estimation approaches will be used to analyze the data and obtain control-oriented dynamical models to study JIT states. The estimated models from all approaches will be contrasted, with the ultimate goal of guiding rigorous, replicable, empirical formulation and study of JIT states to inform a future JITAI.

RESULTS

As is common in system ID, we conducted a series of simulation studies to formulate the experiment. The results of our simulation studies illustrated the plausibility of this approach for generating informative and unique data for studying JIT states. The study began enrolling participants in June 2022, with a final enrollment of 48 participants. Data collection concluded in April 2023. Upon completion of the analyses, the results of this study are expected to be submitted for publication in the fourth quarter of 2023.

CONCLUSIONS

This study will be the first empirical investigation of JIT states that uses system ID methods to inform the optimization of a scalable JITAI for physical activity.

TRIAL REGISTRATION

ClinicalTrials.gov NCT05273437; https://clinicaltrials.gov/ct2/show/NCT05273437.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/52161.

System Identification and Hybrid Model Predictive Control in Personalized mHealth Interventions for Physical Activity

The application of control systems principles in behavioral medicine includes developing interventions that can be individualized to promote healthy behaviors, such as sustained engagement in adequate levels of physical activity (PA). This paper presents the use of system identification and control engineering methods in the design of behavioral interventions through the novel formalism of a control-optimization trial (COT). The multiple stages of a COT, from experimental design in system identification through controller implementation, are illustrated using participant data from Just Walk, an intervention to promote walking behavior in sedentary adults. ARX models for individual participants are estimated using multiple estimation and validation data combinations, with the model leading to the best performance over a weighted norm being selected. This model serves as the internal model in a hybrid MPC controller formulated with three degree-of-freedom (3DoF) tuning that properly balances the requirements of physical activity interventions. Its performance in a realistic closed-loop setting is evaluated via simulation. These results serve as proof of concept for the COT approach, which is currently being evaluated with human participants in the clinical trial YourMove.

Effectiveness of a nurse-led mHealth app to prevent excessive gestational weight gain among overweight and obese women: A randomized controlled trial

OBJECTIVE

To explore the effectiveness of a nurse-led mobile health (mHealth) intervention to prevent excessive gestational weight gain (GWG) in overweight and obese women.

METHODS

A randomized controlled trial with an experimental study design. Ninety-two pregnant women with body mass index (BMI) ≥25 kg/m² at less than 17 weeks gestation were recruited from two prenatal clinics in northern Taiwan from January to June 2020. The experimental group used the MyHealthyWeight (MHW) app and a wearable activity tracker (WAT), and the controls received standard antenatal treatments with no mHealth-based elements. Two hospital follow-up visits were scheduled at 24-26 weeks in the second trimester and 34-36 weeks in the third trimester. A generalized estimating equation (GEE) was used to examine the trajectories and the effectiveness of mHealth on GWG.

RESULTS

No difference in GWG was found between the intervention and control groups at baseline (p > 0.05). The GWG trajectory in the entire cohort of women with obesity exhibited a quadratic pattern (ß = 1.8, 95% confidence interval [CI] = 1.27-2.32), and intervention participants' weekly GWG was gained significantly lower than their controls in the second trimester (p < 0.05). Throughout the pregnancy, the mHealth intervention group had a significantly lower proportion of individuals who exceeded their GWG in both total (21.6% vs. 32.6%) and weekly weight gain (first trimester = 58.7% vs. 65.2%; second trimester = 45% vs. 67.4%; third trimester = 48.6% vs. 55.1%). In particular, among obese women in the third trimester, those in the intervention group gained less gestational weight than their controls. The adjusted body weight difference was 5.44 kg (p = 0.023), signifying the total GWG difference (3.30 vs. 8.74 kg) between the means of the two groups. The GEE model indicated that obese women who were aged 35 years, had prepregnancy exercise habits, perceived self-efficacy of diet, and more physical activity tended to have low GWG (p < 0.05).

CONCLUSIONS

The nurse-led mHealth-based intervention shows promising results in significantly preventing excessive GWG among high-BMI women. More effectiveness was found among the obese subgroup.

CLINICAL RELEVANCE

The mHealth-based intervention would be successfully implemented by nurses to help high-BMI women maintain their optimal body weight and promote healthy behavioral changes, particularly in diet and physical activity during pregnancy.

A framework for design and usability testing of telerehabilitation system for adults with chronic diseases: A panoramic scoping review

Objective

This scoping review aimed to identify the design and usability testing of a telerehabilitation (TR) system, and its characteristics and functionalities that are best-suited for rehabilitating adults with chronic diseases.

Methods

Searches were conducted in PubMed, EBSCO, Web of Science, and Cochrane library for studies published between January 2017 and December 2022. We followed the Joanna Briggs Institute guidelines and the framework by Arksey and O'Malley. Screening was undertaken by two reviewers, and data extraction was undertaken by the first author. Then, the data were further reviewed and discussed thoroughly with the team members.

Results

A total of 31 results were identified, with the core criteria of developing and testing a telerehabilitation system, including a mobile app for cardiovascular diseases, cancer, diabetes, and chronic respiratory disorders. All developed systems resulted from multidisciplinary teams and employed mixed-methods research. We proposed the "input-process-output" framework that identified phases of both system design and usability testing. Through system design, we reported the use of user-centered design, iterative design, users' needs and characteristics, theory underpinning development, and the expert panel in 64%, 75%, 86%, 82%, and 71% of the studies, respectively. We recorded the application of moderated usability testing, unmoderated testing (1), and unmoderated testing (2) in 74%, 63%, and 15% of the studies, respectively. The identified design and testing activities produced a matured system, a high-fidelity prototype, and a released system in 81.5%, 15%, and 3.5%, respectively.

Conclusion

This review provides a framework for TR system design and testing for a wide range of chronic diseases that require prolonged management through remote monitoring using a mobile app. The identified "input-process-output" framework highlights the inputs, design, development, and improvement as components of the system design. It also identifies the "moderated-unmoderated" model for conducting usability testing. This review illustrates characteristics and functionalities of the TR systems and healthcare professional roles.

A Kalman filter-based Hybrid Model Predictive Control Algorithm for Mixed Logical Dynamical Systems: Application to Optimized Interventions for Physical Activity

Hybrid Model Predictive Control (HMPC) is presented as a decision-making tool for novel behavioral interventions to increase physical activity in sedentary adults, such as Just Walk. A broad-based HMPC formulation for mixed logical dynamical (MLD) systems relevant to problems in behavioral medicine is developed and illustrated on a representative participant model arising from the Just Walk study. The MLD model is developed based on the requirement of granting points for meeting daily step goals and categorical input variables. The algorithm features three degrees-of-freedom tuning for setpoint tracking, measured and unmeasured disturbance rejection that facilitates controller robustness; disturbance anticipation further improves performance for upcoming events such as weekends and weather forecasts. To avoid the corresponding mixed-integer quadratic problem (MIQP) from becoming infeasible, slack variables are introduced in the objective function. Simulation results indicate that the proposed HMPC scheme effectively manages hybrid dynamics, setpoint tracking, disturbance rejection, and the transition between the two phases of the intervention (initiation and maintenance) and is suitable for evaluation in clinical trials.

Using SIR Epidemic Modeling and Control to Teach Process Dynamics and Control to Chemical Engineers

The COVID-19 pandemic has brought about unprecedented opportunities to introduce control systems topics in the undergraduate engineering curriculum. This paper describes two computer modeling assignments based on MATLAB with Simulink developed for CHE 461: Process Dynamics and Control taught at Arizona State University during the fall 2020 semester. A myriad of important concepts, among these dynamic modeling using conservation and accounting principles, linearization, state-space system and transfer function model representations, PID feedback control and Internal Model Control design can be applied to the problem and explained to students in the context of a significant world event representing a unique "process" system, notably the COVID-19 pandemic.

[A decision framework for an adaptive behavioral intervention for physical activity using hybrid model predictive control: illustration with Just Walk]

Physical inactivity is a major contributor to morbidity and mortality worldwide. Many current physical activity behavioral interventions have shown limited success addressing the problem from a long-term perspective that includes maintenance. This paper proposes the design of a decision algorithm for a mobile and wireless health (mHealth) adaptive intervention that is based on control engineering concepts. The design process relies on a behavioral dynamical model based on Social Cognitive Theory (SCT), with a controller formulation based on hybrid model predictive control (HMPC) being used to implement the decision scheme. The discrete and logical features of HMPC coincide naturally with the categorical nature of the intervention components and the logical decisions that are particular to an intervention for physical activity. The intervention incorporates an online controller reconfiguration mode that applies changes in the penalty weights to accomplish the transition between the behavioral initiation and maintenance training stages. Controller performance is illustrated using an ARX model estimated from system identification data of a representative participant for Just Walk, a physical activity intervention designed on the basis of control systems principles.

Enhanced Social Cognitive Theory Dynamic Modeling and Simulation Towards Improving the Estimation of "Just-In-Time" States

Insufficient physical activity (PA) is commonplace in society, in spite of its significant impact on personal health and well-being. Improved interventions are clearly needed. One of the challenges faced in behavioral interventions is a lack of understanding of multi-timescale dynamics. In this paper we rely on a dynamical model of Social Cognitive Theory (SCT) to gain insights regarding a control-oriented experimental design for a behavioral intervention to improve PA. The intervention (Just Walk JITAI) is designed with the aim to better understand and estimate ideal times for intervention and support based on the concept of "just-in-time" states. An innovative input signal design strategy is used to study the just-in-time state dynamics through the use of decision rules based on conditions of need, opportunity and receptivity. Model simulations featuring within-day effects are used to assess input signal effectiveness. Scenarios for adherent and non-adherent participants are presented, with the proposed experimental design showing significant potential for reducing notification burden while providing informative data to support future system identification and control design efforts.

Model Predictive Control Strategies for Optimized mHealth Interventions for Physical Activity

Many individuals fail to engage in sufficient physical activity (PA), despite its well-known health benefits. This paper examines Model Predictive Control (MPC) as a means to deliver optimized, personalized behavioral interventions to improve PA, as reflected by the number of steps walked per day. Using a health behavior fluid analogy model representing Social Cognitive Theory, a series of diverse strategies are evaluated in simulated scenarios that provide insights into the most effective means for implementing MPC in PA behavioral interventions. The interplay of measurement, information, and decision is explored, with the results illustrating MPC's potential to deliver feasible, personalized, and user-friendly behavioral interventions, even under circumstances involving limited measurements. Our analysis demonstrates the effectiveness of sensibly formulated constrained MPC controllers for optimizing PA interventions, which is a preliminary though essential step to experimental evaluation of constrained MPC control strategies under real-life conditions.

Advancing Behavioral Intervention and Theory Development for Mobile Health: The HeartSteps II Protocol

Background: Recent advances in mobile and wearable technologies have led to new forms of interventions, called “Just-in-Time Adaptive Interventions” (JITAI). JITAIs interact with the individual at the most appropriate time and provide the most appropriate support depending on the continuously acquired Intensive Longitudinal Data (ILD) on participant physiology, behavior, and contexts. These advances raise an important question: How do we model these data to better understand and intervene on health behaviors? The HeartSteps II study, described here, is a Micro-Randomized Trial (MRT) intended to advance both intervention development and theory-building enabled by the new generation of mobile and wearable technology. Methods: The study involves a year-long deployment of HeartSteps, a JITAI for physical activity and sedentary behavior, with 96 sedentary, overweight, but otherwise healthy adults. The central purpose is twofold: (1) to support the development of modeling approaches for operationalizing dynamic, mathematically rigorous theories of health behavior; and (2) to serve as a testbed for the development of learning algorithms that JITAIs can use to individualize intervention provision in real time at multiple timescales. Discussion and Conclusions: We outline an innovative modeling paradigm to model and use ILD in real- or near-time to individually tailor JITIAs.

Smartphone-Based Interventions to Reduce Sedentary Behavior and Promote Physical Activity Using Integrated Dynamic Models: Systematic Review

BACKGROUND

Traditional psychological theories are inadequate to fully leverage the potential of smartphones and improve the effectiveness of physical activity (PA) and sedentary behavior (SB) change interventions. Future interventions need to consider dynamic models taken from other disciplines, such as engineering (eg, control systems). The extent to which such dynamic models have been incorporated in the development of interventions for PA and SB remains unclear.

OBJECTIVE

This review aims to quantify the number of studies that have used dynamic models to develop smartphone-based interventions to promote PA and reduce SB, describe their features, and evaluate their effectiveness where possible.

METHODS

Databases including PubMed, PsycINFO, IEEE Xplore, Cochrane, and Scopus were searched from inception to May 15, 2019, using terms related to mobile health, dynamic models, SB, and PA. The included studies involved the following: PA or SB interventions involving human adults; either developed or evaluated integrated psychological theory with dynamic theories; used smartphones for the intervention delivery; the interventions were adaptive or just-in-time adaptive; included randomized controlled trials (RCTs), pilot RCTs, quasi-experimental, and pre-post study designs; and were published from 2000 onward. Outcomes included general characteristics, dynamic models, theory or construct integration, and measured SB and PA behaviors. Data were synthesized narratively. There was limited scope for meta-analysis because of the variability in the study results.

RESULTS

A total of 1087 publications were screened, with 11 publications describing 8 studies included in the review. All studies targeted PA; 4 also included SB. Social cognitive theory was the major psychological theory upon which the studies were based. Behavioral intervention technology, control systems, computational agent model, exploit-explore strategy, behavioral analytic algorithm, and dynamic decision network were the dynamic models used in the included studies. The effectiveness of quasi-experimental studies involved reduced SB (1 study; P=.08), increased light PA (1 study; P=.002), walking steps (2 studies; P=.06 and P<.001), walking time (1 study; P=.02), moderate-to-vigorous PA (2 studies; P=.08 and P=.81), and nonwalking exercise time (1 study; P=.31). RCT studies showed increased walking steps (1 study; P=.003) and walking time (1 study; P=.06). To measure activity, 5 studies used built-in smartphone sensors (ie, accelerometers), 3 of which used the phone's GPS, and 3 studies used wearable activity trackers.

CONCLUSIONS

To our knowledge, this is the first systematic review to report on smartphone-based studies to reduce SB and promote PA with a focus on integrated dynamic models. These findings highlight the scarcity of dynamic model-based smartphone studies to reduce SB or promote PA. The limited number of studies that incorporate these models shows promising findings. Future research is required to assess the effectiveness of dynamic models in promoting PA and reducing SB.

TRIAL REGISTRATION

International Prospective Register of Systematic Reviews (PROSPERO) CRD42020139350; https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=139350.

From Reflection to Action: Combining Machine Learning with Expert Knowledge for Nutrition Goal Recommendations

Self-tracking can help personalize self-management interventions for chronic conditions like type 2 diabetes (T2D), but reflecting on personal data requires motivation and literacy. Machine learning (ML) methods can identify patterns, but a key challenge is making actionable suggestions based on personal health data. We introduce GlucoGoalie, which combines ML with an expert system to translate ML output into personalized nutrition goal suggestions for individuals with T2D. In a controlled experiment, participants with T2D found that goal suggestions were understandable and actionable. A 4-week in-the-wild deployment study showed that receiving goal suggestions augmented participants' self-discovery, choosing goals highlighted the multifaceted nature of personal preferences, and the experience of following goals demonstrated the importance of feedback and context. However, we identified tensions between abstract goals and concrete eating experiences and found static text too ambiguous for complex concepts. We discuss implications for ML-based interventions and the need for systems that offer more interactivity, feedback, and negotiation.

Engineering Person-Specific Behavioral Interventions to Promote Physical Activity

Physical activity is dynamic, complex, and often regulated idiosyncratically. In this article, we review how techniques used in control systems engineering are being applied to refine physical activity theory and interventions. We hypothesize that person-specific adaptive behavioral interventions grounded in system identification and model predictive control will lead to greater physical activity than more generic, conventional intervention approaches.

Personalized HeartSteps: A Reinforcement Learning Algorithm for Optimizing Physical Activity

With the recent proliferation of mobile health technologies, health scientists are increasingly interested in developing just-in-time adaptive interventions (JITAIs), typically delivered via notifications on mobile devices and designed to help users prevent negative health outcomes and to promote the adoption and maintenance of healthy behaviors. A JITAI involves a sequence of decision rules (i.e., treatment policies) that take the user's current context as input and specify whether and what type of intervention should be provided at the moment. In this work, we describe a reinforcement learning (RL) algorithm that continuously learns and improves the treatment policy embedded in the JITAI as data is being collected from the user. This work is motivated by our collaboration on designing an RL algorithm for HeartSteps V2 based on data collected HeartSteps V1. HeartSteps is a physical activity mobile health application. The RL algorithm developed in this work is being used in HeartSteps V2 to decide, five times per day, whether to deliver a context-tailored activity suggestion.