Search

This study examined enrollment, retention, engagement, and behavior changes from a caregiver short message service (SMS) component of a larger school-based sugar-sweetened beverage (SSB) reduction intervention. Over 22 weeks, caregivers of seventh graders in 10 Appalachian middle schools received a two-way SMS Baseline Assessment and four monthly follow-up assessments to report their and their child’s SSB intake and select a personalized strategy topic. Between assessments, caregivers received two weekly one-way messages: one information or infographic message and one strategy message. Of 1873 caregivers, 542 (29%) enrolled by completing the SMS Baseline Assessment. Three-quarters completed Assessments 2–5, with 84% retained at Assessment 5. Reminders, used to encourage adherence, improved completion by 19–40%, with 18–33% completing after the first two reminders. Most caregivers (72–93%) selected a personalized strategy and an average of 28% viewed infographic messages. Between Baseline and Assessment 5, daily SSB intake frequency significantly (p < 0.01) declined for caregivers (−0.32 (0.03), effect size (ES) = 0.51) and children (−0.26 (0.01), ES = 0.53). Effect sizes increased when limited to participants who consumed SSB twice or more per week (caregivers ES = 0.65, children ES = 0.67). Findings indicate that an SMS-delivered intervention is promising for engaging rural caregivers of middle school students and improving SSB behaviors.

Few studies have examined how trait emotion dysregulation relates to momentary affective experiences and the emotion regulation (ER) strategies people use in daily life. In the current study, 112 college students completed a trait measure of emotion dysregulation and completed experience sampling and end-of-day surveys over a two- to three-week period, asking about their emotional experiences and ER strategy use. Participants completed a total of 3798 experience sampling (in-the-moment) and 995 nightly diary surveys. We examined the top 40% of each participant's reported instances of negative affect (to capture times when emotions more likely need regulation). Results indicated that a higher level of trait emotion dysregulation was associated with the following in-the-moment responses: (a) higher level of negative affect; (b) greater desire to change emotions; (c) more attempts to regulate emotion; (d) higher relative endorsements of avoidant (e.g. thought suppression) versus engagement (e.g. acceptance) ER strategy use; and (e) lower perceived effectiveness of ER. Further, individuals with a higher (vs. lower) level of trait emotion dysregulation were less able to identify emotions over the course of the day. Findings demonstrate how trait emotion dysregulation may predict emotional experiences and ER in daily life.

Background Nearly half of newly diagnosed breast cancer patients will report clinically significant symptoms of depression and/or anxiety within the first year of diagnosis. Research on the trajectory of distress in cancer patients suggests that targeting patients early in the diagnostic pathway could be particularly impactful. Given the recent rise of smartphone adoption, apps are a convenient and accessible platform from which to deliver mental health support; however, little research has examined their potential impact among newly diagnosed cancer patients. One reason is likely due to the obstacles associated with in-clinic recruitment of newly diagnosed cancer patients for mHealth pilot studies. Methods This article draws from our experiences of a recently completed pilot study to test a suite of mental health apps in newly diagnosed breast cancer patients. Recruitment strategies included in-clinic pamphlets, flyers, and direct communication with clinicians. Surgical oncologists and research staff members approached eligible patients after a medical appointment. Research team members met with patients to provide informed consent and review the study schedule. Results Four domains of in-clinic recruitment challenges emerged: (a) coordination with clinic staff, (b) perceived burden among breast cancer patients, (c) limitations regarding the adoption and use of technology, and (d) availability of resources. Potential solutions are provided for each challenge. Conclusion Recruitment of newly diagnosed cancer patients is a major challenge to conducting mobile intervention studies for researchers on a pilot-study budget. To realize the impact of mobile interventions for the most vulnerable cancer patient populations, health researchers must address barriers to in-clinic recruitment to provide vital preliminary data in proposals of large-scale research projects.

Objective This study investigates a 6-month short message service (SMS) intervention to reduce adolescent sugar-sweetened beverage (SSB) intake. The objectives are to describe caregiver retention and SMS engagement as well as explore differences by caregiver characteristics. Methods Caregivers completed a baseline survey then messages were sent two times per week. Message types included the following: SSB intake assessments, educational information, infographic URLs, and strategies. Engagement was measured through interaction with these messages and included: assessment completion, reminders needed, number of strategies chosen, and URLs clicked. Results Caregivers (n = 357) had an average baseline SSB intake of 23.9 (SD = 26.8) oz/day. Of those, 89% were retained. Caregivers with a greater income and education were retained at a higher rate. Average engagement included: 4.1 (SD = 1.3) of 5 assessments completed with few reminders needed [4.1 (SD = 3.7) of 14 possible], 3.2 (SD = 1.1) of 4 strategies selected, and 1.2 (SD = 1.6) of 5 URLs clicked. Overall, average engagement was relatively high, even where disparities were found. Demographic characteristics that were statistically related to lower engagement included younger age, lower income, lower educational attainment, single caregivers, lower health literacy. Furthermore, caregivers with a reduced intention to change SSB behaviors completed fewer assessments and needed more reminders. Higher baseline SSB intake was associated with lower engagement across all indicators except URL clicks. Conclusions Results can be used to develop targeted retention and engagement strategies (e.g., just-in-time and/or adaptive interventions) in rural SMS interventions for identified demographic subsets. Trial registration Clincialtrials.gov: NCT03740113.

The way people use smartphones provides a window into the relationship between behaviors and mental health. This relationship is of particular significance to individuals with elevated social anxiety, as it helps to reveal when and where their stress increases in relation to social interactions, ultimately leading to more precise treatment delivery and interventions. In this collaboration between engineers and psychologists, we present the first study to use smartphone sensors to examine socially anxious individuals’ fine-grained behaviors around periods in which they engage in some form of social interaction, and how these behaviors differ as a function of location (e.g., at home, at work, or at an unfamiliar location). In a two-week study of 52 college students, we show that there is a significant difference in behaviors for individuals based on social anxiety levels and locations, in that individuals higher (vs. lower) in social anxiety symptoms exhibit more movement (as tracked by the accelerometer) around the time of phone calls, especially in an unfamiliar location (i.e., not home or at work). Finally, we discuss the implications of these findings for developing better interventions for socially anxious individuals.

OBJECTIVE A common method of distress monitoring in cancer patients relies on static and retrospective data collected in-person at the time of a health care provider appointment. Relatively little work has examined the potential usefulness of mobile distress monitoring using cancer patients' smartphones. The current study deployed longitudinal distress monitoring using secure text messaging. METHODS In an observational study, a total of 52 cancer patients receiving active cancer treatment (Mage = 58, 62% female) received a text message once a week for 4 weeks. Text messages contained a secure link to complete online the Patient Health Questionnaire-4 (PHQ-4), a commonly used distress screener. RESULTS Cancer patients completed a distress screener 75% of the time they received a text message. On average, it took less than a minute to complete each mobile distress screener. Geolocation data indicated that cancer patients completed distress screeners across a range of locations. Analyses of model fit of distress scores indicated significant heterogeneity in variability of distress scores over time and across cancer patients (AIC = 630.5). Quantitative feedback from cancer patients at the end of the study indicated high ease of use, ease of learning, and satisfaction of completing mobile distress screeners. CONCLUSIONS These findings support the use of secure text messaging to monitor longitudinal, out of clinic, distress in cancer patients. Findings also highlight the importance of mobile-based approaches to distress screening in order to maximize opportunities to intervene.

Socially anxious individuals typically select more avoidant emotion regulation (ER) strategies than non-anxious individuals, contributing to interpersonal difficulties. The present study utilized smartphone-delivered experience sampling over 14 days to assess how actual and desired social situations predicted reports of ER strategy use in 115 undergraduate students with varying levels of social anxiety symptoms. After controlling for multiple comparisons, results indicated that higher (vs. lower) baseline social anxiety symptoms predicted endorsing at least one of the available eight ER strategies relatively more often than reporting no strategy use, in the context of high negative affect. We did not find the hypothesized positive relationship between social anxiety symptoms and endorsements of avoidant- (e.g., expressive suppression) versus engagement-oriented (e.g., cognitive reappraisal) ER strategies in the context of high negative affect. However, state social desire interacted with trait social anxiety at high negative affect to predict the use of an ER strategy, although the simple effects analyses at high and low levels of social desire were not statistically reliable. Collectively, our results demonstrate the importance of considering both trait-level social anxiety symptoms and in-the-moment social context when studying ER strategy selection. The importance of assessing intrinsic motivational goals and beliefs in the context of ER strategy use is also discussed.

BackgroundHealth interventions delivered via smart devices are increasingly being used to address mental health challenges associated with cancer treatment. Engagement with mobile interventions has been associated with treatment success; however, the relationship between mood and engagement among patients with cancer remains poorly understood. A reason for this is the lack of a data-driven process for analyzing mood and app engagement data for patients with cancer.ObjectiveThis study aimed to provide a step-by-step process for using app engagement metrics to predict continuously assessed mood outcomes in patients with breast cancer.MethodsWe described the steps involved in data preprocessing, feature extraction, and data modeling and prediction. We applied this process as a case study to data collected from patients with breast cancer who engaged with a mobile mental health app intervention (IntelliCare) over 7 weeks. We compared engagement patterns over time (eg, frequency and days of use) between participants with high and low anxiety and between participants with high and low depression. We then used a linear mixed model to identify significant effects and evaluate the performance of the random forest and XGBoost classifiers in predicting weekly mood from baseline affect and engagement features.ResultsWe observed differences in engagement patterns between the participants with high and low levels of anxiety and depression. The linear mixed model results varied by the feature set; these results revealed weak effects for several features of engagement, including duration-based metrics and frequency. The accuracy of predicting depressed mood varied according to the feature set and classifier. The feature set containing survey features and overall app engagement features achieved the best performance (accuracy: 84.6%; precision: 82.5%; recall: 64.4%; F1 score: 67.8%) when used with a random forest classifier.ConclusionsThe results from the case study support the feasibility and potential of our analytic process for understanding the relationship between app engagement and mood outcomes in patients with breast cancer. The ability to leverage both self-report and engagement features to analyze and predict mood during an intervention could be used to enhance decision-making for researchers and clinicians and assist in developing more personalized interventions for patients with breast cancer.

BACKGROUND Health interventions delivered via smart devices are increasingly being used to address mental health challenges associated with cancer treatment. Engagement with mobile interventions has been associated with treatment success; however, the relationship between mood and engagement among patients with cancer remains poorly understood. A reason for this is the lack of a data-driven process for analyzing mood and app engagement data for patients with cancer. OBJECTIVE This study aimed to provide a step-by-step process for using app engagement metrics to predict continuously assessed mood outcomes in patients with breast cancer. METHODS We described the steps involved in data preprocessing, feature extraction, and data modeling and prediction. We applied this process as a case study to data collected from patients with breast cancer who engaged with a mobile mental health app intervention (IntelliCare) over 7 weeks. We compared engagement patterns over time (eg, frequency and days of use) between participants with high and low anxiety and between participants with high and low depression. We then used a linear mixed model to identify significant effects and evaluate the performance of the random forest and XGBoost classifiers in predicting weekly mood from baseline affect and engagement features. RESULTS We observed differences in engagement patterns between the participants with high and low levels of anxiety and depression. The linear mixed model results varied by the feature set; these results revealed weak effects for several features of engagement, including duration-based metrics and frequency. The accuracy of predicting depressed mood varied according to the feature set and classifier. The feature set containing survey features and overall app engagement features achieved the best performance (accuracy: 84.6%; precision: 82.5%; recall: 64.4%; F1 score: 67.8%) when used with a random forest classifier. CONCLUSIONS The results from the case study support the feasibility and potential of our analytic process for understanding the relationship between app engagement and mood outcomes in patients with breast cancer. The ability to leverage both self-report and engagement features to analyze and predict mood during an intervention could be used to enhance decision-making for researchers and clinicians and assist in developing more personalized interventions for patients with breast cancer. CLINICALTRIAL

Objective Ecological momentary assessment (EMA) may help with the development of more targeted interventions for caregivers' depression, yet the use of this method has been limited among cancer caregivers. This study aimed to demonstrate the feasibility of EMA among cancer caregivers and the use of EMA data to understand affective correlates of caregiver depressive symptoms. Methods Caregivers (N = 25) completed a depressive symptom assessment (Patient Health Questionnaire-8) and then received eight EMA survey prompts per day for 7 days. EMA surveys assessed affect on the orthogonal dimensions of valence and arousal. Participants completed feedback surveys regarding the EMA protocol at the conclusion of the week-long study. Results Of 32 caregivers approached, 25 enrolled and participated (78%), which exceeded the a priori feasibility cutoff of 55%. The prompt completion rate (59%, or 762 of 1,286 issued) did not exceed the a priori cutoff of 65%, although completion was not related to caregivers' age, employment status, physical health quality of life, caregiving stress, or depressive symptoms or the patients' care needs (ps > 0.22). Caregivers' feedback about their study experience was generally positive. Mixed-effects location scale modeling showed caregivers' higher depressive symptoms were related to overall higher reported negative affect and lower positive affect, but not to affective variability. Conclusions Findings from this feasibility study refute potential concerns that an EMA design is too burdensome for distressed caregivers. Clinically, findings suggest the potential importance of not only strategies to reduce overall levels of negative affect, but also to increase opportunities for positive affect.

Cortisol is a glucocorticoid hormone that is critical to immune system functioning. Studies show that prolonged exposure to high levels of cortisol can lead to a range of physical health ailments including the progression of tumor growth. The ability to monitor cortisol levels over time can therefore be used to facilitate decision-making during cancer treatment. However, collecting serum or saliva samples to monitor cortisol in situ is inconvenient, costly, and impractical. In this paper, we propose a general predictive modeling process that uses passively sensed actigraphy data to predict underlying salivary cortisol levels using graph representation learning. We compare machine learning models with handcrafted feature engineering and with graph representation learning, which includes Graph2Vec, FeatherGraph, GeoScattering and NetLSD. Our preliminary results generated from data from 10 newly diagnosed pancreatic cancer patients demonstrate that machine learning models with graph representation learning can outperform the handcrafted feature engineering to predict salivary cortisol levels.

Background: User dropout is a widespread concern in the delivery and evaluation of digital (ie, web and mobile apps) health interventions. Researchers have yet to fully realize the potential of the large amount of data generated by these technology-based programs. Of particular interest is the ability to predict who will drop out of an intervention. This may be possible through the analysis of user journey data—self-reported as well as system-generated data—produced by the path (or journey) an individual takes to navigate through a digital health intervention.Objective: The purpose of this study is to provide a step-by-step process for the analysis of user journey data and eventually to predict dropout in the context of digital health interventions. The process is applied to data from an internet-based intervention for insomnia as a way to illustrate its use. The completion of the program is contingent upon completing 7 sequential cores, which include an initial tutorial core. Dropout is defined as not completing the seventh core.Methods: Steps of user journey analysis, including data transformation, feature engineering, and statistical model analysis and evaluation, are presented. Dropouts were predicted based on data from 151 participants from a fully automated web-based program (Sleep Healthy Using the Internet) that delivers cognitive behavioral therapy for insomnia. Logistic regression with L1 and L2 regularization, support vector machines, and boosted decision trees were used and evaluated based on their predictive performance. Relevant features from the data are reported that predict user dropout.Results: Accuracy of predicting dropout (area under the curve [AUC] values) varied depending on the program core and the machine learning technique. After model evaluation, boosted decision trees achieved AUC values ranging between 0.6 and 0.9. Additional handcrafted features, including time to complete certain steps of the intervention, time to get out of bed, and days since the last interaction with the system, contributed to the prediction performance.Conclusions: The results support the feasibility and potential of analyzing user journey data to predict dropout. Theory-driven handcrafted features increased the prediction performance. The ability to predict dropout at an individual level could be used to enhance decision making for researchers and clinicians as well as inform dynamic intervention regimens. Background: User dropout is a widespread concern in the delivery and evaluation of digital (ie, web and mobile apps) health interventions. Researchers have yet to fully realize the potential of the large amount of data generated by these technology-based programs. Of particular interest is the ability to predict who will drop out of an intervention. This may be possible through the analysis of user journey data—self-reported as well as system-generated data—produced by the path (or journey) an individual takes to navigate through a digital health intervention.Objective: The purpose of this study is to provide a step-by-step process for the analysis of user journey data and eventually to predict dropout in the context of digital health interventions. The process is applied to data from an internet-based intervention for insomnia as a way to illustrate its use. The completion of the program is contingent upon completing 7 sequential cores, which include an initial tutorial core. Dropout is defined as not completing the seventh core.Methods: Steps of user journey analysis, including data transformation, feature engineering, and statistical model analysis and evaluation, are presented. Dropouts were predicted based on data from 151 participants from a fully automated web-based program (Sleep Healthy Using the Internet) that delivers cognitive behavioral therapy for insomnia. Logistic regression with L1 and L2 regularization, support vector machines, and boosted decision trees were used and evaluated based on their predictive performance. Relevant features from the data are reported that predict user dropout.Results: Accuracy of predicting dropout (area under the curve [AUC] values) varied depending on the program core and the machine learning technique. After model evaluation, boosted decision trees achieved AUC values ranging between 0.6 and 0.9. Additional handcrafted features, including time to complete certain steps of the intervention, time to get out of bed, and days since the last interaction with the system, contributed to the prediction performance.Conclusions: The results support the feasibility and potential of analyzing user journey data to predict dropout. Theory-driven handcrafted features increased the prediction performance. The ability to predict dropout at an individual level could be used to enhance decision making for researchers and clinicians as well as inform dynamic intervention regimens.

Background Although gamification continues to be a popular approach to increase engagement, motivation, and adherence to behavioral interventions, empirical studies have rarely focused on this topic. There is a need to empirically evaluate gamification models to increase the understanding of how to integrate gamification into interventions. Objective The model of gamification principles for digital health interventions proposes a set of five independent yet interrelated gamification principles. This study aimed to examine the validity and reliability of this model to inform its use in Web- and mobile-based apps. Methods A total of 17 digital health interventions were selected from a curated website of mobile- and Web-based apps (PsyberGuide), which makes independent and unbiased ratings on various metrics. A total of 133 independent raters trained in gamification evaluation techniques were instructed to evaluate the apps and rate the degree to which gamification principles are present. Multiple ratings (n≥20) were collected for each of the five gamification principles within each app. Existing measures, including the PsyberGuide credibility score, mobile app rating scale (MARS), and the app store rating of each app were collected, and their relationship with the gamification principle scores was investigated. Results Apps varied widely in the degree of gamification implemented (ie, the mean gamification rating ranged from 0.17≤m≤4.65 out of 5). Inter-rater reliability of gamification scores for each app was acceptable (κ≥0.5). There was no significant correlation between any of the five gamification principles and the PsyberGuide credibility score (P≥.49 in all cases). Three gamification principles (supporting player archetypes, feedback, and visibility) were significantly correlated with the MARS score, whereas three principles (meaningful purpose, meaningful choice, and supporting player archetypes) were significantly correlated with the app store rating. One gamification principle was statistically significant with both the MARS and the app store rating (supporting player archetypes). Conclusions Overall, the results support the validity and potential utility of the model of gamification principles for digital health interventions. As expected, there was some overlap between several gamification principles and existing app measures (eg, MARS). However, the results indicate that the gamification principles are not redundant with existing measures and highlight the potential utility of a 5-factor gamification model structure in digital behavioral health interventions. These gamification principles may be used to improve user experience and enhance engagement with digital health programs.

Background Nearly half of the patients with breast cancer experience clinically significant mental distress within the first year of receiving their cancer diagnosis. There is an urgent need to identify scalable and cost-efficient ways of delivering empirically supported mental health interventions to patients with breast cancer. Objective The aim of this study was to evaluate the feasibility of in-clinic recruitment for a mobile phone app study and to evaluate the usability and preliminary impact of a suite of mental health apps (IntelliCare) with phone coaching on psychosocial distress symptoms in patients recently diagnosed with breast cancer. Methods This pilot study adopted a within-subject, 7-week pre-post study design. A total of 40 patients with breast cancer were recruited at a US National Cancer Institute–designated clinical cancer center. Self-reported distress (Patient Health Questionnaire-4) and mood symptoms (Patient-Reported Outcomes Measurement Information System depression and anxiety scales) were assessed at baseline and postintervention. App usability was assessed at postintervention. Results The minimum recruitment threshold was met. There was a significant decrease in general distress symptoms, as well as symptoms of depression and anxiety, from baseline to postintervention. Overall, participants reported high levels of ease of app use and learning. Scores for app usefulness and satisfaction were reinforced by some qualitative feedback suggesting that tailoring the apps more for patients with breast cancer could enhance engagement. Conclusions There is a dire need for scalable, supportive interventions in cancer. The results from this study inform how scalable mobile phone–delivered programs with additional phone support can be used to support patients with breast cancer. International Registered Report Identifier (IRRID) RR2-10.2196/11452

Background Behavioral health researchers are increasingly recognizing the potential of mobile phone apps to deliver empirically supported treatments. However, current options for developing apps typically require large amounts of expertise or money. Objective This paper aims to describe a pragmatic do-it-yourself approach for researchers to create and pilot an Android mobile phone app using existing survey software (eg, Qualtrics survey platform). Methods This study was conducted at an academic research center in the United States focused on developing and evaluating behavioral health technologies. The process outlined in this paper was derived and condensed from the steps to building an existing app intervention, iCanThrive, which was developed to enhance mental well-being in women cancer survivors. Results This paper describes an inexpensive, practical process that uses a widely available survey software, such as Qualtrics, to create and pilot a mobile phone intervention that is presented to participants as a Web viewer app that is downloaded from the Google Play store. Health researchers who are interested in using this process to pilot apps are encouraged to inquire about the survey platforms available to them, the level of security those survey platforms provide, and the regulatory guidelines set forth by their institution. Conclusions As app interventions continue to gain interest among researchers and consumers alike, it is important to find new ways to efficiently develop and pilot app interventions before committing a large amount of resources. Mobile phone app interventions are an important component to discovering new ways to reach and support individuals with behavioral or mental health disorders.

Background Psychological distress is a major issue among survivors of women’s cancer who face numerous barriers to accessing in-person mental health treatments. Mobile phone app–based interventions are scalable and have the potential to increase access to mental health care among survivors of women’s cancer worldwide. Objective This study aimed to evaluate the acceptability and preliminary efficacy of a novel app-based intervention with phone coaching in a sample of survivors of women’s cancer. Methods In a single-group, pre-post, 6-week pilot study in the United States, 28 survivors of women’s cancer used iCanThrive, a novel app intervention that teaches skills for coping with stress and enhancing well-being, with added phone coaching. The primary outcome was self-reported symptoms of depression (Center for Epidemiologic Studies Depression Scale). Emotional self-efficacy and sleep disruption were also assessed at baseline, 6-week postintervention, and 4 weeks after the intervention period. Feedback obtained at the end of the study focused on user experience of the intervention. Results There were significant decreases in symptoms of depression and sleep disruption from baseline to postintervention. Sleep disruption remained significantly lower at 4-week postintervention compared with baseline. The iCanThrive app was launched a median of 20.5 times over the intervention period. The median length of use was 2.1 min. Of the individuals who initiated the intervention, 87% (20/23) completed the 6-week intervention. Conclusions This pilot study provides support for the acceptability and preliminary efficacy of the iCanThrive intervention. Future work should validate the intervention in a larger randomized controlled study. It is important to develop scalable interventions that meet the psychosocial needs of different cancer populations. The modular structure of the iCanThrive app and phone coaching could impact a large population of survivors of women’s cancer.

BACKGROUND User dropout is a widespread concern in the delivery and evaluation of digital (ie, web and mobile apps) health interventions. Researchers have yet to fully realize the potential of the large amount of data generated by these technology-based programs. Of particular interest is the ability to predict who will drop out of an intervention. This may be possible through the analysis of user journey data—self-reported as well as system-generated data—produced by the path (or journey) an individual takes to navigate through a digital health intervention. OBJECTIVE The purpose of this study is to provide a step-by-step process for the analysis of user journey data and eventually to predict dropout in the context of digital health interventions. The process is applied to data from an internet-based intervention for insomnia as a way to illustrate its use. The completion of the program is contingent upon completing 7 sequential cores, which include an initial tutorial core. Dropout is defined as not completing the seventh core. METHODS Steps of user journey analysis, including data transformation, feature engineering, and statistical model analysis and evaluation, are presented. Dropouts were predicted based on data from 151 participants from a fully automated web-based program (Sleep Healthy Using the Internet) that delivers cognitive behavioral therapy for insomnia. Logistic regression with L1 and L2 regularization, support vector machines, and boosted decision trees were used and evaluated based on their predictive performance. Relevant features from the data are reported that predict user dropout. RESULTS Accuracy of predicting dropout (area under the curve [AUC] values) varied depending on the program core and the machine learning technique. After model evaluation, boosted decision trees achieved AUC values ranging between 0.6 and 0.9. Additional handcrafted features, including time to complete certain steps of the intervention, time to get out of bed, and days since the last interaction with the system, contributed to the prediction performance. CONCLUSIONS The results support the feasibility and potential of analyzing user journey data to predict dropout. Theory-driven handcrafted features increased the prediction performance. The ability to predict dropout at an individual level could be used to enhance decision making for researchers and clinicians as well as inform dynamic intervention regimens.

Mental health problems are a leading cause of disease burden and disability worldwide. Social anxiety and depression are highly prevalent among college students. The current methods for detecting symptoms are based on client self-report via questionnaires and interviews in traditional clinical settings, but self-report is subject to recall bias and visiting a clinic requires a high level of motivation. Assessment methods that use both actively and passively collected data hold promise for detecting and monitoring social anxiety and depression symptoms as individuals go about their daily lives. The present study, named DemonicSalmon, investigates how social anxiety and depression symptoms manifest in the daily life of 72 students over a two-week study period. Results show a number of significant correlations between the automatic objective sensor data from smartphones and indicators of mental health. The collected data enhances understanding of how students’ social anxiety, depression and affect levels are associated with their mobility, activity levels, and communication patterns. The DemonicSalmon dataset has been made publicly available on the web to enhance collaborations in this important area of research.

Objective The aim of this study was to investigate in a randomized clinical trial the role of sleep-related cognitive variables in the long-term efficacy of an online, fully automated cognitive behavioral therapy intervention for insomnia (CBT-I) (Sleep Healthy Using the Internet [SHUTi]). Method Three hundred and three participants (Mage = 43.3 years; SD = 11.6) were randomly assigned to SHUTi or an online patient education condition and assessed at baseline, postintervention (nine weeks after baseline), and six and 12 months after the intervention period. Cognitive variables were self-reported internal and chance sleep locus of control, dysfunctional beliefs and attitudes about sleep (DBAS), sleep specific self-efficacy, and insomnia knowledge. Primary outcomes were self-reported online ratings of insomnia severity (Insomnia Severity Index), and sleep onset latency and wake after sleep onset from online sleep diaries, collected 12 months after the intervention period. Results Those who received SHUTi had, at postassessment, higher levels of insomnia knowledge (95% confidence interval [CI] = 0.10–0.16) and internal sleep locus of control (95% CI = 0.04–0.55) as well as lower DBAS (95% CI = 1.52–2.39) and sleep locus of control attributed to chance (95% CI = 0.15–0.71). Insomnia knowledge, chance sleep locus of control, and DBAS mediated the relationship between condition and at least one 12-month postassessment sleep outcome. Within the SHUTi condition, changes in each cognitive variable (with the exception of internal sleep locus of control) predicted improvement in at least one sleep outcome one year later. Conclusion Online CBT-I may reduce the enormous public health burden of insomnia by changing underlying cognitive variables that lead to long-term changes in sleep outcomes.

The present research examined the degree to which facets of trait mindfulness were associated with level and changes in psychological distress in response to a repeated carbon dioxide (CO2) breathing challenge. Undergraduate students (N = 93) completed a self-report measure of mindfulness and underwent two 7.5% CO2 challenges, spaced 1 week apart. Subjective distress, physical/fear symptoms, and threat cognitions were assessed at multiple times throughout each administration. A pattern emerged such that although mindfulness facets were not reliably associated with distress at either administration separately, a low (but not high) level of mindfulness was associated with a significant decrease in distress across administrations, likely indicative of habituation, for the facets Describing (β = − 0.25, p

Abstract. The present research presents evidence of the validity and reliability of an abbreviated version of the Perceived Affect Utility Scale (PAUSe). In Study 1, findings using item response theory (IRT) led to the deletion of one-third of the number of items in the PAUSe. Results from confirmatory factor analysis (CFA) supported the structure of the abbreviated version of the PAUSe, the PAUSe-r. Examining effect sizes between the PAUSe-r and instruments measuring personality, emotion, and depression also supported the convergent validity, discriminant validity, and incremental validity of this measure, even after taking into account ideal affect and the actual experience of emotion. In Study 2, test-retest reliability of the PAUSe-r in a sample of college students is presented. In Study 3, using data from a large, nonstudent sample, we replicated the structure of the PAUSe-r, as well as relations between the PAUSe-r and personality variables, that were found in Study 1.

The current meta-analysis investigated the extent to which personality traits changed as a result of intervention, with the primary focus on clinical interventions. We identified 207 studies that had tracked changes in measures of personality traits during interventions, including true experiments and prepost change designs. Interventions were associated with marked changes in personality trait measures over an average time of 24 weeks (e.g., d = .37). Additional analyses showed that the increases replicated across experimental and nonexperimental designs, for nonclinical interventions, and persisted in longitudinal follow-ups of samples beyond the course of intervention. Emotional stability was the primary trait domain showing changes as a result of therapy, followed by extraversion. The type of therapy employed was not strongly associated with the amount of change in personality traits. Patients presenting with anxiety disorders changed the most, and patients being treated for substance use changed the least. The relevance of the results for theory and social policy are discussed. (PsycINFO Database Record

Background Online interventions for insomnia can increase access to treatments for those with limited access to services. What remains unknown is whether individuals from more isolated (vs. more densely populated) regions engage with, and benefit as much from, an online intervention. This secondary analysis examined the relationship of geographical indices with engagement and outcomes of an efficacious, fully automated online cognitive behavioral therapy for insomnia (CBT—I) program (Sleep Healthy Using the Internet-SHUTi). Method 303 participants (Mage = 43.3; SD = 11.6) were randomly assigned to SHUTi or an online patient education condition and assessed at baseline and post intervention. Rural code of participants was determined using participant zip codes. Distance to the nearest sleep medicine provider was calculated as the distance between the center of the nearest provider's city (from a publicly available list of CBT-I providers) and the center of the participants' zip code. Adherence outcomes were number of intervention core completions, sleep diaries, and logins. Sleep outcomes were insomnia severity as well as sleep onset latency and wake after sleep onset derived from online sleep diaries. Results Individuals were from a range of geographic locations. Most lived in fairly densely populated areas; however, there was a large variation in distance to the nearest sleep medicine provider. Findings indicate that the efficacy, adherence, and engagement with SHUTi were not impacted by where people lived. Controlling for age and gender did not impact any of the relationships among geography variables (i.e., distance, ruralness) and adherence or sleep related outcomes. Conclusions Internet interventions must demonstrate that they can overcome obstacles posed by geography. This is the first study to examine the geographic location of participants and its association with engagement with, and outcomes of, online CBT-I., Highlights • Data from a large randomized trial of an automated online CBT-I program (SHUTi) • Study participants had insomnia and were from a range of geographic locations. • Examined the impact of participant geography indices on engagement and outcomes • SHUTi efficacy, adherence, and engagement was not impacted by where people lived.

BACKGROUND Although gamification continues to be a popular approach to increase engagement, motivation, and adherence to behavioral interventions, empirical studies have rarely focused on this topic. There is a need to empirically evaluate gamification models to increase the understanding of how to integrate gamification into interventions. OBJECTIVE The model of gamification principles for digital health interventions proposes a set of five independent yet interrelated gamification principles. This study aimed to examine the validity and reliability of this model to inform its use in Web- and mobile-based apps. METHODS A total of 17 digital health interventions were selected from a curated website of mobile- and Web-based apps (PsyberGuide), which makes independent and unbiased ratings on various metrics. A total of 133 independent raters trained in gamification evaluation techniques were instructed to evaluate the apps and rate the degree to which gamification principles are present. Multiple ratings (n≥20) were collected for each of the five gamification principles within each app. Existing measures, including the PsyberGuide credibility score, mobile app rating scale (MARS), and the app store rating of each app were collected, and their relationship with the gamification principle scores was investigated. RESULTS Apps varied widely in the degree of gamification implemented (ie, the mean gamification rating ranged from 0.17≤m≤4.65 out of 5). Inter-rater reliability of gamification scores for each app was acceptable (κ≥0.5). There was no significant correlation between any of the five gamification principles and the PsyberGuide credibility score (P≥.49 in all cases). Three gamification principles (supporting player archetypes, feedback, and visibility) were significantly correlated with the MARS score, whereas three principles (meaningful purpose, meaningful choice, and supporting player archetypes) were significantly correlated with the app store rating. One gamification principle was statistically significant with both the MARS and the app store rating (supporting player archetypes). CONCLUSIONS Overall, the results support the validity and potential utility of the model of gamification principles for digital health interventions. As expected, there was some overlap between several gamification principles and existing app measures (eg, MARS). However, the results indicate that the gamification principles are not redundant with existing measures and highlight the potential utility of a 5-factor gamification model structure in digital behavioral health interventions. These gamification principles may be used to improve user experience and enhance engagement with digital health programs.

BACKGROUND Nearly half of the patients with breast cancer experience clinically significant mental distress within the first year of receiving their cancer diagnosis. There is an urgent need to identify scalable and cost-efficient ways of delivering empirically supported mental health interventions to patients with breast cancer. OBJECTIVE The aim of this study was to evaluate the feasibility of in-clinic recruitment for a mobile phone app study and to evaluate the usability and preliminary impact of a suite of mental health apps (IntelliCare) with phone coaching on psychosocial distress symptoms in patients recently diagnosed with breast cancer. METHODS This pilot study adopted a within-subject, 7-week pre-post study design. A total of 40 patients with breast cancer were recruited at a US National Cancer Institute–designated clinical cancer center. Self-reported distress (Patient Health Questionnaire-4) and mood symptoms (Patient-Reported Outcomes Measurement Information System depression and anxiety scales) were assessed at baseline and postintervention. App usability was assessed at postintervention. RESULTS The minimum recruitment threshold was met. There was a significant decrease in general distress symptoms, as well as symptoms of depression and anxiety, from baseline to postintervention. Overall, participants reported high levels of ease of app use and learning. Scores for app usefulness and satisfaction were reinforced by some qualitative feedback suggesting that tailoring the apps more for patients with breast cancer could enhance engagement. CONCLUSIONS There is a dire need for scalable, supportive interventions in cancer. The results from this study inform how scalable mobile phone–delivered programs with additional phone support can be used to support patients with breast cancer. INTERNATIONAL REGISTERED REPORT RR2-10.2196/11452

Introduction:. Annually, over 250,000 women are diagnosed with breast cancer with over one-third undergoing mastectomy and contemplating reconstruction. Surgical breast reconstructive options vary in post-operative recovery, yet with a paucity of objective data to inform women of their expected recovery after flap or implant-based reconstruction. As a result, patient decision-making is based primarily on surgeon preference and subjective data regarding perceived invasiveness of surgical options. This study aims to identify recovery outcomes of interest to breast cancer patients and to determine the feasibility of objectively measuring patient recovery after mastectomy and reconstruction using patient-worn actigraphy devices. Methods:. Three survivorship focus groups for patients after mastectomy with and without reconstruction were used to identify recovery outcomes they considered relevant. Cloud storage systems and actigraphy devices were piloted to determine performance. Actigraphy devices were worn by patients peri-operatively to measure post-operative sleep quality and steps taken, normalized to individual patient pre-operative control data. Results:. Focus groups identified sleep quality, return to activity (measurable with actigraphy), and driving as variables impacting surgical decision-making. We prospectively measured outcomes for four women undergoing immediate pre-pectoral tissue expander placement and four women undergoing immediate free flap reconstruction. Actigraphy data demonstrated an initial decrease in activity, increase in sleep variability and increased heart rate that approached the patients’ pre-operative normalized data as they recovered over time. Conclusions:. These data demonstrate that actigraphy data would be of interest to patients making breast reconstruction decisions and that the data can be successfully collected to inform decision-making.

BACKGROUND Psychological distress is a major issue among survivors of women’s cancer who face numerous barriers to accessing in-person mental health treatments. Mobile phone app–based interventions are scalable and have the potential to increase access to mental health care among survivors of women’s cancer worldwide. OBJECTIVE This study aimed to evaluate the acceptability and preliminary efficacy of a novel app-based intervention with phone coaching in a sample of survivors of women’s cancer. METHODS In a single-group, pre-post, 6-week pilot study in the United States, 28 survivors of women’s cancer used iCanThrive, a novel app intervention that teaches skills for coping with stress and enhancing well-being, with added phone coaching. The primary outcome was self-reported symptoms of depression (Center for Epidemiologic Studies Depression Scale). Emotional self-efficacy and sleep disruption were also assessed at baseline, 6-week postintervention, and 4 weeks after the intervention period. Feedback obtained at the end of the study focused on user experience of the intervention. RESULTS There were significant decreases in symptoms of depression and sleep disruption from baseline to postintervention. Sleep disruption remained significantly lower at 4-week postintervention compared with baseline. The iCanThrive app was launched a median of 20.5 times over the intervention period. The median length of use was 2.1 min. Of the individuals who initiated the intervention, 87% (20/23) completed the 6-week intervention. CONCLUSIONS This pilot study provides support for the acceptability and preliminary efficacy of the iCanThrive intervention. Future work should validate the intervention in a larger randomized controlled study. It is important to develop scalable interventions that meet the psychosocial needs of different cancer populations. The modular structure of the iCanThrive app and phone coaching could impact a large population of survivors of women’s cancer.

BACKGROUND Behavioral health researchers are increasingly recognizing the potential of mobile phone apps to deliver empirically supported treatments. However, current options for developing apps typically require large amounts of expertise or money. OBJECTIVE This paper aims to describe a pragmatic do-it-yourself approach for researchers to create and pilot an Android mobile phone app using existing survey software (eg, Qualtrics survey platform). METHODS This study was conducted at an academic research center in the United States focused on developing and evaluating behavioral health technologies. The process outlined in this paper was derived and condensed from the steps to building an existing app intervention, iCanThrive, which was developed to enhance mental well-being in women cancer survivors. RESULTS This paper describes an inexpensive, practical process that uses a widely available survey software, such as Qualtrics, to create and pilot a mobile phone intervention that is presented to participants as a Web viewer app that is downloaded from the Google Play store. Health researchers who are interested in using this process to pilot apps are encouraged to inquire about the survey platforms available to them, the level of security those survey platforms provide, and the regulatory guidelines set forth by their institution. CONCLUSIONS As app interventions continue to gain interest among researchers and consumers alike, it is important to find new ways to efficiently develop and pilot app interventions before committing a large amount of resources. Mobile phone app interventions are an important component to discovering new ways to reach and support individuals with behavioral or mental health disorders.

Cognitive-behavioral therapy for insomnia (CBT-I) shows treatment benefits among individuals with pain interference; however, effects of Internet-delivered CBT-I for this population are unknown. This secondary analysis used randomized clinical trial data from adults assigned to Internet-delivered CBT-I to compare changes in sleep by pre-intervention pain interference. Participants (N = 151) completed the Insomnia Severity Index (ISI) and sleep diaries [sleep onset latency (SOL); wake after sleep onset (WASO)] at baseline, post-assessment, 6- and 12-month follow-ups. Linear mixed-effects models showed no differences between pain interference groups (no, some, moderate/severe) for changes from baseline to any follow-up timepoint for ISI (p = .72) or WASO (p = .88). There was a small difference in SOL between those reporting some versus no or moderate/severe pain interference (p = .04). Predominantly comparable and sustained treatment benefits for both those with and without pain interference suggest that Internet-delivered CBT-I is promising for delivering accessible care to individuals with comorbid pain and insomnia.

Gamification is a popular method used to add entertaining and appealing dimensions to nongaming activities. Researchers of technology-based behavioral and mental-health-focused interventions have shown considerable interest in gamification to enhance engagement and adherence. There have been a number of gamification frameworks proposed, each with differences in focus but with overlapping similarities. A review of these frameworks highlight critical issues in gamification-lack of clear definitions, standards, and a need for an overarching model for applying gamification, rather than simply describing gamification. These issues leave researchers challenged to apply gamification to its full potential. This paper explores gamification as a construct and endeavors to define its core features. A useful way of evaluating the potential utility of gamification features in the context of an intervention is by distinguishing between exogenous applications of gamification (layering game mechanics externally upon a system) and endogenous application of gamification (developing mechanics intrinsic to the given experience). By then comparing and contrasting six gamification frameworks, components are identified that lay at the intersection and a theoretical model is proposed. A theory-driven set of gamification principles, organized into four categories, is developed and presented. Of particular interest is the utilization of this model as it relates to behavioral and mental-health-focused Internet-based interventions. To demonstrate the potential of this gamification framework, the generated principles are overlaid onto the established Model for Internet Interventions, extending it, and providing a more concrete foundation for researchers of Internet interventions. The presented model will assist researchers and developers who are interested in applying gamification to Internet interventions.

Objectives: Despite widespread agreement that personality traits change across the lifespan into older adulthood, the association between changes in personality and depression among older adult cancer survivors is unknown. It was hypothesized that older adults with (vs. without) a past cancer diagnosis would experience an increase in neuroticism, and decreases in conscientiousness, agreeableness, openness, and extraversion, and that changes in these traits would mediate the relationship between receiving a cancer diagnosis and change in depression. Two hypotheses were tested in a cancer survivor sample. First, that increased chronic stressors and decreased physical health would mediate the link between personality change and increased depression. Second, that personality change would mediate the link between changes in chronic stressors/health and increased depression.Method: Secondary data analysis utilizing three waves of data from the Health and Retirement Study. Data was compiled from 5,217 participants, among whom 707 received a cancer diagnosis.Results: Older adults with (vs. without) a cancer diagnosis decreased in conscientiousness, which was associated with increased depression. Among cancer survivors, worsening chronic stressors/health mediated many pathways between personality change and an increased depression. Increased neuroticism mediated the link between worsening health/chronic stressors and increased depression.Conclusion: With the exception of conscientiousness, changes in personality did not mediate the link between cancer survivor status and depression. Among older adult cancer survivors, changes in personality traits may increase depression through worsening physical health and chronic stressors, potentially informing targeted interventions. Interventions that target increased neuroticism may be particularly useful in older adult cancer survivors.

Introduction: The current study investigated whether high and low socially-anxious individuals would show differences in affective forecasting accuracy (i.e., the prediction of emotional states in response to future events) to positive versus negative social evaluation. Method: High (n = 94) and low (n = 98) socially-anxious participants gave a speech and were randomly assigned to receive a positive or negative evaluation. Results: For affective forecasts made proximally (moments before the speech), those low in social anxiety overpredicted their affect to a greater extent to a negative evaluation versus a positive evaluation. In contrast, those high in social anxiety overpredicted their affect to positive and negative evaluations comparably, and failed to adjust their prediction for a future hypothetical negative evaluation—in effect, not learning from their prior forecasting error. Discussion: Results suggest that affective forecasting biases deserve further study as a maintaining factor for social anxiety symptoms.

In three independent samples of undergraduate students, the present research examined the relations between the Big Five personality dimensions and five pleasurable emotions: tranquility, contentment, interest, cheerfulness, and vigor (only three of which, contentment, interest, and cheerfulness, had been examined in past research). Personality was measured using self-report in Studies 1 and 3, and using peer-report in Study 2. Extraversion was strongly associated with cheerfulness and vigor, openness to experience was associated with interest, and neuroticism was negatively associated with most of the pleasurable emotions. Contentment, but not tranquility, was consistently associated with conscientiousness and extraversion. Study 3 also examined the types of activities that people reported engaging in to obtain pleasure. There was some evidence of the types of activities listed being associated with personality, especially extraversion. However, individual differences in the nature of the pleasure-eliciting activities people reported could not account for the associations between personality and the experience of different pleasurable emotions.

Introduction Intervention dropout is an important factor for the evaluation and implementation of digital therapeutics, including in insomnia. Large amounts of individualized data (logins, questionnaires, EMA data) in these interventions can combine to create user journeys - the data generated by the path an individual takes to navigate the digital therapeutic. User journeys can provide insight about how likely users are to drop out of an intervention on an individual level and lead to increased prediction performance. Thus, the goal of this study is to provide a step-by-step guide for the analysis of user journeys and utilize this guide to predict intervention dropout, illustrated with an example from a data in a RCT of digital therapeutic for chronic insomnia, for which outcomes have previously been published. Methods Analysis of user journeys includes data transformation, feature engineering, and statistical model analysis, using machine learning techniques. A framework is established to leverage user journeys to predict various behaviors. For this study, the framework was applied to predict dropouts of 151 participants from a fully automated web-based program (SHUTi) that delivered cognitive behavioral therapy for insomnia. For this task, support vector machines, logistic regression with regularization, and boosted decision trees were applied at different points in 9-week intervention. These techniques were evaluated based on their predictive performance. Results After model evaluation, a decision tree ensemble achieved AUC values ranging between 0.6-0.9 based on application of machine earning techniques. Various handcrafted and theory-driven features (e.g., time to complete certain intervention steps, time to get out of bed after arising, and days since last system interaction contributed to prediction performance. Conclusion Results indicate that utilizing a user journey framework and analysis can predict intervention dropout. Further, handcrafted theory-driven features can increase prediction performance. This prediction of dropout could lead to an enhanced clinical decision-making in digital therapeutics. Support The original study evaluating the efficacy of this intervention has been reported elsewhere and was funded by grant R01 MH86758 from the National Institute of Mental Health.

Despite the proliferation of mobile interventions that are delivered through smartphones, and their potential impact on promoting mental health in chronic health populations, little research has examined how to leverage the advanced technological capabilities of smartphones to also monitor response to app-based interventions. Current methods used to evaluate response to digital interventions rely on static and retrospective self-report measures to infer crucial behavioral and affective patterns, which provide little contextual information and limits generalizability. Importantly, this approach can also increase burden among chronic health populations that are already experiencing a heavy treatment load. Using data collected from 7 recently diagnosed breast cancer patients undergoing active cancer treatment (average time from initial diagnosis to study enrollment is 13 days), we propose a framework for integrating self-report surveys and fine-grained mobile sensing data. Preliminary results demonstrate the feasibility and utility of this framework to reduce burden and improve detection of mood in response to an app-based intervention among chronic health populations.

Momentary experiences of positive and negative emotionality--- also referred to as state affect---are core components of well-being and performance. The ability to unobtrusively monitor state affect could raise individuals' awareness of their mental health status and enable healthcare providers to deliver targeted, just-in-time mental health interventions. In this work, we investigate whether passively sensed smartphone data can be used to recognize individuals' state affect. Our exploratory analysis uses data generated from 220 participants in a two-week study, and our results indicate that fluctuations in participants' negative state affect are associated with various aspects of context including current physical state, geographic location, and time. We also test algorithms that predict participants' negative state affect given contextual features, comparing the impact of historical contextual features on prediction performance.

BACKGROUND Over one-third of cancer patients experience clinically significant mental distress, and distress in caregivers can exceed that of the cancer patients for whom they care. There is an urgent need to identify scalable and cost-efficient ways of delivering mental health interventions to cancer patients and their loved ones. OBJECTIVE The aim of this study is to describe the protocol to pilot a mobile app–based mental health intervention in breast cancer patients and caregivers. METHODS The IntelliCare mental health apps are grounded in evidence-based research in psychology. They have not been examined in cancer populations. This pilot study will adopt a within-subject, pre-post study design to inform a potential phase III randomized controlled trial. A target sample of 50 individuals (with roughly equal numbers of patients and caregivers) at least 18 years of age and fluent in English will be recruited at a US National Cancer Institute designated clinical cancer center. Consent will be obtained in writing and a mobile phone will be provided if needed. Self-report surveys assessing mental health outcomes will be administered at a baseline session and after a 7-week intervention. Before using the apps, participants will receive a 30-min coaching call to explain their purpose and function. A 10-min coaching call 3 weeks later will check on user progress and address questions or barriers to use. Self-report and semistructured interviews with participants at the end of the study period will focus on user experience and suggestions for improving the apps and coaching in future studies. RESULTS This study is ongoing, and recruitment will be completed by the end of 2018. CONCLUSIONS Results from this study will inform how scalable mobile phone-delivered programs can be used to support breast cancer patients and their loved ones. CLINICALTRIAL ClinicalTrials.gov NCT03488745; https://clinicaltrials.gov/ct2/show/NCT03488745 INTERNATIONAL REGISTERED REPOR DERR1-10.2196/11452