Trading pounds for points: Engagement and weight loss in a mobile health intervention

App Engagement as a Predictor of Weight Loss in Blended-Care Interventions: Retrospective Observational Study Using Large-Scale Real-World Data

BACKGROUND

Early weight loss is an established predictor for treatment outcomes in weight management interventions for people with obesity. However, there is a paucity of additional, reliable, and clinically actionable early predictors in weight management interventions. Novel blended-care weight management interventions combine coach and app support and afford new means of structured, continuous data collection, informing research on treatment adherence and outcome prediction.

OBJECTIVE

Against this backdrop, this study analyzes app engagement as a predictor for weight loss in large-scale, real-world, blended-care interventions. We hypothesize that patients who engage more frequently in app usage in blended-care treatment (eg, higher logging activity) lose more weight than patients who engage comparably less frequently at 3 and 6 months of intervention.

METHODS

Real-world data from 19,211 patients in obesity treatment were analyzed retrospectively. Patients were treated with 3 different blended-care weight management interventions, offered in Switzerland, the United Kingdom, and Germany by a digital behavior change provider. The principal component analysis identified an overarching metric for app engagement based on app usage. A median split informed a distinction in higher and lower engagers among the patients. Both groups were matched through optimal propensity score matching for relevant characteristics (eg, gender, age, and start weight). A linear regression model, combining patient characteristics and app-derived data, was applied to identify predictors for weight loss outcomes.

RESULTS

For the entire sample (N=19,211), mean weight loss was -3.24% (SD 4.58%) at 3 months and -5.22% (SD 6.29%) at 6 months. Across countries, higher app engagement yielded more weight loss than lower engagement after 3 but not after 6 months of intervention (P3 months<.001 and P6 months=.59). Early app engagement within the first 3 months predicted percentage weight loss in Switzerland and Germany, but not in the United Kingdom (PSwitzerland<.001, PUnited Kingdom=.12, and PGermany=.005). Higher age was associated with stronger weight loss in the 3-month period (PSwitzerland=.001, PUnited Kingdom=.002, and PGermany<.001) and, for Germany, also in the 6-month period (PSwitzerland=.09, PUnited Kingdom=.46, and PGermany=.03). In Switzerland, higher numbers of patients' messages to coaches were associated with higher weight loss (P3 months<.001 and P6 months<.001). Messages from coaches were not significantly associated with weight loss (all P>.05).

CONCLUSIONS

Early app engagement is a predictor of weight loss, with higher engagement yielding more weight loss than lower engagement in this analysis. This new predictor lends itself to automated monitoring and as a digital indicator for needed or adapted clinical action. Further research needs to establish the reliability of early app engagement as a predictor for treatment adherence and outcomes. In general, the obtained results testify to the potential of app-derived data to inform clinical monitoring practices and intervention design.

Mobile health (m-health) smartphone interventions for adolescents and adults with overweight or obesity

BACKGROUND

Obesity is considered to be a risk factor for various diseases, and its incidence has tripled worldwide since 1975. In addition to potentially being at risk for adverse health outcomes, people with overweight or obesity are often stigmatised. Behaviour change interventions are increasingly delivered as mobile health (m-health) interventions, using smartphone apps and wearables. They are believed to support healthy behaviours at the individual level in a low-threshold manner.

OBJECTIVES

To assess the effects of integrated smartphone applications for adolescents and adults with overweight or obesity.

SEARCH METHODS

We searched CENTRAL, MEDLINE, PsycINFO, CINAHL, and LILACS, as well as the trials registers ClinicalTrials.gov and World Health Organization International Clinical Trials Registry Platform on 2 October 2023 (date of last search for all databases). We placed no restrictions on the language of publication.

SELECTION CRITERIA

Participants were adolescents and adults with overweight or obesity. Eligible interventions were integrated smartphone apps using at least two behaviour change techniques. The intervention could target physical activity, cardiorespiratory fitness, weight loss, healthy diet, or self-efficacy. Comparators included no or minimal intervention (NMI), a different smartphone app, personal coaching, or usual care. Eligible studies were randomised controlled trials of any duration with a follow-up of at least three months.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methodology and the RoB 2 tool. Important outcomes were physical activity, body mass index (BMI) and weight, health-related quality of life, self-efficacy, well-being, change in dietary behaviour, and adverse events. We focused on presenting studies with medium- (6 to < 12 months) and long-term (≥ 12 months) outcomes in our summary of findings table, following recommendations in the core outcome set for behavioural weight management interventions.

MAIN RESULTS

We included 18 studies with 2703 participants. Interventions lasted from 2 to 24 months. The mean BMI in adults ranged from 27 to 50, and the median BMI z-score in adolescents ranged from 2.2 to 2.5. Smartphone app versus no or minimal intervention Thirteen studies compared a smartphone app versus NMI in adults; no studies were available for adolescents. The comparator comprised minimal health advice, handouts, food diaries, smartphone apps unrelated to weight loss, and waiting list. Measures of physical activity: at 12 months' follow-up, a smartphone app compared to NMI probably reduces moderate to vigorous physical activity (MVPA) slightly (mean difference (MD) -28.9 min/week (95% confidence interval (CI) -85.9 to 28; 1 study, 650 participants; moderate-certainty evidence)). We are very uncertain about the results of estimated energy expenditure and cardiorespiratory fitness at eight months' follow-up. A smartphone app compared with NMI probably results in little to no difference in changes in total activity time at 12 months' follow-up and leisure time physical activity at 24 months' follow-up. Anthropometric measures: a smartphone app compared with NMI may reduce BMI (MD of BMI change -2.6 kg/m2, 95% CI -6 to 0.8; 2 studies, 146 participants; very low-certainty evidence) at six to eight months' follow-up, but the evidence is very uncertain. At 12 months' follow-up, a smartphone app probably resulted in little to no difference in BMI change (MD -0.1 kg/m2, 95% CI -0.4 to 0.3; 1 study; 650 participants; moderate-certainty evidence). A smartphone app compared with NMI may result in little to no difference in body weight change (MD -2.5 kg, 95% CI -6.8 to 1.7; 3 studies, 1044 participants; low-certainty evidence) at 12 months' follow-up. At 24 months' follow-up, a smartphone app probably resulted in little to no difference in body weight change (MD 0.7 kg, 95% CI -1.2 to 2.6; 1 study, 245 participants; moderate-certainty evidence). A smartphone app compared with NMI may result in little to no difference in self-efficacy for a physical activity score at eight months' follow-up, but the results are very uncertain. A smartphone app probably results in little to no difference in quality of life and well-being at 12 months (moderate-certainty evidence) and in little to no difference in various measures used to inform dietary behaviour at 12 and 24 months' follow-up. We are very uncertain about adverse events, which were only reported narratively in two studies (very low-certainty evidence). Smartphone app versus another smartphone app Two studies compared different versions of the same app in adults, showing no or minimal differences in outcomes. One study in adults compared two different apps (calorie counting versus ketogenic diet) and suggested a slight reduction in body weight at six months in favour of the ketogenic diet app. No studies were available for adolescents. Smartphone app versus personal coaching Only one study compared a smartphone app with personal coaching in adults, presenting data at three months. Two studies compared these interventions in adolescents. A smartphone app resulted in little to no difference in BMI z-score compared to personal coaching at six months' follow-up (MD 0, 95% CI -0.2 to 0.2; 1 study; 107 participants). Smartphone app versus usual care Only one study compared an app with usual care in adults but only reported data at three months on participant satisfaction. No studies were available for adolescents. We identified 34 ongoing studies.

AUTHORS' CONCLUSIONS

The available evidence is limited and does not demonstrate a clear benefit of smartphone applications as interventions for adolescents or adults with overweight or obesity. While the number of studies is growing, the evidence remains incomplete due to the high variability of the apps' features, content and components, which complicates direct comparisons and assessment of their effectiveness. Comparisons with either no or minimal intervention or personal coaching show minor effects, which are mostly not clinically significant. Minimal data for adolescents also warrants further research. Evidence is also scarce for low- and middle-income countries as well as for people with different socio-economic and cultural backgrounds. The 34 ongoing studies suggest sustained interest in the topic, with new evidence expected to emerge within the next two years. In practice, clinicians and healthcare practitioners should carefully consider the potential benefits, limitations, and evolving research when recommending smartphone apps to adolescents and adults with overweight or obesity.

Enhancing digital health services: A machine learning approach to personalized exercise goal setting

Background

The utilization of digital health has increased recently, and these services provide extensive guidance to encourage users to exercise frequently by setting daily exercise goals to promote a healthy lifestyle. These comprehensive guides evolved from the consideration of various personalized behavioral factors. Nevertheless, existing approaches frequently neglect the users' dynamic behavior and the changing in their health conditions.

Objective

This study aims to fill this gap by developing a machine learning algorithm that dynamically updates auto-suggestion exercise goals using retrospective data and realistic behavior trajectory.

Methods

We conducted a methodological study by designing a deep reinforcement learning algorithm to evaluate exercise performance, considering fitness-fatigue effects. The deep reinforcement learning algorithm combines deep learning techniques to analyze time series data and infer user's exercise behavior. In addition, we use the asynchronous advantage actor-critic algorithm for reinforcement learning to determine the optimal exercise intensity through exploration and exploitation. The personalized exercise data and biometric data used in this study were collected from publicly available datasets, encompassing walking, sports logs, and running.

Results

In our study, we conducted the statistical analyses/inferential tests to compare the effectiveness of machine learning approach in exercise goal setting across different exercise goal-setting strategies. The 95% confidence intervals demonstrated the robustness of these findings, emphasizing the superior outcomes of the machine learning approach.

Conclusions

Our study demonstrates the adaptability of machine learning algorithm to users' exercise preferences and behaviors in exercise goal setting, emphasizing the substantial influence of goal design on service effectiveness.

Ethical and Information Governance Considerations for Promoting Digital Social Interventions in Primary Care

Promoting online peer support beyond the informal sector to statutory health services requires ethical considerations and evidence-based knowledge about its impact on patients, health care professionals, and the wider health care system. Evidence on the effectiveness of digital interventions in primary care is sparse, and definitive guidance is lacking on the ethical concerns arising from the use of social media as a means for health-related interventions and research. Existing literature examining ethical issues with digital interventions in health care mainly focuses on apps, electronic health records, wearables, and telephone or video consultations, without necessarily covering digital social interventions, and does not always account for primary care settings specifically. Here we address the ethical and information governance aspects of undertaking research on the promotion of online peer support to patients by primary care clinicians, related to medical and public health ethics.

Mitigating Issues With/of/for True Personalization

A common but false perception persists about the level and type of personalization in the offerings of contemporary software, information systems, and services, known as Personalization Myopia: this involves a tendency for researchers to think that there are many more personalized services than there genuinely are, for the general audience to think that they are offered personalized services when they really are not, and for practitioners to have a mistaken idea of what makes a service personalized. And yet in an era, which mashes up large amounts of data, business analytics, deep learning, and persuasive systems, true personalization is a most promising approach for innovating and developing new types of systems and services—including support for behavior change. The potential of true personalization is elaborated in this article, especially with regards to persuasive software features and the oft-neglected fact that users change over time.

Challenges in Participant Engagement and Retention using Mobile Health Apps: A Literature Review (Preprint)

Background

Mobile health (mHealth) apps are revolutionizing the way clinicians and researchers monitor and manage the health of their participants. However, many studies using mHealth apps are hampered by substantial participant dropout or attrition, which may impact the representativeness of the sample and the effectiveness of the study. Therefore, it is imperative for researchers to understand what makes participants stay with mHealth apps or studies using mHealth apps.

Objective

This study aimed to review the current peer-reviewed research literature to identify the notable factors and strategies used in adult participant engagement and retention.

Methods

We conducted a systematic search of PubMed, MEDLINE, and PsycINFO databases for mHealth studies that evaluated and assessed issues or strategies to improve the engagement and retention of adults from 2015 to 2020. We followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Notable themes were identified and narratively compared among different studies. A binomial regression model was generated to examine the factors affecting retention.

Results

Of the 389 identified studies, 62 (15.9%) were included in this review. Overall, most studies were partially successful in maintaining participant engagement. Factors related to particular elements of the app (eg, feedback, appropriate reminders, and in-app support from peers or coaches) and research strategies (eg, compensation and niche samples) that promote retention were identified. Factors that obstructed retention were also identified (eg, lack of support features, technical difficulties, and usefulness of the app). The regression model results showed that a participant is more likely to drop out than to be retained.

Conclusions

Retaining participants is an omnipresent challenge in mHealth studies. The insights from this review can help inform future studies about the factors and strategies to improve participant retention.

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.

Influence of Personality on mHealth Use in Patients with Diabetes: Prospective Pilot Study

Background

Mobile technology for health (mHealth) interventions are increasingly being used to help improve self-management among patients with diabetes; however, these interventions have not been adopted by a large number of patients and often have high dropout rates. Patient personality characteristics may play a critical role in app adoption and active utilization, but few studies have focused on addressing this question.

Objective

This study aims to address a gap in understanding of the relationship between personality traits and mHealth treatment for patients with diabetes. We tested the role of the five-factor model of personality traits (openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism) in mHealth adoption preference and active utilization.

Methods

We developed an mHealth app (DiaSocial) aimed to encourage diabetes self-management. We recruited 98 patients with diabetes—each patient freely chose whether to receive the standard care or the mHealth app intervention. Patient demographic information and patient personality characteristics were assessed at baseline. App usage data were collected to measure user utilization of the app. Patient health outcomes were assessed with lab measures of glycated hemoglobin (HbA_1c level). Logistic regression models and linear regression were employed to explore factors predicting the relationship between mHealth use (adoption and active utilization) and changes in health outcome.

Results

Of 98 study participants, 46 (47%) downloaded and used the app. Relatively younger patients with diabetes were 9% more likely to try and use the app (P=.02, odds ratio [OR] 0.91, 95% CI 0.85-0.98) than older patients with diabetes were. Extraversion was negatively associated with adoption of the mHealth app (P=.04, OR 0.71, 95% CI 0.51-0.98), and openness to experience was positively associated with adoption of the app (P=.03, OR 1.73, 95% CI 1.07-2.80). Gender (P=.43, OR 0.66, 95% CI 0.23-1.88), education (senior: P=.99, OR 1.00, 95% CI 0.32-3.11; higher: P=.21, OR 2.51, 95% CI 0.59-10.66), and baseline HbA_1c level (P=.36, OR 0.79, 95% CI 0.47-1.31) were not associated with app adoption. Among those who adopted the app, a low education level (senior versus primary P=.003; higher versus primary P=.03) and a high level of openness to experience (P=.048, OR 2.01, 95% CI 1.01-4.00) were associated with active app utilization. Active users showed a significantly greater decrease in HbA_1c level than other users (ΔHbA_1c=−0.64, P=.05).

Conclusions

This is one of the first studies to investigate how different personality traits influence the adoption and active utilization of an mHealth app among patients with diabetes. The research findings suggest that personality is a factor that should be considered when trying to identify patients who would benefit the most from apps for diabetes management.

Unpacking mHealth interventions: A systematic review of behavior change techniques used in randomized controlled trials assessing mHealth effectiveness

Objective

Mobile health interventions have surged in popularity but their implementation varies widely and evidence of effectiveness is mixed. We sought to advance understanding of the diversity of behavior change techniques in mHealth interventions, especially those that leverage advanced mobile technologies.

Methods

We conducted a systematic review of articles published between 2007 and 2017 in high-impact journals in medicine, medical informatics, and health psychology to identify randomized controlled trials in which the effectiveness of an mobile health intervention was tested. Search terms included a mix of general (e.g. mobile health), hardware (e.g. Android, iPhone), and format (e.g. SMS, application) terms.

Results

In a systematic review of 21 studies, we found the techniques of personalization, feedback and monitoring, and associations were most commonly used in mobile health interventions, but there remains considerable opportunity to leverage more sophisticated aspects of ubiquitous computing. We found that prompts and cues were the most common behavior change techniques used in effective trials, but there was notable overlap in behavior change techniques used in ineffective trials.

Conclusions

Our results identify techniques that are commonly used in mobile health interventions and highlight pathways to advance the science of mobile health.

Designing for engagement with self-monitoring: A user-centered approach with low-income, Latino adults with Type 2 Diabetes

BACKGROUND AND SIGNIFICANCE

Data-driven interventions for health can help to personalize self-management of Type 2 Diabetes (T2D), but lack of sustained engagement with self-monitoring among disadvantaged populations may widen existing health disparities. Prior work developing approaches to increase motivation and engagement with self-monitoring holds promise, but little is known about applicability of these approaches to underserved populations.

OBJECTIVE

To explore how low-income, Latino adults with T2D respond to different design concepts for data-driven solutions in health that require self-monitoring, and what features resonate with them the most.

MATERIAL AND METHODS

We developed a set of mockups that incorporated different design features for promoting engagement with self-monitoring in T2D. We conducted focus groups to examine individuals' perceptions and attitudes towards mockups. Multiple interdisciplinary researchers analyzed data using directed content analysis.

RESULTS

We conducted 14 focus groups with 25 English- and Spanish-speaking adults with T2D. All participants reacted positively to external incentives. Social connectedness and healthcare expert feedback were also well liked because they enhanced current social practices and presented opportunities for learning. However, attitudes were more mixed towards goal setting, and very few participants responded positively to self-discovery and personalized decision support features. Instead, participants wished for personalized recommendations for meals and other health behaviors based on their personal health data.

CONCLUSION

This study suggests connections between individuals' degree of internal motivation and motivation for self-monitoring in health and their attitude towards designs of self-monitoring apps. We relate our findings to the self-determination continuum in self-determination theory (SDT) and propose it as a blueprint for aligning incentives for self-monitoring to different levels of motivation.

Defining Adherence to Mobile Dietary Self-Monitoring and Assessing Tracking Over Time: Tracking at Least Two Eating Occasions per Day Is Best Marker of Adherence within Two Different Mobile Health Randomized Weight Loss Interventions

BACKGROUND

Mobile dietary self-monitoring methods allow for objective assessment of adherence to self-monitoring; however, the best way to define self-monitoring adherence is not known.

OBJECTIVE

The objective was to identify the best criteria for defining adherence to dietary self-monitoring with mobile devices when predicting weight loss.

DESIGN

This was a secondary data analysis from two 6-month randomized trials: Dietary Intervention to Enhance Tracking with Mobile Devices (n=42 calorie tracking app or n=39 wearable Bite Counter device) and Self-Monitoring Assessment in Real Time (n=20 kcal tracking app or n=23 photo meal app).

PARTICIPANTS/SETTING

Adults (n=124; mean body mass index=34.7±5.6) participated in one of two remotely delivered weight-loss interventions at a southeastern university between 2015 and 2017.

INTERVENTION

All participants received the same behavioral weight loss information via twice-weekly podcasts. Participants were randomly assigned to a specific diet tracking method.

MAIN OUTCOME MEASURES

Seven methods of tracking adherence to self-monitoring (eg, number of days tracked, and number of eating occasions tracked) were examined, as was weight loss at 6 months.

STATISTICAL ANALYSES PERFORMED

Linear regression models estimated the strength of association (R²) between each method of tracking adherence and weight loss, adjusting for age and sex.

RESULTS

Among all study completers combined (N=91), adherence defined as the overall number of days participants tracked at least two eating occasions explained the most variance in weight loss at 6 months (R²=0.27; P<0.001). Self-monitoring declined over time; all examined adherence methods had fewer than half the sample still tracking after Week 10.

CONCLUSIONS

Using the total number of days at least two eating occasions are tracked using a mobile self-monitoring method may be the best way to assess self-monitoring adherence during weight loss interventions. This study shows that self-monitoring rates decline quickly and elucidates potential times for early interventions to stop the reductions in self-monitoring.

Using social media for health research: Methodological and ethical considerations for recruitment and intervention delivery

As the popularity and diversity of social media platforms increases so does their utility for health research. Using social media for recruitment into clinical studies and/or delivering health behavior interventions may increase reach to a broader audience. However, evidence supporting the efficacy of these approaches is limited, and key questions remain with respect to optimal benchmarks, intervention development and methodology, participant engagement, informed consent, privacy, and data management. Little methodological guidance is available to researchers interested in using social media for health research. In this Tutorial, we summarize the content of the 2017 Society for Behavioral Medicine Pre-Conference Course entitled 'Using Social Media for Research,' at which the authors presented their experiences with methodological and ethical issues relating to social media-enabled research recruitment and intervention delivery. We identify common pitfalls and provide recommendations for recruitment and intervention via social media. We also discuss the ethical and responsible conduct of research using social media for each of these purposes.