Characterizing user engagement with health app data: a data mining approach.
ABSTRACT: The use of mobile health applications (apps) especially in the area of lifestyle behaviors has increased, thus providing unprecedented opportunities to develop health programs that can engage people in real-time and in the real-world. Yet, relatively little is known about which factors relate to the engagement of commercially available apps for health behaviors. This exploratory study examined behavioral engagement with a weight loss app, Lose It! and characterized higher versus lower engaged groups. Cross-sectional, anonymized data from Lose It! were analyzed (n = 12,427,196). This dataset was randomly split into 24 subsamples and three were used for this study (total n = 1,011,008). Classification and regression tree methods were used to identify subgroups of user engagement with one subsample, and descriptive analyses were conducted to examine other group characteristics associated with engagement. Data mining validation methods were conducted with two separate subsamples. On average, users engaged with the app for 29 days. Six unique subgroups were identified, and engagement for each subgroup varied, ranging from 3.5 to 172 days. Highly engaged subgroups were primarily distinguished by the customization of diet and exercise. Those less engaged were distinguished by weigh-ins and the customization of diet. Results were replicated in further analyses. Commercially-developed apps can reach large segments of the population, and data from these apps can provide insights into important app features that may aid in user engagement. Getting users to engage with a mobile health app is critical to the success of apps and interventions that are focused on health behavior change.
Project description:<h4>Background</h4>Understanding how users engage with electronic screening and brief intervention (eSBI) is a critical research objective to improve effectiveness of app-based interventions to reduce harmful alcohol consumption. Although quantitative measures of engagement provide a strong indicator of how the user engages with an app at the group level, they do not elucidate finer-grained details of how apps function from an individual, experiential perspective and why, or how, users engage with an intervention in a particular manner.<h4>Objective</h4>The aim of this study was to (1) understand why and how participants engaged with the BRANCH app, (2) explore facilitators and barriers to engagement with app features, (3) explore how the BRANCH app impacted drinking behavior, (4) use these data to identify typologies of users of the BRANCH app in terms of engagement behaviors, and (5) identify future eSBI app design implications.<h4>Methods</h4>In total, 20 one-to-one semistructured telephone interviews were conducted with participants recruited from a randomized controlled trial, which evaluated the effectiveness of engagement-promoting strategies in the BRANCH app targeting harmful drinking in young adults (aged 18-30 years). The topic guide explored users' current engagement levels with existing health promotion apps, their views toward the effectiveness of such apps, and what they liked and disliked about BRANCH, specifically focusing on how they engaged with the app. Framework analysis was used to develop typologies of user engagement.<h4>Results</h4>The study identified 3 typologies of engagers. Trackers were defined by their motivations to use health-tracking apps to monitor and understand quantified self-data. They did not have intentions necessarily to cut down and predominantly used only the drinking diary. Cut-downers were motivated to use the app because they wanted to reduce their alcohol consumption Unlike Trackers, they did not use a range of different health apps daily, but saw the BRANCH app as an opportunity to test out a different method of trying to cut down their alcohol use. This typology used more features than Trackers, such as the goal setting function. Noncommitters were characterized as a group of users who were initially enthusiastic about using the app; however, this enthusiasm quickly waned and they gained no benefit from it.<h4>Conclusions</h4>This was the first study to identify typologies of user engagement with eSBI apps. Although in need of replication, it provides a first step in understanding independent categories of eSBI users, who may benefit from apps tailored to a user's typology or motivation. It also provides new evidence to suggest that apps may be used more effectively as a tool to raise awareness of drinking, instead of reducing alcohol use, and be a step in the care pathway, identifying at-risk individuals and signposting them to more intensive treatment.<h4>Trial registration</h4>International Standard Randomised Controlled Trial Number ISRCTN70980706; http://www.isrctn.com /ISRCTN70980706 (Archived by WebCite at http://www.webcitation.org/73vfDXYEZ).
Project description:Smartphone applications (apps) might be able to reach pregnant smokers who do not engage with face-to-face support. However, we do not know how far pregnant smokers will engage with smoking cessation apps or what components are likely to be effective. This study aimed to assess pregnant smokers' engagement with the SmokeFree Baby app (v1) and to assess the short-term efficacy of selected components ("modules") for smoking abstinence. Positive outcomes would provide a basis for further development and evaluation. SmokeFree Baby was developed drawing on behavior change theories and relevant evidence. Pregnant smokers (18+) who were interested in quitting and set a quit date were recruited. Following multiphase optimization development principles, participants (N = 565) were randomly allocated to one of 32 (2 × 2 × 2 × 2 × 2) experimental groups in a full factorial design to evaluate five modules (each in minimal and full version: identity, health information, stress management, face-to-face support, and behavioral substitution). Measures of engagement included duration and frequency of engagement with the app. Smoking abstinence was measured by self-reported number of smoke-free days up to 4 weeks from the quit date. Participants engaged with the app for a mean of 4.5 days (SD = 8.5) and logged in a mean of 2.9 times (SD = 3.1). Main effects of the modules on the number of smoke-free days were not statistically significant (identity: p = .782, health information: p = .905, stress management: p = .103, face-to-face support: p = .397, behavioral substitution: p = .945). Despite systematic development and usability testing, engagement with SmokeFree Baby (v1) was low and the app did not appear to increase smoking abstinence during pregnancy.
Project description:Given the widespread adoption and technical possibilities of mobile technology, mobile health apps could be potentially effective tools to intervene in people's daily routines and stimulate physical activity. Self-determination theory and the motivational technology model both suggest that mobile technology can promote health behaviour change by allowing users to customize their online experience when using mobile health apps. However, we know very little about why and for whom customization is most effective. Using a between-subjects experimental design, we tested the effects of customization in mobile health apps among a convenience sample (N?=?203). We assessed the effects of customization on perceived active control over mobile health apps, autonomous motivation to use mobile health apps, and intention to engage in physical activity, and tested the moderating role of need for autonomy. Structural equation modelling showed that customization in mobile health apps does not increase perceived active control, autonomous motivation, or the intention to engage in physical activity. However, an interaction effect between customization and need for autonomy showed that customization in mobile health apps leads to higher intentions to engage in physical activity for those with a greater need for autonomy, but not for those with a lesser need for autonomy. The implications for theory and practice are discussed.
Project description:BACKGROUND:User engagement is key to the effectiveness of digital mental health interventions. Considerable research has examined the clinical outcomes of overall engagement with mental health apps (eg, frequency and duration of app use). However, few studies have examined how specific app use behaviors can drive change in outcomes. Understanding the clinical outcomes of more nuanced app use could inform the design of mental health apps that are more clinically effective to users. OBJECTIVE:This study aimed to classify user behaviors in a suite of mental health apps and examine how different types of app use are related to depression and anxiety outcomes. We also compare the clinical outcomes of specific types of app use with those of generic app use (ie, intensity and duration of app use) to understand what aspects of app use may drive symptom improvement. METHODS:We conducted a secondary analysis of system use data from an 8-week randomized trial of a suite of 13 mental health apps. We categorized app use behaviors through a mixed methods analysis combining qualitative content analysis and principal component analysis. Regression analyses were used to assess the association between app use and levels of depression and anxiety at the end of treatment. RESULTS:A total of 3 distinct clusters of app use behaviors were identified: learning, goal setting, and self-tracking. Each specific behavior had varied effects on outcomes. Participants who engaged in self-tracking experienced reduced depression symptoms, and those who engaged with learning and goal setting at a moderate level (ie, not too much or not too little) also had an improvement in depression. Notably, the combination of these 3 types of behaviors, what we termed "clinically meaningful use," accounted for roughly the same amount of variance as explained by the overall intensity of app use (ie, total number of app use sessions). This suggests that our categorization of app use behaviors succeeded in capturing app use associated with better outcomes. However, anxiety outcomes were neither associated with specific behaviors nor generic app use. CONCLUSIONS:This study presents the first granular examination of user interactions with mental health apps and their effects on mental health outcomes. It has important implications for the design of mobile health interventions that aim to achieve greater user engagement and improved clinical efficacy.
Project description:Mobile applications (apps) have been increasingly utilized to access the latest and abundant information related to genetics/genomics for resources, risk assessments, and individualized recommendations. Nevertheless, the number and quality of the current apps in genetics/genomics remain unknown. Thus, in this review, we aimed to identify existing genetic/genomic apps, summarize their characteristics, and examine their quality. A systematic search of genetics/genomics apps was conducted on Apple Store and Google Play. We adapted a validated evaluation scale, Mobile App Rating Scale (MARS), to examine the quality of genetics/genomics apps. Eighty-eight genetics/genomics apps, with the cost ranging from free to $49.99, formed the final sample. Findings showed that the majority of the apps had reference/resource as a feature (95.5%), had health professional students as the target audience (86.4%), and did not focus on specific diseases (78.5%). Only 21.6% of the apps were developed by reliable or authoritative agencies, and the apps' overall quality was slightly above average based on the criteria of the MARS. Therefore, while genetics/genomics mobile apps might be useful resources, their quality still needs improvement, especially with respect to the credibility and evidence-based items of app information as well as the customization items of app engagement; caution must be taken when using those apps.
Project description:BACKGROUND:There is mixed evidence to support current ambitions for mobile health (mHealth) apps to improve chronic health and well-being. One proposed explanation for this variable effect is that users do not engage with apps as intended. The application of analytics, defined as the use of data to generate new insights, is an emerging approach to study and interpret engagement with mHealth interventions. OBJECTIVE:This study aimed to consolidate how analytic indicators of engagement have previously been applied across clinical and technological contexts, to inform how they might be optimally applied in future evaluations. METHODS:We conducted a scoping review to catalog the range of analytic indicators being used in evaluations of consumer mHealth apps for chronic conditions. We categorized studies according to app structure and application of engagement data and calculated descriptive data for each category. Chi-square and Fisher exact tests of independence were applied to calculate differences between coded variables. RESULTS:A total of 41 studies met our inclusion criteria. The average mHealth evaluation included for review was a two-group pretest-posttest randomized controlled trial of a hybrid-structured app for mental health self-management, had 103 participants, lasted 5 months, did not provide access to health care provider services, measured 3 analytic indicators of engagement, segmented users based on engagement data, applied engagement data for descriptive analyses, and did not report on attrition. Across the reviewed studies, engagement was measured using the following 7 analytic indicators: the number of measures recorded (76%, 31/41), the frequency of interactions logged (73%, 30/41), the number of features accessed (49%, 20/41), the number of log-ins or sessions logged (46%, 19/41), the number of modules or lessons started or completed (29%, 12/41), time spent engaging with the app (27%, 11/41), and the number or content of pages accessed (17%, 7/41). Engagement with unstructured apps was mostly measured by the number of features accessed (8/10, P=.04), and engagement with hybrid apps was mostly measured by the number of measures recorded (21/24, P=.03). A total of 24 studies presented, described, or summarized the data generated from applying analytic indicators to measure engagement. The remaining 17 studies used or planned to use these data to infer a relationship between engagement patterns and intended outcomes. CONCLUSIONS:Although researchers measured on average 3 indicators in a single study, the majority reported findings descriptively and did not further investigate how engagement with an app contributed to its impact on health and well-being. Researchers are gaining nuanced insights into engagement but are not yet characterizing effective engagement for improved outcomes. Raising the standard of mHealth app efficacy through measuring analytic indicators of engagement may enable greater confidence in the causal impact of apps on improved chronic health and well-being.
Project description:BACKGROUND:Most adults do not engage in sufficient physical activity to maintain good health. Smartphone apps are increasingly used to support physical activity but typically focus on tracking behaviors with no support for the complex process of behavior change. Tracking features do not engage all users, and apps could better reach their targets by engaging users in reflecting their reasons, capabilities, and opportunities to change. Motivational interviewing supports this active engagement in self-reflection and self-regulation by fostering psychological needs proposed by the self-determination theory (ie, autonomy, competence, and relatedness). However, it is unknown whether digitalized motivational interviewing in a smartphone app engages users in this process. OBJECTIVE:This study aimed to describe the theory- and evidence-based development of the Precious app and to examine how digitalized motivational interviewing using a smartphone app engages users in the behavior change process. Specifically, we aimed to determine if use of the Precious app elicits change talk in participants and how they perceive autonomy support in the app. METHODS:A multidisciplinary team built the Precious app to support engagement in the behavior change process. The Precious app targets reflective processes with motivational interviewing and spontaneous processes with gamified tools, and builds on the principles of self-determination theory and control theory by using 7 relational techniques and 12 behavior change techniques. The feasibility of the app was tested among 12 adults, who were asked to interact with the prototype and think aloud. Semistructured interviews allowed participants to extend their statements. Participants' interactions with the app were video recorded, transcribed, and analyzed with deductive thematic analysis to identify the theoretical themes related to autonomy support and change talk. RESULTS:Participants valued the autonomy supportive features in the Precious app (eg, freedom to pursue personally relevant goals and receive tailored feedback). We identified the following five themes based on the theory-based theme autonomy support: valuing the chance to choose, concern about lack of autonomy, expecting controlling features, autonomous goals, and autonomy supportive feedback. The motivational interviewing features actively engaged participants in reflecting their outcome goals and reasons for activity, producing several types of change talk and very little sustain talk. The types of change talk identified were desire, need, reasons, ability, commitment, and taking steps toward change. CONCLUSIONS:The Precious app takes a unique approach to engage users in the behavior change process by targeting both reflective and spontaneous processes. It allows motivational interviewing in a mobile form, supports psychological needs with relational techniques, and targets intrinsic motivation with gamified elements. The motivational interviewing approach shows promise, but the impact of its interactive features and tailored feedback needs to be studied over time. The Precious app is undergoing testing in a series of n-of-1 randomized controlled trials.
Project description:BACKGROUND:Early nutrition interventions to improve food knowledge and skills are critical in enhancing the diet quality of children and reducing the lifelong risk of chronic disease. Despite the rise of mobile health (mHealth) apps and their known effectiveness for improving health behaviors, few evidence-based apps exist to help engage children in learning about nutrition and healthy eating. OBJECTIVE:This study aimed to describe the iterative development and user testing of Foodbot Factory, a novel nutrition education gamified app for children to use at home or in the classroom and to present data from user testing experiments conducted to evaluate the app. METHODS:An interdisciplinary team of experts in nutrition, education (pedagogy), and game design led to the creation of Foodbot Factory. First, a literature review and an environmental scan of the app marketplace were conducted, and stakeholders were consulted to define the key objectives and content of Foodbot Factory. Dietitian and teacher stakeholders identified priority age groups and learning objectives. Using a quasi-experimental mixed method design guided by the Iterative Convergent Design for Mobile Health Usability Testing approach, five app user testing sessions were conducted among students (ages 9-12 years). During gameplay, engagement and usability were assessed via direct observations with a semistructured form. After gameplay, qualitative interviews and questionnaires were used to assess user satisfaction, engagement, usability, and knowledge gained. RESULTS:The environmental scan data revealed that few evidence-based nutrition education apps existed for children. A literature search identified key nutrients of concern for Canadian children and techniques that could be incorporated into the app to engage users in learning. Foodbot Factory included characters (2 scientists and Foodbots) who initiate fun and engaging dialogue and challenges (minigames), with storylines incorporating healthy eating messages that align with the established learning objectives. A total of five modules were developed: drinks, vegetables and fruit, grain foods, animal protein foods, and plant protein foods. Seven behavior change techniques and three unique gamified components were integrated into the app. Data from each user testing session were used to inform and optimize the next app iteration. The final user testing session demonstrated that participants agreed that they wanted to play Foodbot Factory again (12/17, 71%), that the app is easy to use (12/17, 71%) and fun (14/17, 88%), and that the app goals were clearly presented (15/17, 94%). CONCLUSIONS:Foodbot Factory is an engaging and educational mHealth intervention for the Canadian public that is grounded in evidence and developed by an interdisciplinary team of experts. The use of an iterative development approach is a demonstrated method to improve engagement, satisfaction, and usability with each iteration. Children find Foodbot Factory to be fun and easy to use, and can engage children in learning about nutrition.
Project description:BACKGROUND:IntelliCare is a mental health app platform with 14 apps that are elemental, simple and brief to use, and eclectic. Although a variety of apps may improve engagement, leading to better outcomes, they may require navigation aids such as recommender systems that can quickly direct a person to a useful app. OBJECTIVE:As the first step toward developing navigation and recommender tools, this study explored app-use patterns across the IntelliCare platform and their relationship with depression and anxiety outcomes. METHODS:This is a secondary analysis of the IntelliCare Field Trial, which recruited people with depression or anxiety. Participants of the trial received 8 weeks of coaching, primarily by text, and weekly random recommendations for apps. App-use metrics included frequency and lifetime use. Depression and anxiety, measured using the Patient Health Questionnaire-9 and Generalized Anxiety Disorder-7, respectively, were assessed at baseline and end of treatment. Cluster analysis was utilized to determine patterns of app use; ordinal logistic regression models and log-rank tests were used to determine if these use metrics alone, or in combination, predicted improvement or remission in depression or anxiety. RESULTS:The analysis included 96 people who generally followed recommendations to download and try new apps each week. Apps were clustered into 5 groups: Thinking (apps that targeted or relied on thinking), Calming (relaxation and insomnia), Checklists (apps that used checklists), Activity (behavioral activation and activity), and Other. Both overall frequency of use and lifetime use predicted response for depression and anxiety. The Thinking, Calming, and Checklist clusters were associated with improvement in depression and anxiety, and the Activity cluster was associated with improvement in Anxiety only. However, the use of clusters was less strongly associated with improvement than individual app use. CONCLUSIONS:Participants in the field trial remained engaged with a suite of apps for the full 8 weeks of the trial. App-use patterns did fall into clusters, suggesting that some knowledge about the use of one app may be useful in selecting another app that the person is more likely to use and may help suggest apps based on baseline symptomology and personal preference.
Project description:BACKGROUND:Diabetes is a significant public health issue. Saudi Arabia has the highest prevalence of type 2 diabetes mellitus (T2DM) in the Arab world. Currently, it affects 31.6% of the general population, and the prevalence of T2DM is predicted to rise to 45.36% by 2030. Mobile health (mHealth) offers improved and cost-effective care to people with T2DM. However, the efficiency of engagement strategies and features of this technology need to be reviewed and standardized according to stakeholder and expert perspectives. OBJECTIVE:The main objective of this study was to identify the most agreed-upon features for T2DM self-management mobile apps; the secondary objective was to identify the most agreed-upon strategies that prompt users to use these apps. METHODS:In this study, a 4-round modified Delphi method was applied by experts in the domain of diabetes care. RESULTS:In total, 11 experts with a mean age of 47.09 years (SD 11.70) consented to participate in the study. Overall, 36 app features were generated. The group of experts displayed weak agreement in their ranking of intervention components (Kendall W=0.275; P<.001). The top 5 features included insulin dose adjustment according to carbohydrate counting and blood glucose readings (5.36), alerting a caregiver of abnormal or critical readings (6.09), nutrition education (12.45), contacts for guidance if required (12.64), and offering patient-specific education tailored to the user's goals, needs, and blood glucose readings (12.90). In total, 21 engagement strategies were generated. Overall, the experts showed a moderate degree of consensus in their strategy rankings (Kendall W=0.454; P<.001). The top 5 engagement strategies included a user-friendly design (educational and age-appropriate design; 2.82), a free app (3.73), allowing the user to communicate or send information/data to a health care provider (HCP; 5.36), HCPs prescribing the mobile app in the clinic and asking about patients' app use compliance during clinical visits (6.91), and flexibility and customization (7.91). CONCLUSIONS:This is the first study in the region consisting of a local panel of experts from the diabetes field gathering together. We used an iterative process to combine the experts' opinions into a group consensus. The results of this study could thus be useful for health app developers and HCPs and inform future decision making on the topic.