Evaluating Machine Learning-Based Automated Personalized Daily Step Goals Delivered Through a Mobile Phone App: Randomized Controlled Trial

PEAK Mood, Mind, and Marks: a pilot study of an intervention to support university students' mental and cognitive health through physical exercise

Introduction

Regular exercise has the potential to enhance university students' mental and cognitive health. The PEAK Mood, Mind and Marks program (i.e., PEAK) is a neuroscience-informed intervention developed using the Behaviour Change Wheel to support students to exercise three or more times per week to benefit their mental and cognitive health. This pilot study assessed the impact of PEAK on exercise, mental and cognitive health, and implementation outcomes.

Methods

PEAK was delivered to 115 undergraduate university students throughout a 12-week university semester. The primary outcome was weekly exercise frequency. Secondary outcomes were: time spent engaged in moderate-vigorous exercise, sedentary behaviour and perceived mental health and cognitive health. All were measured via online self-report questionnaires. Qualitative interviews with 15 students investigated influences on engagement, the acceptability and appropriateness of PEAK, and its mechanisms of behaviour change. Paired t-tests, Wilcoxon Signed-Rank tests and template analysis were used to analyse quantitative and qualitative data, respectively.

Results

On average, 48.4% of students engaged in the recommended frequency of three or more exercise sessions per week. This proportion decreased towards the end of PEAK. Sedentary behaviour significantly decreased from baseline to end-point, and moderate-vigorous exercise significantly increased among students' who were non-exercisers. Mental wellbeing, stress, loneliness, and sense of belonging to the university significantly improved. There were no significant changes in psychological distress. Concentration, memory, and productivity significantly improved. Sixty-eight percent of students remained engaged in one or more components of PEAK at end-point. Qualitative data indicated students found PEAK to be acceptable and appropriate, and that it improved aspects of their capability, opportunity, and motivation to exercise.

Conclusions

Students are receptive to an exercise-based program to support their mental and cognitive health. Students exercise frequency decreased; however, these figures are likely a conservative estimate of students exercise engagement. Students valued the neuroscience-informed approach to motivational and educational content and that the program's goals aligned with their academic goals. Students identified numerous areas PEAK's content and implementation can be optimised, including use of a single digital delivery platform, more opportunities to connect with peers and to expand the content's cultural inclusivity.

Age Unplugged: A Brief Narrative Review on the Intersection of Digital Tools, Sedentary and Physical Activity Behaviors in Community-Dwelling Older Adults

This brief narrative review assesses how digital technologies-such as wearables, mobile health apps, and various digital tools such as computers, game consoles, tablets, smartphones, and extended reality systems-can influence sedentary and physical activity behaviors among community-dwelling older adults. Each section highlights the central role of these technologies in promoting active aging through increased motivation, engagement and customized experiences. It underlines the critical importance of functionality, usability and adaptability of devices and confirms the effectiveness of digital interventions in increasing physical activity and reducing sedentary behavior. The sustainable impact of these technologies needs to be further investigated, with a focus on adapting digital health strategies to the specific needs of older people. The research advocates an interdisciplinary approach and points out that such collaborations are essential for the development of accessible, effective and ethical solutions. This perspective emphasizes the potential of digital tools to improve the health and well-being of the aging population and recommends their strategic integration into health promotion and policy making.

A scoping review of methodologies for applying artificial intelligence to physical activity interventions

PURPOSE

This scoping review aimed to offer researchers and practitioners an understanding of artificial intelligence (AI) applications in physical activity (PA) interventions; introduce them to prevalent machine learning (ML), deep learning (DL), and reinforcement learning (RL) algorithms; and encourage the adoption of AI methodologies.

METHODS

A scoping review was performed in PubMed, Web of Science, Cochrane Library, and EBSCO focusing on AI applications for promoting PA or predicting related behavioral or health outcomes. AI methodologies were summarized and categorized to identify synergies, patterns, and trends informing future research. Additionally, a concise primer on predominant AI methodologies within the realm of PA was provided to bolster understanding and broader application.

RESULTS

The review included 24 studies that met the predetermined eligibility criteria. AI models were found effective in detecting significant patterns of PA behavior and associations between specific factors and intervention outcomes. Most studies comparing AI models to traditional statistical approaches reported higher prediction accuracy for AI models on test data. Comparisons of different AI models yielded mixed results, likely due to model performance being highly dependent on the dataset and task. An increasing trend of adopting state-of-the-art DL and RL models over standard ML was observed, addressing complex human-machine communication, behavior modification, and decision-making tasks. Six key areas for future AI adoption in PA interventions emerged: personalized PA interventions, real-time monitoring and adaptation, integration of multimodal data sources, evaluation of intervention effectiveness, expanding access to PA interventions, and predicting and preventing injuries.

CONCLUSION

The scoping review highlights the potential of AI methodologies for advancing PA interventions. As the field progresses, staying informed and exploring emerging AI-driven strategies is essential for achieving significant improvements in PA interventions and fostering overall well-being.

Precision Nutrition to Improve Risk Factors of Obesity and Type 2 Diabetes

PURPOSE OF REVIEW

Existing dietary and lifestyle interventions and recommendations, to improve the risk factors of obesity and type 2 diabetes with the target to mitigate this double global epidemic, have produced inconsistent results due to interpersonal variabilities in response to these conventional approaches, and inaccuracies in dietary assessment methods. Precision nutrition, an emerging strategy, tailors an individual's key characteristics such as diet, phenotype, genotype, metabolic biomarkers, and gut microbiome for personalized dietary recommendations to optimize dietary response and health. Precision nutrition is suggested to be an alternative and potentially more effective strategy to improve dietary intake and prevention of obesity and chronic diseases. The purpose of this narrative review is to synthesize the current research and examine the state of the science regarding the effect of precision nutrition in improving the risk factors of obesity and type 2 diabetes.

RECENT FINDINGS

The results of the research review indicate to a large extent significant evidence supporting the effectiveness of precision nutrition in improving the risk factors of obesity and type 2 diabetes. Deeper insights and further rigorous research into the diet-phenotype-genotype and interactions of other components of precision nutrition may enable this innovative approach to be adapted in health care and public health to the special needs of individuals. Precision nutrition provides the strategy to make individualized dietary recommendations by integrating genetic, phenotypic, nutritional, lifestyle, medical, social, and other pertinent characteristics about individuals, as a means to address the challenges of generalized dietary recommendations. The evidence presented in this review shows that precision nutrition markedly improves risk factors of obesity and type 2 diabetes, particularly behavior change.

Behavioral Analytics for Myopic Agents

Many multi-agent systems have a single coordinator providing incentives to a large number of agents. Two challenges faced by the coordinator are a finite budget from which to allocate incentives, and an initial lack of knowledge about the utility function of the agents. Here, we present a behavioral analytics approach for solving the coordinator's problem when the agents make decisions by maximizing utility functions that depend on prior system states, inputs, and other parameters that are initially unknown. Our behavioral analytics framework involves three steps: first, we develop a model that describes the decision-making process of an agent; second, we use data to estimate the model parameters for each agent and predict their future decisions; and third, we use these predictions to optimize a set of incentives that will be provided to each agent. The framework and approaches we propose in this paper can then adapt incentives as new information is collected. Furthermore, we prove that the incentives computed by this approach are asymptotically optimal with respect to a loss function that describes the coordinator's objective. We optimize incentives with a decomposition scheme, where each sub-problem solves the coordinator's problem for a single agent, and the master problem is a pure integer program. We conclude with a simulation study to evaluate the effectiveness of our approach for designing a personalized weight loss program. The results show that our approach maintains efficacy of the program while reducing its costs by up to 60%, while adaptive heuristics provide substantially less savings.

Digital Health Solutions for Weight Loss and Obesity: A Narrative Review

Personal exercise programmes have long been used and prescribed for weight loss and the improvement of quality of life in obese patients. While individualised programmes are usually the preferred option, they can be more costly and challenging to deliver in person. A move to digital programmes with a wider reach has commenced, and demand has increased due to the SARS-CoV-2 pandemic. In this review, we evaluate the current status of digital exercise programme delivery and its evolution over the past decade, with a focus on personalisation. We used specific keywords to search for articles that met our predetermined inclusion and exclusion criteria in order to provide valuable evidence and insights for future research. We identified 55 studies in total in four key areas of focus, from the more recent development of apps and personal digital assistants to web-based programmes and text or phone call interventions. In summary, we observed that apps may be useful for a low-intensity approach and can improve adherence to programmes through self-monitoring, but they are not always developed in an evidence-based manner. Engagement and adherence are important determinants of weight loss and subsequent weight maintenance. Generally, professional support is required to achieve weight loss goals.

Mining Sensor Data to Assess Changes in Physical Activity Behaviors in Health Interventions: Systematic Review

BACKGROUND

Sensors are increasingly used in health interventions to unobtrusively and continuously capture participants' physical activity in free-living conditions. The rich granularity of sensor data offers great potential for analyzing patterns and changes in physical activity behaviors. The use of specialized machine learning and data mining techniques to detect, extract, and analyze these patterns has increased, helping to better understand how participants' physical activity evolves.

OBJECTIVE

The aim of this systematic review was to identify and present the various data mining techniques employed to analyze changes in physical activity behaviors from sensors-derived data in health education and health promotion intervention studies. We addressed two main research questions: (1) What are the current techniques used for mining physical activity sensor data to detect behavior changes in health education or health promotion contexts? (2) What are the challenges and opportunities in mining physical activity sensor data for detecting physical activity behavior changes?

METHODS

The systematic review was performed in May 2021 using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. We queried the Association for Computing Machinery (ACM), IEEE Xplore, ProQuest, Scopus, Web of Science, Education Resources Information Center (ERIC), and Springer literature databases for peer-reviewed references related to wearable machine learning to detect physical activity changes in health education. A total of 4388 references were initially retrieved from the databases. After removing duplicates and screening titles and abstracts, 285 references were subjected to full-text review, resulting in 19 articles included for analysis.

RESULTS

All studies used accelerometers, sometimes in combination with another sensor (37%). Data were collected over a period ranging from 4 days to 1 year (median 10 weeks) from a cohort size ranging between 10 and 11615 (median 74). Data preprocessing was mainly carried out using proprietary software, generally resulting in step counts and time spent in physical activity aggregated predominantly at the daily or minute level. The main features used as input for the data mining models were descriptive statistics of the preprocessed data. The most common data mining methods were classifiers, clusters, and decision-making algorithms, and these focused on personalization (58%) and analysis of physical activity behaviors (42%).

CONCLUSIONS

Mining sensor data offers great opportunities to analyze physical activity behavior changes, build models to better detect and interpret behavior changes, and allow for personalized feedback and support for participants, especially where larger sample sizes and longer recording times are available. Exploring different data aggregation levels can help detect subtle and sustained behavior changes. However, the literature suggests that there is still work remaining to improve the transparency, explicitness, and standardization of the data preprocessing and mining processes to establish best practices and make the detection methods easier to understand, scrutinize, and reproduce.

Mechanisms of an App-Based Physical Activity Intervention and Maintenance in Community-Dwelling Women: Mediation Analyses of a Randomized Controlled Trial

BACKGROUND

Understanding the mechanism of interventions that increase physical activity (PA) is critical to developing robust intervention strategies.

AIMS

This study aims to examine the mediation effects of hypothesized changes in self-efficacy, social support, and barriers on daily changes in accelerometer-measured steps and the duration of moderate to vigorous PA over 3-month intervention and 6-month maintenance periods with a mobile phone-based PA education program.

METHODS

Data were analyzed for a total of 210 physically inactive women who were randomized. The mean (SD) age was 52.4 (11.0) years. The framework of Baron and Kenny and the Sobel test were used to evaluate the proportion of the treatment effect explained by mediation factors.

RESULTS

Postintervention PA changes were mediated by a reduction in self-efficacy and barriers and an increase in social support from friends during the intervention and maintenance periods (P ≤ .05). However, social support from family was significant only during the intervention, but not the maintenance (P = .90). Barriers to PA had the largest mediation effect on the intervention, explaining 13% to 16% of the 3-month intervention effect and 14% to 19% of the 6-month maintenance effect on daily steps and duration of moderate to vigorous PA minutes (P ≤ .05).

CONCLUSIONS

Incorporating strategies for overcoming PA barriers and promoting social support for PA is important for the design of interventions for physically inactive women. However, a reduction in self-efficacy was observed in the intervention group at 3 and 9 months as compared with the control group. This unexpected finding requires further investigation.

Assessing the Pragmatic Nature of mHealth Interventions Promoting Physical Activity: A Systematic Review and Meta-Analysis (Preprint)

BACKGROUND

Mobile health (mHealth) apps can promote physical activity; however, the pragmatic nature (ie, how well research translates into real-world settings) of these studies is unknown. The impact of study design choices, for example, intervention duration, on intervention effect sizes is also understudied.

OBJECTIVE

This review and meta-analysis aims to describe the pragmatic nature of recent mHealth interventions for promoting physical activity and examine the associations between study effect size and pragmatic study design choices.

METHODS

The PubMed, Scopus, Web of Science, and PsycINFO databases were searched until April 2020. Studies were eligible if they incorporated apps as the primary intervention, were conducted in health promotion or preventive care settings, included a device-based physical activity outcome, and used randomized study designs. Studies were assessed using the Reach, Effectiveness, Adoption, Implementation, Maintenance (RE-AIM) and Pragmatic-Explanatory Continuum Indicator Summary-2 (PRECIS-2) frameworks. Study effect sizes were summarized using random effect models, and meta-regression was used to examine treatment effect heterogeneity by study characteristics.

RESULTS

Overall, 3555 participants were included across 22 interventions, with sample sizes ranging from 27 to 833 (mean 161.6, SD 193.9, median 93) participants. The study populations' mean age ranged from 10.6 to 61.5 (mean 39.6, SD 6.5) years, and the proportion of males included across all studies was 42.8% (1521/3555). Additionally, intervention lengths varied from 2 weeks to 6 months (mean 60.9, SD 34.9 days). The primary app- or device-based physical activity outcome differed among interventions: most interventions (17/22, 77%) used activity monitors or fitness trackers, whereas the rest (5/22, 23%) used app-based accelerometry measures. Data reporting across the RE-AIM framework was low (5.64/31, 18%) and varied within specific dimensions (Reach=44%; Effectiveness=52%; Adoption=3%; Implementation=10%; Maintenance=12.4%). PRECIS-2 results indicated that most study designs (14/22, 63%) were equally explanatory and pragmatic, with an overall PRECIS-2 score across all interventions of 2.93/5 (SD 0.54). The most pragmatic dimension was flexibility (adherence), with an average score of 3.73 (SD 0.92), whereas follow-up, organization, and flexibility (delivery) appeared more explanatory with means of 2.18 (SD 0.75), 2.36 (SD 1.07), and 2.41 (SD 0.72), respectively. An overall positive treatment effect was observed (Cohen d=0.29, 95% CI 0.13-0.46). Meta-regression analyses revealed that more pragmatic studies (-0.81, 95% CI -1.36 to -0.25) were associated with smaller increases in physical activity. Treatment effect sizes were homogenous across study duration, participants' age and gender, and RE-AIM scores.

CONCLUSIONS

App-based mHealth physical activity studies continue to underreport several key study characteristics and have limited pragmatic use and generalizability. In addition, more pragmatic interventions observe smaller treatment effects, whereas study duration appears to be unrelated to the effect size. Future app-based studies should more comprehensively report real-world applicability, and more pragmatic approaches are needed for maximal population health impacts.

TRIAL REGISTRATION

PROSPERO CRD42020169102; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=169102.

Adding Mobile Elements to Online Physical Activity Interventions Targeted at Adults Aged 50 Years and Older: Protocol for a Systematic Design

BACKGROUND

Physical activity (PA) can increase mental and physical health in adults aged 50 years and older. However, it has been shown that PA guidelines are often not met within this population. Therefore, our research group developed 2 computer-tailored intervention programs in the last decade to stimulate PA: Active Plus and I Move. Although these programs were proven effective, positive effects diminished over time and attrition rates were relatively high. To respond to this, we will integrate 3 interactive mobile elements into the existing programs: activity tracker, ecological momentary intervention program, and virtual coach app.

OBJECTIVE

The goal of the research is to define systematic and evidence-based steps for extending our online computer-based PA intervention programs with 3 interactive mobile elements.

METHODS

Components often included in other (eHealth) design models were identified as key components and served as a base for the definition of systematic steps: exploration of context, involvement of the target population, prototype and intervention testing, and implementation. Based on these key components, 10 systematic steps were defined. The initial step is a literature search, with the results serving as a base for development of the low-fidelity prototypes in step 2. The pilot phase comprises the 3rd to 6th steps and includes semistructured interviews, pilot tests, and adaptations of the prototypes with intensive involvement of the target population of adults aged 50 years and older, where particular attention will be paid to lower educated persons. The 7th step is an effect evaluation in the form of a randomized controlled trial. During the 8th step, the most effective intervention programs will be selected and reinforced. These reinforced intervention programs will be used during the design of an implementation plan in the 9th step and the subsequent field study in the 10th step.

RESULTS

The project will be executed from December 2019 to December 2023. During this period, the systematic approach presented will be practically executed according to the methodological procedures described.

CONCLUSIONS

Based on the 4 identified key components, we were able to design an evidence-based systematic design approach for separately adding 3 mobile elements to our existing online PA intervention programs. The 10 steps are presented as a useful approach to guide future eHealth design studies.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/31677.

SNapp, a Tailored Smartphone App Intervention to Promote Walking in Adults of Low Socioeconomic Position: Development and Qualitative Pilot Study (Preprint)

BACKGROUND

Adults of low socioeconomic position (SEP) are generally less physically active than those who are more socioeconomically advantaged, which increases their cardiovascular disease incidence risk. Moreover, individuals of low SEP are often less easily reached with physical activity (PA) interventions than individuals of higher SEP. Smartphone apps have been presented as a promising platform for delivering PA interventions to difficult-to-reach individuals of low SEP. Although PA apps are widely available, they are rarely based on health behavior theories and most predominantly offer generic PA advice. Consequently, it is unlikely that available apps are the most effective PA intervention tools.

OBJECTIVE

To respond to these areas for improvement, we developed SNapp, an app-based intervention encouraging adults of low SEP to increase PA by providing tailored coaching messages targeting walking behavior. This study aimed to describe SNapp's stepwise development and pilot evaluation process.

METHODS

We applied a stepwise approach: analyzing the health problem, developing a program framework, developing tailoring assessments, writing tailored messages, automating the tailoring process, and implementing and evaluating the program in a qualitative pilot study (11 participants).

RESULTS

SNapp consisted of several elements. First, an app was developed to collect step count and geolocation data using smartphone sensor functionalities. In addition, a survey measure was created to assess users' behavior change technique (BCT) preferences. These 3 data types were used to tailor SNapp's coaching messages to stimulate walking. This allows SNapp to offer feedback on performance levels, contextually tailored prompts when users are near green spaces, and coaching content that aligns with individual BCT preferences. Finally, a server-based Python program that interacts with databases containing user data and tailored messages was built using Microsoft Azure to select and automatically send messages to users through Telegram messenger. Pilot study findings indicated that SNapp was rated positively, with participants reporting that its design, technical functioning, and message content were acceptable. Participants suggested additional functionalities that are worth considering for future updates.

CONCLUSIONS

SNapp is an app-based intervention that aims to promote walking in adults of low SEP by offering tailored coaching messages. Its development is theory based, and it is among the first to incorporate contextualized feedback and content tailored to individual BCT preferences. The effectiveness of SNapp will be evaluated in a 12-month real-life parallel cluster-randomized controlled trial.

Users' Perspective on the AI-Based Smartphone PROTEIN App for Personalized Nutrition and Healthy Living: A Modified Technology Acceptance Model (mTAM) Approach

The ubiquitous nature of smartphone ownership, its broad application and usage, along with its interactive delivery of timely feedback are appealing for health-related behavior change interventions via mobile apps. However, users' perspectives about such apps are vital in better bridging the gap between their design intention and effective practical usage. In this vein, a modified technology acceptance model (mTAM) is proposed here, to explain the relationship between users' perspectives when using an AI-based smartphone app for personalized nutrition and healthy living, namely, PROTEIN, and the mTAM constructs toward behavior change in their nutrition and physical activity habits. In particular, online survey data from 85 users of the PROTEIN app within a period of 2 months were subjected to confirmatory factor analysis (CFA) and regression analysis (RA) to reveal the relationship of the mTAM constructs, i.e., perceived usefulness (PU), perceived ease of use (PEoU), perceived novelty (PN), perceived personalization (PP), usage attitude (UA), and usage intention (UI) with the users' behavior change (BC), as expressed via the acceptance/rejection of six related hypotheses (H1-H6), respectively. The resulted CFA-related parameters, i.e., factor loading (FL) with the related p-value, average variance extracted (AVE), and composite reliability (CR), along with the RA results, have shown that all hypotheses H1-H6 can be accepted (p < 0.001). In particular, it was found that, in all cases, FL > 0.5, CR > 0.7, AVE > 0.5, indicating that the items/constructs within the mTAM framework have good convergent validity. Moreover, the adjusted coefficient of determination (R 2) was found within the range of 0.224-0.732, justifying the positive effect of PU, PEoU, PN, and PP on the UA, that in turn positively affects the UI, leading to the BC. Additionally, using a hierarchical RA, a significant change in the prediction of BC from UA when the UI is used as a mediating variable was identified. The explored mTAM framework provides the means for explaining the role of each construct in the functionality of the PROTEIN app as a supportive tool for the users to improve their healthy living by adopting behavior change in their dietary and physical activity habits. The findings herein offer insights and references for formulating new strategies and policies to improve the collaboration among app designers, developers, behavior scientists, nutritionists, physical activity/exercise physiology experts, and marketing experts for app design/development toward behavior change.

A Systematic and Comparative Review of Behavior Change Strategies in Stress Management Apps: Opportunities for Improvement

Stress is one of the significant triggers of several physiological and psychological illnesses. Mobile health apps have been used to deliver various stress management interventions and coping strategies over the years. However, little work exists on persuasive strategies employed in stress management apps to promote behavior change. To address this gap, we review 150 stress management apps on both Google Play and Apple's App Store in three stages. First, we deconstruct and compare the persuasive/behavior change strategies operationalized in the apps using the Persuasive Systems Design (PSD) framework and Cialdini's Principles of Persuasion. Our results show that the most frequently employed strategies are personalization, followed by self-monitoring, and trustworthiness, while social support strategies such as competition, cooperation and social comparison are the least employed. Second, we compare our findings within the stress management domain with those from other mental health domains to uncover further insights. Finally, we reflect on our findings and offer eight design recommendations to improve the effectiveness of stress management apps and foster future research.

Status of the research in fitness apps: A bibliometric analysis

•

A bibliometric analysis of the fitness apps research field to gain insight into the state of the art.

•

Scopus and Web of Science were used to collect the data (481 records).

•

Statistical analysis and science mapping were used to analyze the data.

•

Provides basic data, research classifications and future research directions in the area.

Fitness applications have undergone considerable development in the last few years and becoming popular and significant in both academic and practical areas. However, contributions to the systematic mapping of this field continue to be lacking. This paper constitutes the first bibliometric study in this field to better understand the current state of research. We examined 481 records from databases Scopus and Web of Science (Core Collection) using several bibliometric analysis methods. All the records on this emerging topic were published between 2011 and 2019. We processed these records using statistical analysis and science mapping. The bibliometric analysis included the year of publication, journal name, citation, author, country, and particularly, research methodology. Additionally, we used the VOSViewer software to perform bibliometric mapping of co-authorship, co-citation of authors, and co-occurrence of keywords. This field of study, it was found, is currently in its precursor stage, contributing primarily to the fields of medicine, computer science, and health sciences. The United States appeared to have made the largest contribution to this field. However, author productivity, number of citations, and number of core journals all indicated a high degree of fragmentation of research in this filed. Remarkably, scientific research in this area is expected to progress tremendously over time. Overall, this study provides basic data and research classifications for the initial phase of research and research direction for future research in this area.

REinforcement learning to improve non-adherence for diabetes treatments by Optimising Response and Customising Engagement (REINFORCE): study protocol of a pragmatic randomised trial

INTRODUCTION

Achieving optimal diabetes control requires several daily self-management behaviours, especially adherence to medication. Evidence supports the use of text messages to support adherence, but there remains much opportunity to improve their effectiveness. One key limitation is that message content has been generic. By contrast, reinforcement learning is a machine learning method that can be used to identify individuals' patterns of responsiveness by observing their response to cues and then optimising them accordingly. Despite its demonstrated benefits outside of healthcare, its application to tailoring communication for patients has received limited attention. The objective of this trial is to test the impact of a reinforcement learning-based text messaging programme on adherence to medication for patients with type 2 diabetes.

METHODS AND ANALYSIS

In the REinforcement learning to Improve Non-adherence For diabetes treatments by Optimising Response and Customising Engagement (REINFORCE) trial, we are randomising 60 patients with suboptimal diabetes control treated with oral diabetes medications to receive a reinforcement learning intervention or control. Subjects in both arms will receive electronic pill bottles to use, and those in the intervention arm will receive up to daily text messages. The messages will be individually adapted using a reinforcement learning prediction algorithm based on daily adherence measurements from the pill bottles. The trial's primary outcome is average adherence to medication over the 6-month follow-up period. Secondary outcomes include diabetes control, measured by glycated haemoglobin A1c, and self-reported adherence. In sum, the REINFORCE trial will evaluate the effect of personalising the framing of text messages for patients to support medication adherence and provide insight into how this could be adapted at scale to improve other self-management interventions.

ETHICS AND DISSEMINATION

This study was approved by the Mass General Brigham Institutional Review Board (IRB) (USA). Findings will be disseminated through peer-reviewed journals, clinicaltrials.gov reporting and conferences.

TRIAL REGISTRATION NUMBER

Clinicaltrials.gov (NCT04473326).

Conversational Physical Activity Coaches for Spanish and English Speaking Women: A User Design Study

Introduction: Digital technologies, including text messaging and mobile phone apps, can be leveraged to increase people's physical activity and manage health. Chatbots, powered by artificial intelligence, can automatically interact with individuals through natural conversation. They may be more engaging than one-way messaging interventions. To our knowledge, physical activity chatbots have not been developed with low-income participants, nor in Spanish-the second most dominant language in the U.S. We recommend best practices for physical activity chatbots in English and Spanish for low-income women. Methods: We designed a prototype physical activity text-message based conversational agent based on various psychotherapeutic techniques. We recruited participants through SNAP-Ed (Supplemental Nutrition Assistance Program Education) in California (Alameda County) and Tennessee (Shelby County). We conducted qualitative interviews with participants during testing of our prototype chatbot, held a Wizard of Oz study, and facilitated a co-design workshop in Spanish with a subset of our participants. Results: We included 10 Spanish- and 8 English-speaking women between 27 and 41 years old. The majority was Hispanic/Latina (n = 14), 2 were White and 2 were Black/African American. More than half were monolingual Spanish speakers, and the majority was born outside the US (>50% in Mexico). Most participants were unfamiliar with chatbots and were initially skeptical. After testing our prototype, most users felt positively about health chatbots. They desired a personalized chatbot that addresses their concerns about privacy, and stressed the need for a comprehensive system to also aid with nutrition, health information, stress, and involve family members. Differences between English and monolingual Spanish speakers were found mostly in exercise app use, digital literacy, and the wish for family inclusion. Conclusion: Low-income Spanish- and English-speaking women are interested in using chatbots to improve their physical activity and other health related aspects. Researchers developing health chatbots for this population should focus on issues of digital literacy, app familiarity, linguistic and cultural issues, privacy concerns, and personalization. Designing and testing this intervention for and with this group using co-creation techniques and involving community partners will increase the probability that it will ultimately be effective.

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.

Systematic review of context-aware digital behavior change interventions to improve health

Health risk behaviors are leading contributors to morbidity, premature mortality associated with chronic diseases, and escalating health costs. However, traditional interventions to change health behaviors often have modest effects, and limited applicability and scale. To better support health improvement goals across the care continuum, new approaches incorporating various smart technologies are being utilized to create more individualized digital behavior change interventions (DBCIs). The purpose of this study is to identify context-aware DBCIs that provide individualized interventions to improve health. A systematic review of published literature (2013-2020) was conducted from multiple databases and manual searches. All included DBCIs were context-aware, automated digital health technologies, whereby user input, activity, or location influenced the intervention. Included studies addressed explicit health behaviors and reported data of behavior change outcomes. Data extracted from studies included study design, type of intervention, including its functions and technologies used, behavior change techniques, and target health behavior and outcomes data. Thirty-three articles were included, comprising mobile health (mHealth) applications, Internet of Things wearables/sensors, and internet-based web applications. The most frequently adopted behavior change techniques were in the groupings of feedback and monitoring, shaping knowledge, associations, and goals and planning. Technologies used to apply these in a context-aware, automated fashion included analytic and artificial intelligence (e.g., machine learning and symbolic reasoning) methods requiring various degrees of access to data. Studies demonstrated improvements in physical activity, dietary behaviors, medication adherence, and sun protection practices. Context-aware DBCIs effectively supported behavior change to improve users' health behaviors.

Personalized mobile technologies for lifestyle behavior change: A systematic review, meta-analysis, and meta-regression

Given that the one-size-fits-all approach to mobile health interventions have limited effects, a personalized approach might be necessary to promote healthy behaviors and prevent chronic conditions. Our systematic review aims to evaluate the effectiveness of personalized mobile interventions on lifestyle behaviors (i.e., physical activity, diet, smoking and alcohol consumption), and identify the effective key features of such interventions. We included any experimental trials that tested a personalized mobile app or fitness tracker and reported any lifestyle behavior measures. We conducted a narrative synthesis for all studies, and a meta-analysis of randomized controlled trials. Thirty-nine articles describing 31 interventions were included (n = 77,243, 64% women). All interventions personalized content and rarely personalized other features. Source of data included system-captured (12 interventions), user-reported (11 interventions) or both (8 interventions). The meta-analysis showed a moderate positive effect on lifestyle behavior outcomes (standardized difference in means [SDM] 0.663, 95% CI 0.228 to 1.10). A meta-regression model including source of data found that interventions that used system-captured data for personalization were associated with higher effectiveness than those that used user-reported data (SDM 1.48, 95% CI 0.76 to 2.19). In summary, the field is in its infancy, with preliminary evidence of the potential efficacy of personalization in improving lifestyle behaviors. Source of data for personalization might be important in determining intervention effectiveness. To fully exploit the potential of personalization, future high-quality studies should investigate the integration of multiple data from different sources and include personalized features other than content.

The potential of artificial intelligence in enhancing adult weight loss: a scoping review

OBJECTIVE

To present an overview of how artificial intelligence (AI) could be used to regulate eating and dietary behaviours, exercise behaviours and weight loss.

DESIGN

A scoping review of global literature published from inception to 15 December 2020 was conducted according to Arksey and O'Malley's five-step framework. Eight databases (CINAHL, Cochrane-Central, Embase, IEEE Xplore, PsycINFO, PubMed, Scopus and Web of Science) were searched. Included studies were independently screened for eligibility by two reviewers with good interrater reliability (k = 0·96).

RESULTS

Sixty-six out of 5573 potential studies were included, representing more than 2031 participants. Three tenets of self-regulation were identified - self-monitoring (n 66, 100 %), optimisation of goal setting (n 10, 15·2 %) and self-control (n 10, 15·2 %). Articles were also categorised into three AI applications, namely machine perception (n 50), predictive analytics only (n 6) and real-time analytics with personalised micro-interventions (n 10). Machine perception focused on recognising food items, eating behaviours, physical activities and estimating energy balance. Predictive analytics focused on predicting weight loss, intervention adherence, dietary lapses and emotional eating. Studies on the last theme focused on evaluating AI-assisted weight management interventions that instantaneously collected behavioural data, optimised prediction models for behavioural lapse events and enhance behavioural self-control through adaptive and personalised nudges/prompts. Only six studies reported average weight losses (2·4-4·7 %) of which two were statistically significant.

CONCLUSION

The use of AI for weight loss is still undeveloped. Based on the current study findings, we proposed a framework on the applicability of AI for weight loss but cautioned its contingency upon engagement and contextualisation.

Differences in objectively measured daily physical activity patterns related to depressive symptoms in community dwelling women - mPED trial

Physical activity (PA) is an effective depression treatment. However, knowledge on how variation in day-to-day PA relates to depression in women is lacking. The purposes of this study were to 1) compare overall objectively measured baseline daily steps and duration of moderate to vigorous PA (MVPA) and 2) examine differences in steps and MVPA on days of the week between women aged 25–65 years, who were physically inactive, with high and low depressive symptoms, enrolled in a run-in period of the mobile phone based physical activity education (mPED) trial. The Center for Epidemiological Studies Depression Scale was used to categorize low/high depressive symptom groups. We used linear mixed-effects models to examine the associations between steps and MVPA and depression-status overall and by day of the week, adjusting for selected demographic variables and their interactions with day of the week. 274 women were included in the final analysis, of which 58 had high depressive symptoms. Overall physical activity levels did not differ. However, day of the week modified the associations of depression with MVPA (p = 0.015) and daily steps (p = 0.08). Women with high depression were characterized by reduced activity at the end of the week (Posthoc: Friday: 791 fewer steps, 95% CI: 73–1509, p = 0.03; 8.8 lower MVPA, 95% CI: 2.16–15.5, p = 0.0098) compared to women with low depression, who showed increased activity. Day of the week might be an important target for personalization of physical activity interventions. Future work should evaluate potential causes of daily activity alterations in depression in women.

The Design and Development of a Personalized Leisure Time Physical Activity Application Based on Behavior Change Theories, End-User Perceptions, and Principles From Empirical Data Mining

Introduction: Many adults do not reach the recommended physical activity (PA) guidelines, which can lead to serious health problems. A promising method to increase PA is the use of smartphone PA applications. However, despite the development and evaluation of multiple PA apps, it remains unclear how to develop and design engaging and effective PA apps. Furthermore, little is known on ways to harness the potential of artificial intelligence for developing personalized apps. In this paper, we describe the design and development of the Playful data-driven Active Urban Living (PAUL): a personalized PA application.

Methods: The two-phased development process of the PAUL apps rests on principles from the behavior change model; the Integrate, Design, Assess, and Share (IDEAS) framework; and the behavioral intervention technology (BIT) model. During the first phase, we explored whether location-specific information on performing PA in the built environment is an enhancement to a PA app. During the second phase, the other modules of the app were developed. To this end, we first build the theoretical foundation for the PAUL intervention by performing a literature study. Next, a focus group study was performed to translate the theoretical foundations and the needs and wishes in a set of user requirements. Since the participants indicated the need for reminders at a for-them-relevant moment, we developed a self-learning module for the timing of the reminders. To initialize this module, a data-mining study was performed with historical running data to determine good situations for running.

Results: The results of these studies informed the design of a personalized mobile health (mHealth) application for running, walking, and performing strength exercises. The app is implemented as a set of modules based on the persuasive strategies “monitoring of behavior,” “feedback,” “goal setting,” “reminders,” “rewards,” and “providing instruction.” An architecture was set up consisting of a smartphone app for the user, a back-end server for storage and adaptivity, and a research portal to provide access to the research team.

Conclusions: The interdisciplinary research encompassing psychology, human movement sciences, computer science, and artificial intelligence has led to a theoretically and empirically driven leisure time PA application. In the current phase, the feasibility of the PAUL app is being assessed.

Unraveling Mobile Health Exercise Interventions for Adults: Scoping Review on the Implementations and Designs of Persuasive Strategies

Background

It is unclear why some physical activity (PA) mobile health (mHealth) interventions successfully promote PA whereas others do not. One possible explanation is the variety in PA mHealth interventions—not only do interventions differ in the selection of persuasive strategies but also the design and implementation of persuasive strategies can vary. However, limited studies have examined the different designs and technical implementations of strategies or explored if they indeed influenced the effectiveness of the intervention.

Objective

This scoping review sets out to explore the different technical implementations and design characteristics of common and likely most effective persuasive strategies, namely, goal setting, monitoring, reminders, rewards, sharing, and social comparison. Furthermore, this review aims to explore whether previous mHealth studies examined the influence of the different design characteristics and technical operationalizations of common persuasive strategies on the effectiveness of the intervention to persuade the user to engage in PA.

Methods

An unsystematic snowball and gray literature search was performed to identify the literature that evaluated the persuasive strategies in experimental trials (eg, randomized controlled trial, pre-post test). Studies were included if they targeted adults, if they were (partly) delivered by a mobile system, if they reported PA outcomes, if they used an experimental trial, and when they specifically compared the effect of different designs or implementations of persuasive strategies. The study methods, implementations, and designs of persuasive strategies, and the study results were systematically extracted from the literature by the reviewers.

Results

A total of 29 experimental trials were identified. We found a heterogeneity in how the strategies are being implemented and designed. Moreover, the findings indicated that the implementation and design of the strategy has an influence on the effectiveness of the PA intervention. For instance, the effectiveness of rewarding was shown to vary between types of rewards; rewarding goal achievement seems to be more effective than rewarding each step taken. Furthermore, studies comparing different ways of goal setting suggested that assigning a goal to users might appear to be more effective than letting the user set their own goal, similar to using adaptively tailored goals as opposed to static generic goals. This study further demonstrates that only a few studies have examined the influence of different technical implementations on PA behavior.

Conclusions

The different implementations and designs of persuasive strategies in mHealth interventions should be critically considered when developing such interventions and before drawing conclusions on the effectiveness of the strategy as a whole. Future efforts are needed to examine which implementations and designs are most effective to improve the translation of theory-based persuasive strategies into practical delivery forms.

A systematic review of tailored mHealth interventions for physical activity promotion among adults

The purpose of this systematic review was to critically examine the effectiveness of tailored mHealth interventions for promoting physical activity (PA) in adult populations. Cochrane Library Central Register of Controlled Trials, Medline, SportDiscus, PubMed, PsycINFO, and ProQuest databases were searched systematically in June 2019. Studies were eligible if they were experimentally designed studies, included adult populations (18+ years), and consisted of a tailored intervention that was delivered via a mobile device (i.e., cell phone, tablet). The primary outcome was change in PA. Risk of bias was assessed using the Cochrane Risk of Bias 2 tool. Sixteen articles were reviewed. Ten studies reported significant positive outcomes for the intervention groups compared to the controls. Three studies reported significant improvements in PA for the tailored intervention arms compared to the non-tailored treatment arms. Four of six studies that reported no between group differences used SMS to deliver tailored materials. Differences on tailoring dimension, PA outcomes, and measurement tools were not identified between studies. Tailored mHealth interventions appear to be promising for promoting PA among adults. Most interventions used multiple intervention components. Additional research is needed to identify best practices and to make programs scalable.

Examining the Impact of a Mobile Health Behavior Change Intervention with a brief in-person component for Cancer Survivors with Overweight/Obesity: Randomized Controlled Trial (Preprint)

Background

Cancer survivorship in Ireland is increasing in both frequency and longevity. However, a significant proportion of cancer survivors do not reach the recommended physical activity levels and have overweight. This has implications for both physical and psychological health, including an increased risk of subsequent and secondary cancers. Mobile health (mHealth) interventions demonstrate potential for positive health behavior change, but there is little evidence for the efficacy of mobile technology in improving health outcomes in cancer survivors with overweight or obesity.

Objective

This study aims to investigate whether a personalized mHealth behavior change intervention improves physical and psychological health outcomes in cancer survivors with overweight or obesity.

Methods

A sample of 123 cancer survivors (BMI≥25 kg/m²) was randomly assigned to the standard care control (n=61) or intervention (n=62) condition. Group allocation was unblinded. The intervention group attended a 4-hour tailored lifestyle education and information session with physiotherapists, a dietician, and a clinical psychologist to support self-management of health behavior. Over the following 12 weeks, participants engaged in personalized goal setting to incrementally increase physical activity (with feedback and review of goals through SMS text messaging contact with the research team). Direct measures of physical activity were collected using a Fitbit accelerometer. Data on anthropometric, functional exercise capacity, dietary behavior, and psychological measures were collected at face-to-face assessments in a single hospital site at baseline (T0), 12 weeks (T1; intervention end), and 24 weeks (T2; follow-up).

Results

The rate of attrition was 21% (13/61) for the control condition and 14% (9/62) for the intervention condition. Using intent-to-treat analysis, significant reductions in BMI (F_2,242=4.149; P=.02; ηp²=0.033) and waist circumference (F_2,242=3.342; P=.04; ηp²=0.027) were observed in the intervention group. Over the 24-week study, BMI was reduced by 0.52 in the intervention condition, relative to a nonsignificant reduction of 0.11 in the control arm. Waist circumference was reduced by 3.02 cm in the intervention condition relative to 1.82 cm in the control condition. Physical activity level was significantly higher in the intervention group on 8 of the 12 weeks of the intervention phase and on 5 of the 12 weeks of the follow-up period, accounting for up to 2500 additional steps per day (mean 2032, SD 270).

Conclusions

The results demonstrate that for cancer survivors with a BMI≥25 kg/m², lifestyle education and personalized goal setting using mobile technology can yield significant changes in clinically relevant health indicators. Further research is needed to elucidate the mechanisms of behavior change and explore the capacity for mHealth interventions to improve broader health and well-being outcomes in the growing population of cancer survivors.

Trial Registration

ISRCTN Registry ISRCTN18676721; https://www.isrctn.com/ISRCTN18676721

International Registered Report Identifier (IRRID)

RR2-10.2196/13214

mHealth app using machine learning to increase physical activity in diabetes and depression: clinical trial protocol for the DIAMANTE Study

INTRODUCTION

Depression and diabetes are highly disabling diseases with a high prevalence and high rate of comorbidity, particularly in low-income ethnic minority patients. Though comorbidity increases the risk of adverse outcomes and mortality, most clinical interventions target these diseases separately. Increasing physical activity might be effective to simultaneously lower depressive symptoms and improve glycaemic control. Self-management apps are a cost-effective, scalable and easy access treatment to increase physical activity. However, cutting-edge technological applications often do not reach vulnerable populations and are not tailored to an individual's behaviour and characteristics. Tailoring of interventions using machine learning methods likely increases the effectiveness of the intervention.

METHODS AND ANALYSIS

In a three-arm randomised controlled trial, we will examine the effect of a text-messaging smartphone application to encourage physical activity in low-income ethnic minority patients with comorbid diabetes and depression. The adaptive intervention group receives messages chosen from different messaging banks by a reinforcement learning algorithm. The uniform random intervention group receives the same messages, but chosen from the messaging banks with equal probabilities. The control group receives a weekly mood message. We aim to recruit 276 adults from primary care clinics aged 18-75 years who have been diagnosed with current diabetes and show elevated depressive symptoms (Patient Health Questionnaire depression scale-8 (PHQ-8) >5). We will compare passively collected daily step counts, self-report PHQ-8 and most recent haemoglobin A1c from medical records at baseline and at intervention completion at 6-month follow-up.

ETHICS AND DISSEMINATION

The Institutional Review Board at the University of California San Francisco approved this study (IRB: 17-22608). We plan to submit manuscripts describing our user-designed methods and testing of the adaptive learning algorithm and will submit the results of the trial for publication in peer-reviewed journals and presentations at (inter)-national scientific meetings.

TRIAL REGISTRATION NUMBER

NCT03490253; pre-results.

Reducing cardiometabolic risk in adults with a low socioeconomic position: protocol of the Supreme Nudge parallel cluster-randomised controlled supermarket trial

Background

Unhealthy lifestyle behaviours such as unhealthy dietary intake and insufficient physical activity (PA) tend to cluster in adults with a low socioeconomic position (SEP), putting them at high cardiometabolic disease risk. Educational approaches aiming to improve lifestyle behaviours show limited effect in this population. Using environmental and context-specific interventions may create opportunities for sustainable behaviour change. In this study protocol, we describe the design of a real-life supermarket trial combining nudging, pricing and a mobile PA app with the aim to improve lifestyle behaviours and lower cardiometabolic disease risk in adults with a low SEP.

Methods

The Supreme Nudge trial includes nudging and pricing strategies cluster-randomised on the supermarket level, with: i) control group receiving no intervention; ii) group 1 receiving healthy food nudges (e.g., product placement or promotion); iii) group 2 receiving nudges and pricing strategies (taxing of unhealthy foods and subsidizing healthy foods). In collaboration with a Dutch supermarket chain we will select nine stores located in low SEP neighbourhoods, with the nearest competitor store at > 1 km distance and managed by a committed store manager. Across the clusters, a personalized mobile coaching app targeting walking behaviour will be randomised at the individual level, with: i) control group; ii) a group receiving the mobile PA app. All participants (target n = 1485) should be Dutch-speaking, aged 45–75 years with a low SEP and purchase more than half of their household grocery shopping at the selected supermarkets. Participants will be recruited via advertisements and mail-invitations followed by community-outreach methods. Primary outcomes are changes in systolic blood pressure, LDL-cholesterol, HbA1c and dietary intake after 12 months follow-up. Secondary outcomes are changes in diastolic blood pressure, blood lipid markers, waist circumference, steps per day, and behavioural factors including healthy food purchasing, food decision style, social cognitive factors related to nudges and to walking behaviours and customer satisfaction after 12 months follow-up. The trial will be reflexively monitored to support current and future implementation.

Discussion

The findings can guide future research and public health policies on reducing lifestyle-related health inequalities, and contribute to a supermarket-based health promotion intervention implementation roadmap.

Trial registration

Dutch Trial Register ID NL7064, 30th of May, 2018

Mobile Apps for Health Behavior Change in Physical Activity, Diet, Drug and Alcohol Use, and Mental Health: Systematic Review

Background

With a growing focus on patient interaction with health management, mobile apps are increasingly used to deliver behavioral health interventions. The large variation in these mobile health apps—their target patient group, health behavior, and behavioral change strategies—has resulted in a large but incohesive body of literature.

Objective

This systematic review aimed to assess the effectiveness of mobile apps in improving health behaviors and outcomes and to examine the inclusion and effectiveness of behavior change techniques (BCTs) in mobile health apps.

Methods

PubMed, EMBASE, CINAHL, and Web of Science were systematically searched for articles published between 2014 and 2019 that evaluated mobile apps for health behavior change. Two authors independently screened and selected studies according to the eligibility criteria. Data were extracted and the risk of bias was assessed by one reviewer and validated by a second reviewer.

Results

A total of 52 randomized controlled trials met the inclusion criteria and were included in the analysis—37 studies focused on physical activity, diet, or a combination of both, 11 on drug and alcohol use, and 4 on mental health. Participant perceptions were generally positive—only one app was rated as less helpful and satisfactory than the control—and the studies that measured engagement and usability found relatively high study completion rates (mean 83%; n=18, N=39) and ease-of-use ratings (3 significantly better than control, 9/15 rated >70%). However, there was little evidence of changed behavior or health outcomes.

Conclusions

There was no strong evidence in support of the effectiveness of mobile apps in improving health behaviors or outcomes because few studies found significant differences between the app and control groups. Further research is needed to identify the BCTs that are most effective at promoting behavior change. Improved reporting is necessary to accurately evaluate the mobile health app effectiveness and risk of bias.

Mobile Apps for Weight Management: A Review of the Latest Evidence to Inform Practice

Over the last decade, mobile technology has emerged as a potentially useful platform to facilitate weight management and tackle the current obesity epidemic. Clinicians are being more frequently asked to give advice about the usefulness of mobile apps and many individuals have already integrated apps into their attempts to manage weight. Hence, it is imperative for clinicians involved in weight management to be aware of the latest developments and knowledge about available mobile apps and their usefulness in this field. A number of newly published studies have demonstrated promising results of mobile-based interventions for weight management across different populations, but the extent of their effectiveness remains widely debated. This narrative literature review synthesizes the latest evidence, primarily from randomized controlled trials (RCTs), regarding the clinical use of mobile applications for weight management, as well as highlight key limitations associated with their use and directions for future research and practice. Overall, evidence suggests that mobile applications may be useful as low-intensity approaches or adjuncts to conventional weight management strategies. However, there is insufficient evidence to support their use as stand-alone intensive approaches to weight management. Further research is needed to clarify the extent of utility of these applications, as well as the measures required to maximize their potential both as stand-alone approaches and adjuncts to more intensive programs.

Applying machine learning to predict future adherence to physical activity programs

BACKGROUND

Identifying individuals who are unlikely to adhere to a physical exercise regime has potential to improve physical activity interventions. The aim of this paper is to develop and test adherence prediction models using objectively measured physical activity data in the Mobile Phone-Based Physical Activity Education program (mPED) trial. To the best of our knowledge, this is the first to apply Machine Learning methods to predict exercise relapse using accelerometer-recorded physical activity data.

METHODS

We use logistic regression and support vector machine methods to design two versions of a Discontinuation Prediction Score (DiPS), which uses objectively measured past data (e.g., steps and goal achievement) to provide a numerical quantity indicating the likelihood of exercise relapse in the upcoming week. The respective prediction accuracy of these two versions of DiPS are compared, and then numerical simulation is performed to explore the potential of using DiPS to selectively allocate financial incentives to participants to encourage them to increase physical activity.

RESULTS

we had access to a physical activity trial data that were continuously collected every 60 sec every day for 9 months in 210 participants. By using the first 15 weeks of data as training and test on weeks 16-30, we show that both versions of DiPS have a test AUC of 0.9 with high sensitivity and specificity in predicting the probability of exercise adherence. Simulation results assuming different intervention regimes suggest the potential benefit of using DiPS as a score to allocate resources in physical activity intervention programs in reducing costs over other allocation schemes.

CONCLUSIONS

DiPS is capable of making accurate and robust predictions for future weeks. The most predictive features are steps and physical activity intensity. Furthermore, the use of DiPS scores can be a promising approach to determine when or if to provide just-in-time messages and step goal adjustments to improve compliance. Further studies on the use of DiPS in the design of physical activity promotion programs are warranted.

TRIAL REGISTRATION

ClinicalTrials.gov NCT01280812 Registered on January 21, 2011.

A loyalty scheme to encourage physical activity in office workers: a cluster RCT

Background

Increasing physical activity in the workplace can provide physical and mental health benefits for employees and economic benefits for the employer through reduced absenteeism and increased productivity. However, there is limited evidence on effective behaviour change interventions in workplace settings that led to maintained physical activity. This study aimed to address this gap and contribute to the evidence base for effective and cost-effective workplace interventions.

Objectives

To determine the effectiveness and cost-effectiveness of the Physical Activity Loyalty scheme, a multicomponent intervention based on concepts similar to those that underpin a high-street loyalty card, which was aimed at encouraging habitual physical activity behaviour and maintaining increases in mean number of steps per day.

Design

A cluster randomised controlled trial with an embedded economic evaluation, behavioural economic experiments, mediation analyses and process evaluation.

Setting

Office-based employees from public sector organisations in Belfast and Lisburn city centres in Northern Ireland.

Participants

A total of 853 participants [mean age 43.6 years (standard deviation 9.6 years); 71% of participants were female] were randomly allocated by cluster to either the intervention group or the (waiting list) control group.

Intervention

The 6-month intervention consisted of financial incentives (retail vouchers), feedback and other evidence-based behaviour change techniques. Sensors situated in the vicinity of the workplaces allowed participants to monitor their accumulated minutes of physical activity.

Main outcome measures

The primary outcome was mean number of steps per day recorded using a sealed pedometer (Yamax Digiwalker CW-701; Yamax, Tasley, UK) worn on the waist for 7 consecutive days and at 6 and 12 months post intervention. Secondary outcomes included health, mental well-being, quality of life, work absenteeism and presenteeism, and the use of health-care resources.

Results

The mean number of steps per day were significantly lower for the intervention group than the control group [6990 mean number of steps per day (standard deviation 3078) vs. 7576 mean number of steps per day (standard deviation 3345), respectively], with an adjusted mean difference of –336 steps (95% confidence interval –612 to –60 steps; p = 0.02) at 6 months post baseline, but not significantly lower at 12 months post baseline. There was a small but significant enhancement of mental well-being in the intervention group (difference between groups for the Warwick–Edinburgh Mental Wellbeing Scale of 1.34 points, 95% confidence interval 0.48 to 2.20 points), but not for the other secondary outcomes. An economic evaluation suggested that, overall, the scheme was not cost-effective compared with no intervention. The intervention was £25.85 (95% confidence interval –£29.89 to £81.60) more costly per participant than no intervention and had no effect on quality-adjusted life-years (incremental quality-adjusted life-years –0.0000891, 95% confidence interval –0.008 to 0.008).

Limitations

Significant restructuring of participating organisations during the study resulted in lower than anticipated recruitment and retention rates. Technical issues affected intervention fidelity.

Conclusions

Overall, assignment to the intervention group resulted in a small but significant decline in the mean pedometer-measured steps per day at 6 months relative to baseline, compared with the waiting list control group. The Physical Activity Loyalty scheme was deemed not to be cost-effective compared with no intervention, primarily because no additional quality-adjusted life-years were gained through the intervention. Research to better understand the mechanisms of physical activity behaviour change maintenance will help the design of future interventions.

Trial registration

Current Controlled Trials ISRCTN17975376.

Funding

This project was funded by the National Institute for Health Research (NIHR) Public Health Research programme and will be published in full in Public Health Research; Vol. 7, No. 15. See the NIHR Journals Library website for further project information.

Behavioral Modeling in Weight Loss Interventions

Designing systems with human agents is difficult because it often requires models that characterize agents' responses to changes in the system's states and inputs. An example of this scenario occurs when designing treatments for obesity. While weight loss interventions through increasing physical activity and modifying diet have found success in reducing individuals' weight, such programs are difficult to maintain over long periods of time due to lack of patient adherence. A promising approach to increase adherence is through the personalization of treatments to each patient. In this paper, we make a contribution towards treatment personalization by developing a framework for predictive modeling using utility functions that depend upon both time-varying system states and motivational states evolving according to some modeled process corresponding to qualitative social science models of behavior change. Computing the predictive model requires solving a bilevel program, which we reformulate as a mixed-integer linear program (MILP). This reformulation provides the first (to our knowledge) formulation for Bayesian inference that uses empirical histograms as prior distributions. We study the predictive ability of our framework using a data set from a weight loss intervention, and our predictive model is validated by comparison to standard machine learning approaches. We conclude by describing how our predictive model could be used for optimization, unlike standard machine learning approaches which cannot.

Community Engagement in the Development of an mHealth-Enabled Physical Activity and Cardiovascular Health Intervention (Step It Up): Pilot Focus Group Study

BACKGROUND

Community-based participatory research is an effective tool for improving health outcomes in minority communities. Few community-based participatory research studies have evaluated methods of optimizing smartphone apps for health technology-enabled interventions in African Americans.

OBJECTIVE

This study aimed to utilize focus groups (FGs) for gathering qualitative data to inform the development of an app that promotes physical activity (PA) among African American women in Washington, DC.

METHODS

We recruited a convenience sample of African American women (N=16, age range 51-74 years) from regions of Washington, DC metropolitan area with the highest burden of cardiovascular disease. Participants used an app created by the research team, which provided motivational messages through app push notifications and educational content to promote PA. Subsequently, participants engaged in semistructured FG interviews led by moderators who asked open-ended questions about participants' experiences of using the app. FGs were audiorecorded and transcribed verbatim, with subsequent behavioral theory-driven thematic analysis. Key themes based on the Health Belief Model and emerging themes were identified from the transcripts. Three independent reviewers iteratively coded the transcripts until consensus was reached. Then, the final codebook was approved by a qualitative research expert.

RESULTS

In this study, 10 main themes emerged. Participants emphasized the need to improve the app by optimizing automation, increasing relatability (eg, photos that reflect target demographic), increasing educational material (eg, health information), and connecting with community resources (eg, cooking classes and exercise groups).

CONCLUSIONS

Involving target users in the development of a culturally sensitive PA app is an essential step for creating an app that has a higher likelihood of acceptance and use in a technology-enabled intervention. This may decrease health disparities in cardiovascular diseases by more effectively increasing PA in a minority population.

An Adaptive, Data-Driven Personalized Advisor for Increasing Physical Activity

In recent years, there has been growing interest in the use of fitness trackers and smartphone applications for promoting physical activity. Many of these applications use accelerometers to estimate the level of activity that users engage in and provide visual reports of a user's step counts. When provided, most recommendations are limited to popular general health advice. In our study, we develop an approach for providing data-driven and personalized recommendations for intraday activity planning. We generate an hour-by-hour activity plan that is based on the user's probability of adhering to the plan. The user's probability of adherence to the plan is personalized, based on his/her past activity patterns and current activity target. Using this approach, we can tailor notifications (e.g., reminders, encouragement) to each user. We can also dynamically update the user's activity plan at mid-day, if his/her actual activity deviates sufficiently from the original plan. In this paper, we describe an implementation of our approach and report our technical findings with respect to identifying typical activity patterns from historical data, predicting whether an activity target will be achieved, and adapting an activity plan based on a user's actual performance throughout the day.

Can the artificial intelligence technique of reinforcement learning use continuously-monitored digital data to optimize treatment for weight loss?

Behavioral weight loss (WL) trials show that, on average, participants regain lost weight unless provided long-term, intensive-and thus costly-intervention. Optimization solutions have shown mixed success. The artificial intelligence principle of "reinforcement learning" (RL) offers a new and more sophisticated form of optimization in which the intensity of each individual's intervention is continuously adjusted depending on patterns of response. In this pilot, we evaluated the feasibility and acceptability of a RL-based WL intervention, and whether optimization would achieve equivalent benefit at a reduced cost compared to a non-optimized intensive intervention. Participants (n = 52) completed a 1-month, group-based in-person behavioral WL intervention and then (in Phase II) were randomly assigned to receive 3 months of twice-weekly remote interventions that were non-optimized (NO; 10-min phone calls) or optimized (a combination of phone calls, text exchanges, and automated messages selected by an algorithm). The Individually-Optimized (IO) and Group-Optimized (GO) algorithms selected interventions based on past performance of each intervention for each participant, and for each group member that fit into a fixed amount of time (e.g., 1 h), respectively. Results indicated that the system was feasible to deploy and acceptable to participants and coaches. As hypothesized, we were able to achieve equivalent Phase II weight losses (NO = 4.42%, IO = 4.56%, GO = 4.39%) at roughly one-third the cost (1.73 and 1.77 coaching hours/participant for IO and GO, versus 4.38 for NO), indicating strong promise for a RL system approach to weight loss and maintenance.

Personalised nutrition and health

Jose Ordovas and colleagues consider that nutrition interventions tailored to individual characteristics and behaviours have promise but more work is needed before they can deliver