Predictive modeling of addiction lapses in a mobile health application

Predicting high-risk periods for weight regain following initial weight loss

OBJECTIVE

The aim of this study was to develop a predictive algorithm of "high-risk" periods for weight regain after weight loss.

METHODS

Longitudinal mixed-effects models and random forest regression were used to select predictors and develop an algorithm to predict weight regain on a week-to-week basis, using weekly questionnaire and self-monitoring data (including daily e-scale data) collected over 40 weeks from 46 adults who lost ≥5% of baseline weight during an initial 12-week intervention (Study 1). The algorithm was evaluated in 22 adults who completed the same Study 1 intervention but lost <5% of baseline weight and in 30 adults recruited for a separate 30-week study (Study 2).

RESULTS

The final algorithm retained the frequency of self-monitoring caloric intake and weight plus self-report ratings of hunger and the importance of weight-management goals compared with competing life demands. In the initial training data set, the algorithm predicted weight regain the following week with a sensitivity of 75.6% and a specificity of 45.8%; performance was similar (sensitivity: 81%-82%, specificity: 30%-33%) in testing data sets.

CONCLUSIONS

Weight regain can be predicted on a proximal, week-to-week level. Future work should investigate the clinical utility of adaptive interventions for weight-loss maintenance and develop more sophisticated predictive models of weight regain.

Acceptability of Personal Sensing Among People With Alcohol Use Disorder: Observational Study

BACKGROUND

Personal sensing may improve digital therapeutics for mental health care by facilitating early screening, symptom monitoring, risk prediction, and personalized adaptive interventions. However, further development and the use of personal sensing requires a better understanding of its acceptability to people targeted for these applications.

OBJECTIVE

We aimed to assess the acceptability of active and passive personal sensing methods in a sample of people with moderate to severe alcohol use disorder using both behavioral and self-report measures. This sample was recruited as part of a larger grant-funded project to develop a machine learning algorithm to predict lapses.

METHODS

Participants (N=154; n=77, 50% female; mean age 41, SD 11.9 years; n=134, 87% White and n=150, 97% non-Hispanic) in early recovery (1-8 weeks of abstinence) were recruited to participate in a 3-month longitudinal study. Participants were modestly compensated for engaging with active (eg, ecological momentary assessment [EMA], audio check-in, and sleep quality) and passive (eg, geolocation, cellular communication logs, and SMS text message content) sensing methods that were selected to tap into constructs from the Relapse Prevention model by Marlatt. We assessed 3 behavioral indicators of acceptability: participants' choices about their participation in the study at various stages in the procedure, their choice to opt in to provide data for each sensing method, and their adherence to a subset of the active methods (EMA and audio check-in). We also assessed 3 self-report measures of acceptability (interference, dislike, and willingness to use for 1 year) for each method.

RESULTS

Of the 192 eligible individuals screened, 191 consented to personal sensing. Most of these individuals (169/191, 88.5%) also returned 1 week later to formally enroll, and 154 participated through the first month follow-up visit. All participants in our analysis sample opted in to provide data for EMA, sleep quality, geolocation, and cellular communication logs. Out of 154 participants, 1 (0.6%) did not provide SMS text message content and 3 (1.9%) did not provide any audio check-ins. The average adherence rate for the 4 times daily EMA was .80. The adherence rate for the daily audio check-in was .54. Aggregate participant ratings indicated that all personal sensing methods were significantly more acceptable (all P<.001) compared with neutral across subjective measures of interference, dislike, and willingness to use for 1 year. Participants did not significantly differ in their dislike of active methods compared with passive methods (P=.23). However, participants reported a higher willingness to use passive (vs active) methods for 1 year (P=.04).

CONCLUSIONS

These results suggest that active and passive sensing methods are acceptable for people with alcohol use disorder over a longer period than has previously been assessed. Important individual differences were observed across people and methods, indicating opportunities for future improvement.

Leveraging Mobile Phone Sensors, Machine Learning and Explainable Artificial Intelligence to Predict Imminent Same-Day Binge Drinking Events to Support Just-In-Time Adaptive Interventions: A Feasibility Study

BACKGROUND

Digital Just-In-Time Adaptive Interventions (JITAIs) can reduce binge drinking events (BDEs: consuming 4+/5+ drinks per occasion for women/men) in young adults, but need to be optimized for timing and content. Delivering just-in-time support messages in the hours prior to BDEs could improve intervention impact.

OBJECTIVE

We determined the feasibility of developing a machine learning model to accurately predict future, that is, same-day, 1 to 6-hours prior BDEs using smartphone sensor data. We aimed to identify the most informative phone sensor features associated with BDEs on weekend and weekdays, respectively, to determine the key features that explain prediction model performance.

METHODS

We collected phone sensor data from 75 young adults (ages 21-25; mean =22.4, SD=1.9) with risky drinking behavior who reported drinking behavior over 14 weeks. Participants in this secondary analysis were enrolled in a clinical trial. We developed machine learning models testing different algorithms (e.g., XGBoost, decision tree) to predict same-day BDEs (versus low-risk drinking events and non-drinking periods) using smartphone sensor data (e.g., accelerometer, GPS). We tested various "prediction distance" time windows (more proximal: 1-hour; to distant: 6-hour) from drinking onset. We also tested various analysis time windows (i.e., amount of data to be analyzed), ranging from 1 to 12 hours prior to drinking onset, because this determines the amount of data that needs to be stored on the phone to compute the model. Explainable AI (XAI) was used to explore interactions between the most informative phone sensor features contributing to BDEs.

RESULTS

The XGBoost model performed best in predicting imminent same-day BDE, with 95.0% accuracy on weekends and 94.3% accuracy on weekdays (F1 score = 0.95 and 0.94, respectively). This XGBoost model needed 12- and 9-hours of phone sensor data at 3- and 6- hours prediction distance from the onset of drinking, on weekends and weekdays, respectively, prior to predicting same-day BDEs. The most informative phone sensor features for BDE prediction were time (e.g., time of day) and GPS-derived, such as radius of gyration (an indicator of travel). Interactions among key features (e.g., time of day, GPS-derived features) contributed to prediction of same-day BDE.

CONCLUSIONS

We demonstrated the feasibility and potential use of smartphone sensor data and machine learning to accurately predict imminent (same-day) BDEs in young adults. The prediction model provides "windows of opportunity" and with the adoption of XAI, we identified "key contributing features" to trigger JITAI prior to the onset of BDEs, with the potential to reduce the likelihood of BDEs in young adults.

CLINICALTRIAL

Developing and Testing a Smartphone Application to Enhance Adherence to Voice Therapy: A Pilot Study

The present study aimed to develop a smartphone application (app) that addressed identified barriers to success in voice therapy; accessibility, and poor adherence to home practice. The study objectives were (1) to investigate if app use enhanced adherence to the home practice of voice therapy and (2) to test app usability. Maximizing the effectiveness of voice therapy is vital as voice disorders are detrimental to personal and professional quality of life. A single-blinded randomized clinical trial was completed for the first objective. Participants included normophonic individuals randomly assigned to the app group or the traditional group. The primary outcome measure was adherence measured as the number of missed home practice tasks. The second objective was completed through usability testing and a focus group discussion. The app group (n = 12) missed approximately 50% less home practice tasks as compared to the traditional group (n = 13) and these results were statistically significant (p = 0.04). Dropout rates were comparable between the two groups. Usability results were positive for good usability with high perceived usefulness and perceived ease of use. App use resulted in improved adherence to home practice tasks. App usability results were positive, and participants provided specific areas of improvement which are achievable. Areas for improvement include app engagement and willingness to pay.

Leveraging mHealth and Wearable Sensors to Manage Alcohol Use Disorders: A Systematic Literature Review

Background: Alcohol use disorder (AUD) is a condition prevalent in many countries around the world, and the public burden of its treatment is close to $130 billion. mHealth offers several possible interventions to assist in the treatment of AUD. Objectives: To analyze the effectiveness of mHealth and wearable sensors to manage AUD from evidence published over the last 10 years. Methods: Following the Kruse Protocol and PRISMA 2020, four databases were queried (PubMed, CINAHL, Web of Science, and Science Direct) to identify studies with strong methodologies (n = 25). Results: Five interventions were identified, and 20/25 were effective at reducing alcohol consumption. Other interventions reported a decrease in depression and an increase in medication compliance. Primary barriers to the adoption of mHealth interventions are a requirement to train users, some are equally as effective as the traditional means of treatment, cost, and computer literacy. Conclusion: While not all mHealth interventions demonstrated statistically significant reduction in alcohol consumption, most are still clinically effective to treat AUD and provide a patient with their preference of a technologically inclined treatment Most interventions require training of users and some technology literacy, the barriers identified were very few compared with the litany of positive results.

Processing incomplete questionnaire data into continuous digital biomarkers for addiction monitoring

Purpose

eHealth systems allow efficient daily smartphone-based collection of self-reported data on mood, wellbeing, routines, and motivation; however, missing data is frequent. Within addictive disorders, missing data may reflect lack of motivation to stay sober. We hypothesize that qualitative questionnaire data contains valuable information, which after proper handling of missing data becomes more useful for practitioners.

Methods

Anonymized data from daily questionnaires containing 11 questions was collected with an eHealth system for 751 patients with alcohol use disorder (AUD). Two digital continuous biomarkers were composed from 9 wellbeing questions (WeBe-i) and from two questions representing motivation/self-confidence to remain sober (MotSC-i). To investigate possible loss of information in the process of composing the digital biomarkers, performance of neural networks to predict exacerbation events (relapse) in alcohol use disorder was compared.

Results

Long short-term memory (LSTM) neural networks predicted a coming exacerbation event 1–3 days (AUC 0.68–0.70) and 5–7 days (AUC 0.65–0.68) in advance on unseen patients. The predictive capability of digital biomarkers and raw questionnaire data was equal, indicating no loss of information. The transformation into digital biomarkers enable a continuous graphical display of each patient’s clinical course and a combined interpretation of qualitative and quantitative aspects of recovery on a time scale.

Conclusion

By transforming questionnaire data with large proportion of missing data into continuous digital biomarkers, the information captured by questionnaires can be more easily used in clinical practice. Information, assessed by the capability to predict exacerbation events of AUD, is preserved when processing raw questionnaire data into digital biomarkers.

Intensive Longitudinal Data Collection Using Microinteraction Ecological Momentary Assessment: Pilot and Preliminary Results

Background

Ecological momentary assessment (EMA) uses mobile technology to enable in situ self-report data collection on behaviors and states. In a typical EMA study, participants are prompted several times a day to answer sets of multiple-choice questions. Although the repeated nature of EMA reduces recall bias, it may induce participation burden. There is a need to explore complementary approaches to collecting in situ self-report data that are less burdensome yet provide comprehensive information on an individual’s behaviors and states. A new approach, microinteraction EMA (μEMA), restricts EMA items to single, cognitively simple questions answered on a smartwatch with single-tap assessments using a quick, glanceable microinteraction. However, the viability of using μEMA to capture behaviors and states in a large-scale longitudinal study has not yet been demonstrated.

Objective

This paper describes the μEMA protocol currently used in the Temporal Influences on Movement & Exercise (TIME) Study conducted with young adults, the interface of the μEMA app used to gather self-report responses on a smartwatch, qualitative feedback from participants after a pilot study of the μEMA app, changes made to the main TIME Study μEMA protocol and app based on the pilot feedback, and preliminary μEMA results from a subset of active participants in the TIME Study.

Methods

The TIME Study involves data collection on behaviors and states from 246 individuals; measurements include passive sensing from a smartwatch and smartphone and intensive smartphone-based hourly EMA, with 4-day EMA bursts every 2 weeks. Every day, participants also answer a nightly EMA survey. On non–EMA burst days, participants answer μEMA questions on the smartwatch, assessing momentary states such as physical activity, sedentary behavior, and affect. At the end of the study, participants describe their experience with EMA and μEMA in a semistructured interview. A pilot study was used to test and refine the μEMA protocol before the main study.

Results

Changes made to the μEMA study protocol based on pilot feedback included adjusting the single-question selection method and smartwatch vibrotactile prompting. We also added sensor-triggered questions for physical activity and sedentary behavior. As of June 2021, a total of 81 participants had completed at least 6 months of data collection in the main study. For 662,397 μEMA questions delivered, the compliance rate was 67.6% (SD 24.4%) and the completion rate was 79% (SD 22.2%).

Conclusions

The TIME Study provides opportunities to explore a novel approach for collecting temporally dense intensive longitudinal self-report data in a sustainable manner. Data suggest that μEMA may be valuable for understanding behaviors and states at the individual level, thus possibly supporting future longitudinal interventions that require within-day, temporally dense self-report data as people go about their lives.

Predicting Substance Use Treatment Failure with Transfer Learning

Transfer learning, which involves repurposing a trained model on a related task, may allow for better predictions with substance use data than models that are trained using the target data alone. This approach may also be useful for small clinical datasets. The current study examined a method of classifying substance use treatment success using transfer learning. Transfer learning was used to classify data from a nationwide database. We trained a convolutional neural network on a heroin use treatment dataset, then trained and tested on a smaller opioid use treatment dataset. We compared this model with a baseline model that did not benefit from transfer learning, and a tuned random forest (RF). The goal was to see if model weights transfer across related substances and from large to small datasets. The transfer model outperformed the RF model and baseline model. These findings suggest leveraging the power of large datasets for transfer learning may be an effective approach in predicting substance use disorder (SUD) treatment outcomes. It is possible to achieve a score that performs better than RF using transfer learning.

Digital Public Health Surveillance Tools for Alcohol Use and HIV Risk Behaviors

There is a need for real-time and predictive data on alcohol use both broadly and specific to HIV. However, substance use and HIV data often suffer from lag times in reporting as they are typically measured from surveys, clinical case visits and other methods requiring extensive time for collection and analysis. Social big data might help to address this problem and be used to provide near real-time assessments of people's alcohol use and/or alcohol. This manuscript describes three types of social data sources (i.e., social media data, internet search data, and wearable device data) that might be used in surveillance of alcohol and HIV, and then discusses the implications and potential of implementing them as additional tools for public health surveillance.

Smartphone Alcohol Use Disorder Recovery Apps: a Survey of Behavioral Intention to Use (Preprint)

Background

Alcohol use disorder (AUD) carries a huge health and economic cost to society. Effective interventions exist but numerous challenges limit their adoption, especially in a pandemic context. AUD recovery apps (AUDRA) have emerged as a potential complement to in-person interventions. They are easy to access and show promising results in terms of efficacy. However, they rely on individual adoption decisions and remain underused.

Objective

The aim of this survey study is to explore the beliefs that determine the intention to use AUDRA.

Methods

We conducted a cross-sectional survey study of people with AUD. We used the Unified Theory of Acceptance and Use of Technology, which predicts use and behavioral intention to use based on performance expectancy, effort expectancy, social influence, and facilitating conditions. Participants were recruited directly from 2 sources; first, respondents at addiction treatment facilities in Ontario, Canada, were contacted in person, and they filled a paper form; second, members from AUD recovery support groups on social media were contacted and invited to fill an internet-based survey. The survey was conducted between October 2019 and June 2020.

Results

The final sample comprised 159 participants (124 involved in the web-based survey and 35 in the paper-based survey) self-identifying somewhat or very much with AUD. Most participants (n=136, 85.5%) were aware of AUDRA and those participants scored higher on performance expectancy, effort expectancy, and social influence. Overall, the model explains 35.4% of the variance in the behavioral intention to use AUDRA and 11.1% of the variance in use. Social influence (P=.31), especially for women (P=.23), and effort expectancy (P=.25) were key antecedents of behavioral intention. Facilitating conditions were not significant overall but were moderated by age (P=.23), suggesting that it matters for older participants. Performance expectancy did not predict behavioral intention, which is unlike many other technologies but confirms other findings associated with mobile health (mHealth). Open-ended questions suggest that privacy concerns may significantly influence the use of AUDRA.

Conclusions

This study suggests that unlike many other technologies, the adoption of AUDRA is not mainly determined by utilitarian factors such as performance expectancy. Rather, effort expectancy and social influence play a key role in determining the intention to use AUDRA.

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

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

A reinforcement learning based algorithm for personalization of digital, just-in-time, adaptive interventions

Suboptimal health related behaviors and habits; and resulting chronic diseases are responsible for majority of deaths globally. Studies show that providing personalized support to patients yield improved results by preventing and/or timely treatment of these problems. Digital, just-in-time and adaptive interventions are mobile phone-based notifications that are being utilized to support people wherever and whenever necessary in coping with their health problems. In this research, we propose a reinforcement learning-based mechanism to personalize interventions in terms of timing, frequency and preferred type(s). We simultaneously employ two reinforcement learning models, namely intervention-selection and opportune-moment-identification; capturing and exploiting changes in people's long-term and momentary contexts respectively. While the intervention-selection model adapts the intervention delivery with respect to type and frequency, the opportune-moment-identification model tries to find the most opportune moments to deliver interventions throughout a day. We propose two accelerator techniques over the standard reinforcement learning algorithms to boost learning performance. First, we propose a customized version of eligibility traces for rewarding past actions throughout an agent's trajectory. Second, we utilize the transfer learning method to reuse knowledge across multiple learning environments. We validate the proposed approach in a simulated experiment where we simulate four personas differing in their daily activities, preferences on specific intervention types and attitudes towards the targeted behavior. Our experiments show that the proposed approach yields better results compared to the standard reinforcement learning algorithms and successfully capture the simulated variations associated with the personas.

Substance Use Disorders and Telehealth in the COVID-19 Pandemic Era: A New Outlook

During the current coronavirus disease 2019 epidemic, many outpatient chemical dependency treatment programs and clinics are decreasing their number of in-person patient contacts. This has widened an already large gap between patients with substance use disorders (SUDs) who need treatment and those who have actually received treatment. For a disorder where group therapy has been the mainstay treatment option for decades, social distancing, shelter in place, and treatment discontinuation have created an urgent need for alternative approaches to addiction treatment. In an attempt to continue some care for patients in need, many medical institutions have transitioned to a virtual environment to promote safe social distancing. Although there is ample evidence to support telemedical interventions, these can be difficult to implement, especially in the SUD population. This article reviews current literature for the use of telehealth interventions in the treatment of SUDs and offers recommendations on safe and effective implementation strategies based on the current literature.

Predicting alcohol dependence treatment outcomes: a prospective comparative study of clinical psychologists versus 'trained' machine learning models

BACKGROUND AND AIMS

Clinical staff are typically poor at predicting alcohol dependence treatment outcomes. Machine learning (ML) offers the potential to model complex clinical data more effectively. This study tested the predictive accuracy of ML algorithms demonstrated to be effective in predicting alcohol dependence outcomes, compared with clinical judgement and traditional linear regression.

DESIGN

Prospective study. ML models were trained on 1016 previously treated patients (training-set) who attended a hospital-based alcohol and drug clinic. ML models (n = 27), clinical psychologists (n = 10) and a 'traditional' logistic regression model (n = 1) predicted treatment outcome during the initial treatment session of an alcohol dependence programme.

SETTING

A 12-week cognitive behavioural therapy (CBT)-based abstinence programme for alcohol dependence in a hospital-based alcohol and drug clinic in Australia.

PARTICIPANTS

Prospective predictions were made for 220 new patients (test-set; 70.91% male, mean age = 35.78 years, standard deviation = 9.19). Sixty-nine (31.36%) patients successfully completed treatment.

MEASUREMENTS

Treatment success was the primary outcome variable. The cross-validated training-set accuracy of ML models was used to determine optimal parameters for selecting models for prospective prediction. Accuracy, sensitivity, specificity, area under the receiver operator curve (AUC), Brier score and calibration curves were calculated and compared across predictions.

FINDINGS

The mean aggregate accuracy of the ML models (63.06%) was higher than the mean accuracy of psychologist predictions (56.36%). The most accurate ML model achieved 70% accuracy, as did logistic regression. Both were more accurate than psychologists (P < 0.05) and had superior calibration. The high specificity for the selected ML (79%) and logistic regression (90%) meant they were significantly (P < 0.001) more effective than psychologists (50%) at correctly identifying patients whose treatment was unsuccessful. For ML and logistic regression, high specificity came at the expense of sensitivity (26 and 31%, respectively), resulting in poor prediction of successful patients.

CONCLUSIONS

Machine learning models and logistic regression appear to be more accurate than psychologists at predicting treatment outcomes in an abstinence programme for alcohol dependence, but sensitivity is low.

NIATx-TI versus typical product training on e-health technology implementation: a clustered randomized controlled trial study protocol

BACKGROUND

Substance use disorders (SUDs) lead to tens-of-thousands of overdose deaths and other forms of preventable deaths in the USA each year. This results in over $500 billion per year in societal and economic costs as well as a considerable amount of grief for loved ones of affected individuals. Despite these health and societal consequences, only a small percentage of people seek treatment for SUDs, and the majority of those that seek help fail to achieve long-term sobriety. E-health applications in healthcare have proven to be effective at sustaining treatment and reaching patients traditional treatment pathways would have missed. However, e-health adoption and sustainment rates in healthcare are poor, especially in the SUD treatment sector. Implementation engineering can address this gap in the e-health field by augmenting existing implementation models, which explain organizational and individual e-health behaviors retrospectively, with prospective resources that can guide implementation.

METHODS

This cluster randomized control trial is designed to test two implementation strategies at adopting an evidence-based mobile e-health technology for SUD treatment. The proposed e-health implementation model is the Network for the Improvement of Addiction Treatment-Technology Implementation (NIATx-TI) Framework. This project, based in Iowa, will compare a control condition (using a typical software product training approach that includes in-person staff training followed by access to on-line support) to software implementation utilizing NIATx-TI, which includes change management training, followed by coaching on how to implement and use the mobile application. While e-health spans many modalities and health disciplines, this project will focus on implementing the Addiction Comprehensive Health Enhancement Support System (A-CHESS), an evidence-based SUD treatment recovery app framework. This trial will be conducted in Iowa at 46 organizational sites within 12 SUD treatment agencies. The control arm consists of 23 individual treatment sites based at five organizations, and the intervention arm consists of 23 individual SUD treatment sites based at seven organizations DISCUSSION: This study addresses an issue of substantial public health significance: enhancing the uptake of the growing inventory of patient-centered evidence-based addiction treatment e-health technologies.

TRIAL REGISTRATION

ClinicalTrials.gov , NCT03954184 . Posted 17 May 2019.

Prediction of stress and drug craving ninety minutes in the future with passively collected GPS data

Just-in-time adaptive interventions (JITAIs), typically smartphone apps, learn to deliver therapeutic content when users need it. The challenge is to "push" content at algorithmically chosen moments without making users trigger it with effortful input. We trained a randomForest algorithm to predict heroin craving, cocaine craving, or stress (reported via smartphone app 3x/day) 90 min into the future, using 16 weeks of field data from 189 outpatients being treated for opioid-use disorder. We used only one form of continuous input (along with person-level demographic data), collected passively: an indicator of environmental exposures along the past 5 h of movement, as assessed by GPS. Our models achieved excellent overall accuracy-as high as 0.93 by the end of 16 weeks of tailoring-but this was driven mostly by correct predictions of absence. For predictions of presence, "believability" (positive predictive value, PPV) usually peaked in the high 0.70s toward the end of the 16 weeks. When the prediction target was more rare, PPV was lower. Our findings complement those of other investigators who use machine learning with more broadly based "digital phenotyping" inputs to predict or detect mental and behavioral events. When target events are comparatively subtle, like stress or drug craving, accurate detection or prediction probably needs effortful input from users, not passive monitoring alone. We discuss ways in which accuracy is difficult to achieve or even assess, and warn that high overall accuracy (including high specificity) can mask the abundance of false alarms that low PPV reveals.

Mobile Sensing in Substance Use Research: A Scoping Review

Background: Addictive disorders and substance use are significant health challenges worldwide, and relapse is a core component of addictive disorders. The dynamics surrounding relapse and especially the immediate period before it occurs is only partly understood, much due to difficulties collecting reliable and sufficient data from this narrow period. Mobile sensing has been an important way to improve data quality and enhance predictive capabilities for symptom worsening within physical and mental health care, but is less developed within substance use research. Methodology: This scoping review aimed to reviewing the currently available research on mobile sensing of substance use and relapse in substance use disorders. The search was conducted in January 2019 using PubMed and Web of Science. Results: Six articles were identified, all concerning subjects using alcohol. In the studies a range of mobile sensors and derived aggregated features were employed. Data collected through mobile sensing were predominantly used to make dichotomous inference on ongoing substance use or not and in some cases on the quantity of substance intake. Only one of the identified studies predicted later substance use. A range of statistical machine learning techniques was employed. Conclusions: The research on mobile sensing in this field remains scarce. The issues requiring further attention include more research on clinical populations in naturalistic settings, use of a priori knowledge in statistical modeling, focus on prediction of substance use rather than purely identification, and finally research on other substances than alcohol.

Measures of Effectiveness, Efficiency, and Quality of Telemedicine in the Management of Alcohol Abuse, Addiction, and Rehabilitation: Systematic Review

Background

More than 18 million Americans are currently suffering from alcohol use disorder (AUD): a compulsive behavior of alcohol use as a result of a chronic, relapsing brain disease. With alcohol-related injuries being one of the leading causes of preventable deaths, there is a dire need to find ways to assist those suffering from alcohol dependence. There still exists a gap in knowledge as to the potential of telemedicine in improving health outcomes for those patients suffering from AUD.

Objective

The purpose of this systematic review was to evaluate the measures of effectiveness, efficiency, and quality that result from the utilization of telemedicine in the management of alcohol abuse, addiction, and rehabilitation.

Methods

This review was conducted utilizing the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The articles used in this analysis were gathered using keywords inclusive of both telemedicine and alcohol abuse, which were then searched in the Cumulative Index to Nursing and Allied Health Literature, Cochrane, and MEDLINE (PubMed) databases. A total of 22 articles were chosen for analysis.

Results

The results indicated that telemedicine reduced alcohol consumption. Other common outcomes included reduced depression (4/35, 11%), increased patient satisfaction (3/35, 9%), increase in accessibility (3/35, 9%), increased quality of life (2/35, 6%), and decreased cost (1/35, 3%). Interventions included mobile health (11/22, 50%), electronic health (6/22, 27%), telephone (3/33, 14%), and 2-way video (2/22, 9%). Studies were conducted in 3 regions: the United States (13/22, 59%), the European Union (8/22, 36%), and Australia (1/22, 5%).

Conclusions

Telemedicine was found to be an effective tool in reducing alcohol consumption and increasing patients’ accessibility to health care services or health providers. The group of articles for analysis suggested that telemedicine may be effective in reducing health care costs and improving the patient’s quality of life. Although telemedicine shows promise as an effective way to manage alcohol-related disorders, it should be further investigated before implementation.

Design and Content Quality of Alcohol-Related German, French and Italian Self-Tracking Applications

Background: Research on the increasing use of mobile technology in the addiction field is mainly focused on data collection and brief interventions. The acceptance and outcomes of autonomous self-tracking and self-governance as key elements for behavior change are under-researched. Purpose/Objectives: The objective of the study was to conduct a quality assessment of design and content features of self-tracking smartphone applications related to alcohol use, available in German, Italian, or French. Methods: A total of 25 self-tracking applications were identified, of which 17 could be assessed with the Mobile App Rating Scale (MARS), the System Usability Scale (SUS), and an additional content quality checklist based on the theoretical self-change framework (n = 13). Results: The scale design analysis showed a rather positive picture. Using the SUS, only six cases were below the reference average (x = 68), and three were clearly above average. Application of the MARS showed higher scores among the self-tracking applications in this study than among the health applications reviewed in the original MARS study. Better design quality goes together with better basic content quality. However, a closer look at the "interactivity scores" and the "risk/information barometer," as well as at the individual subtopics of the 10-point content checklist revealed major shortcomings. Conclusions/Importance: Improvements are necessary for consumer information in app stores, increased availability of alcohol-related self-tracking applications, transparent quality assurance regarding evidence-based content, and user-friendly design quality, to provide guidance for potential users on how to successfully navigate a highly unstable digital environment.

Prevention, screening, and treatment for heavy drinking and alcohol use disorder

Heavy drinking and alcohol use disorder are major public health problems. Practitioners not specialising in alcohol treatment are often unaware of the guidelines for preventing, identifying, and treating heavy drinking and alcohol use disorder. However, a consensus exists that clinically useful and valuable tools are available to address these issues. Here, we review existing information and developments from the past 5 years in these areas. We also include information on heavy drinking and alcohol use disorder among individuals with co-occurring psychiatric disorders, including drug use disorders. Areas covered include prevention; screening, brief intervention, and referral for treatment; evidence-based behavioural interventions; medication-assisted treatment; technology-based interventions (eHealth and mHealth); and population-level interventions. We also discuss the key topics for future research.

The utility and impact of the addiction comprehensive health enhancement support system (ACHESS) on substance abuse treatment adherence among youth in an intensive outpatient program

Youth experiencing substance use disorders often are susceptible to relapse because traditional support systems can be expensive, geographically dispersed, operated on limited schedules and lacking in peer support. The continuity of care offered via the digital Addiction Comprehensive Health Enhancement Support System (ACHESS) system holds promise in preventing relapse because of its portability and capability to foster virtually anytime/anywhere, cost-effective access to supportive interventions. The aim of this mixed-methods study was to evaluate the utility and impact of ACHESS on treatment adherence among youth with substance use disorders in an intensive outpatient program in the US Midwest. Data on 28 clients using ACHESS during 2016-17 were compared to retrospective data on 28 carefully-matched others treated without ACHESS during 2014-16. Fifty-four percent of the study group successfully completed treatment as opposed to 42.9% of those in the comparison group. Staff focus group findings highlighted how some features of ACHESS were effectively integrated into the care model and appeared to positively impact outcomes, while other elements of the application offered little utility. We suggest further study of ACHESS among larger samples of youth with substance use disorders in intensive outpatient programs to assess its efficacy in supporting adherence to treatment.

An expandable approach for design and personalization of digital, just-in-time adaptive interventions

Objective

We aim to deliver a framework with 2 main objectives: 1) facilitating the design of theory-driven, adaptive, digital interventions addressing chronic illnesses or health problems and 2) producing personalized intervention delivery strategies to support self-management by optimizing various intervention components tailored to people's individual needs, momentary contexts, and psychosocial variables.

Materials and Methods

We propose a template-based digital intervention design mechanism enabling the configuration of evidence-based, just-in-time, adaptive intervention components. The design mechanism incorporates a rule definition language enabling experts to specify triggering conditions for interventions based on momentary and historical contextual/personal data. The framework continuously monitors and processes personal data space and evaluates intervention-triggering conditions. We benefit from reinforcement learning methods to develop personalized intervention delivery strategies with respect to timing, frequency, and type (content) of interventions. To validate the personalization algorithm, we lay out a simulation testbed with 2 personas, differing in their various simulated real-life conditions.

Results

We evaluate the design mechanism by presenting example intervention definitions based on behavior change taxonomies and clinical guidelines. Furthermore, we provide intervention definitions for a real-world care program targeting diabetes patients. Finally, we validate the personalized delivery mechanism through a set of hypotheses, asserting certain ways of adaptation in the delivery strategy, according to the differences in simulation related to personal preferences, traits, and lifestyle patterns.

Conclusion

While the design mechanism is sufficiently expandable to meet the theoretical and clinical intervention design requirements, the personalization algorithm is capable of adapting intervention delivery strategies for simulated real-life conditions.

OnTrack: development and feasibility of a smartphone app designed to predict and prevent dietary lapses

Given that the overarching goal of weight loss programs is to remain adherent to a dietary prescription, specific moments of nonadherence known as "dietary lapses" can threaten weight control via the excess energy intake they represent and by provoking future lapses. Just-in-time adaptive interventions could be particularly useful in preventing dietary lapses because they use real-time data to generate interventions that are tailored and delivered at a moment computed to be of high risk for a lapse. To this end, we developed a smartphone application (app) called OnTrack that utilizes machine learning to predict dietary lapses and deliver a targeted intervention designed to prevent the lapse from occurring. This study evaluated the feasibility, acceptability, and preliminary effectiveness of OnTrack among weight loss program participants. An open trial was conducted to investigate subjective satisfaction, objective usage, algorithm performance, and changes in lapse frequency and weight loss among individuals (N = 43; 86% female; body mass index = 35.6 kg/m2) attempting to follow a structured online weight management plan for 8 weeks. Participants were adherent with app prompts to submit data, engaged with interventions, and reported high levels of satisfaction. Over the course of the study, participants averaged a 3.13% weight loss and experienced a reduction in unplanned lapses. OnTrack, the first Just-in-time adaptive intervention for dietary lapses was shown to be feasible and acceptable, and OnTrack users experienced weight loss and lapse reduction over the study period. These data provide the basis for further development and evaluation.

Digital Interventions for Mental Disorders: Key Features, Efficacy, and Potential for Artificial Intelligence Applications

Mental disorders are highly prevalent and often remain untreated. Many limitations of conventional face-to-face psychological interventions could potentially be overcome through Internet-based and mobile-based interventions (IMIs). This chapter introduces core features of IMIs, describes areas of application, presents evidence on the efficacy of IMIs as well as potential effect mechanisms, and delineates how Artificial Intelligence combined with IMIs may improve current practices in the prevention and treatment of mental disorders in adults. Meta-analyses of randomized controlled trials clearly show that therapist-guided IMIs can be highly effective for a broad range of mental health problems. Whether the effects of unguided IMIs are also clinically relevant, particularly under routine care conditions, is less clear. First studies on IMIs for the prevention of mental disorders have shown promising results. Despite limitations and challenges, IMIs are increasingly implemented into routine care worldwide. IMIs are also well suited for applications of Artificial Intelligence and Machine Learning, which provides ample opportunities to improve the identification and treatment of mental disorders. Together with methodological innovations, these approaches may also deepen our understanding of how psychological interventions work, and why. Ethical and professional restraints as well as potential contraindications of IMIs, however, should also be considered. In sum, IMIs have a high potential for improving the prevention and treatment of mental health disorders across various indications, settings, and populations. Therefore, implementing IMIs into routine care as both adjunct and alternative to face-to-face treatment is highly desirable. Technological advancements may further enhance the variability and flexibility of IMIs, and thus even further increase their impact in people's lives in the future.

A P5 Approach to m-Health: Design Suggestions for Advanced Mobile Health Technology

In recent years, technology has been developed as an important resource for health care management, especially in regard to chronic conditions. In the broad field of eHealth, mobile technology (mHealth) is increasingly used to empower patients not only in disease management but also in the achievement of positive experiences and experiential growth. mHealth tools are considered powerful because, unlike more traditional Internet-based tools, they allow patients to be continuously monitored and followed by their own mobile devices and to have continual access to resources (e.g., mobile apps or functions) supporting health care management activities. However, the literature has shown that, in many cases, such technology not accepted and/or adopted in the long term by its users. To address this issue, this article reviews the main factors influencing mHealth technology acceptance/adoption in health care. Finally, based on the main aspects emerging from the review, we propose an innovative approach to mHealth design and implementation, namely P5 mHealth. Relying on the P5 approach to medicine and health care, this approach provides design suggestions to address mHealth adoption issues already at the initial stages of development of the technologies.

What Do You Say Before You Relapse? How Language Use in a Peer-to-peer Online Discussion Forum Predicts Risky Drinking among Those in Recovery

Increasingly, individuals with alcohol use disorder (AUD) seek and provide support for relapse prevention in text-based online environments such as discussion forums. This paper investigates whether language use within a peer-to-peer discussion forum can predict future relapse among individuals treated for AUD. A total of 104 AUD sufferers who had completed residential treatment participated in a mobile phone-based relapse-prevention program, where they communicated via an online forum over the course of a year. We extracted patterns of language use on the forum within the first four months on study using Linguistic Inquiry and Word Count (LIWC), a dictionary-based text analysis program. Participants reported their incidence of risky drinking via a survey at 4, 8, and 12 months. A logistic regression model was built to predict the likelihood that individuals would engage in risky drinking within a year based on their language use, while controlling for baseline characteristics and rates of utilizing the mobile system. Results show that all baseline characteristics and system use factors explained just 13% of the variance in relapse, whereas a small number of linguistic cues, including swearing and cognitive mechanism words, accounted for an additional 32% of the total 45% of variance in relapse explained by the model. Effective models for predicting relapse are needed. Messages exchanged on AUD forums could provide an unobtrusive and cost-effective window into the future health outcomes of AUD sufferers, and their psychological underpinnings. As online communication expands, models that leverage user-submitted text toward predicting relapse will be increasingly scalable and actionable.

Feasibility and Effects of Digital Interventions to Support People in Recovery From Substance Use Disorders: Systematic Review

Background

The development and evaluation of digital interventions aimed at preventing or treating substance use–related problems and disorders is a rapidly growing field. Previous reviews of such interventions reveal a large and complex picture with regard to targeted users, use, and efficacy.

Objective

The objective of this review was to investigate the feasibility and effects of interventions developed specifically for digital platforms. These interventions are focused on supporting people in recovery from substance use disorders by helping them achieve their substance use goals and develop a more satisfying life situation.

Methods

The review is based on a systematic search in MEDLINE, Embase, PsycInfo, and Cochrane Library databases. Of the 1149 identified articles, 722 were excluded as obviously not relevant. Of the remaining articles, 21 were found to be previous reviews, 269 were on interventions aimed at reducing hazardous alcohol or cannabis use, and 94 were on digitized versions of standard treatment methods. The remaining 43 articles were all read in full and systematically scored by both authors.

Results

The 43 articles cover 28 unique interventions, of which 33 have been published after 2013. The interventions are aimed at different target groups (defined by age, substance, or comorbidity). Based on the number of features or modules, the interventions can be categorized as simple or complex. Fourteen of the 18 simple interventions and 9 of the 10 complex interventions have been studied with quantitative controlled methodologies. Thirteen of the 18 simple interventions are integrated in other treatment or support systems, mainly delivered as mobile phone apps, while 6 of the 10 complex interventions are designed as stand-alone interventions, most often delivered on a platform combining desktop/Web and mobile phone technologies. The interventions were generally easy to implement, but in most cases the implementation of the complex interventions was found to be dependent on sustained organizational support. Between 70% and 90% of the participants found the interventions to be useful and easy to use. The rates of sustained use were also generally high, except for simple interventions with an open internet-based recruitment and some information and education modules of the complex interventions. Across all interventions, slightly more than half (55%) of the studies with control groups generated positive findings on 1 or more substance use outcomes, with 57% of the interventions also found to be efficacious in 1 or more studies. In the positive studies, effects were typically in the small to moderate range, with a few studies yielding larger effects. Largely due to the inclusion of stronger control conditions, studies of simple interventions were less likely to produce positive effects.

Conclusions

The digital interventions included in this review are in general feasible but are not consistently effective in helping people in recovery from substance use disorder reduce their substance use or achieving other recovery goals.

Nicotine dependence, internalizing symptoms, mood variability and daily tobacco use among young adult smokers

INTRODUCTION

Cigarette use among young adults continues to rise. As young adults transition to college and assume other adult roles and responsibilities, they are at risk for the development of mental health problems and for the progression of substance use problems. Previous studies suggest that individual differences in negative and positive mood contribute to cigarette use in established college-aged smokers, but less is known whether fluctuations in mood influence daily cigarette use, controlling for trait levels of internalizing symptoms and nicotine dependence.

METHODS

Data for this study came from a sample of college students (N=39, 59% female, mean age 20.4years) who reported regular cigarette use and participated in a 21-day ecological momentary assessment (EMA) study assessing within-individual variation in cigarette use and mood.

RESULTS

A three-level hierarchical linear model accounting for the structure of 1896 occasions of cigarette use nested within days and individuals indicated that within-individual variability in positive mood was associated with cigarette use at each occasion, after taking into account baseline levels of nicotine dependence and internalizing problems.

CONCLUSIONS

Daily shifts in positive moods are importantly associated with consuming cigarettes throughout the day.

Reprint of Using ecological momentary assessments to predict relapse after adult substance use treatment

BACKGROUND

A key component of relapse prevention is to self-monitor the internal (feelings or cravings) and external (people, places, activities) factors associated with relapse. Smartphones can deliver ecological momentary assessments (EMA) to help individuals self-monitor. The purpose of this exploratory study was to develop a model for predicting an individual's risk of future substance use after each EMA and validate it using a multi-level model controlling for repeated measures on persons.

METHODS

Data are from 21,897 observations from 43 adults following their initial episode of substance use treatment in Chicago from 2015 to 2016. Participants were provided smartphones for six months and asked to complete two to three minute EMAs at five random times per day (81% completion). In any given EMA, 2.7% reported substance use and 8% reported any use in the next five completed EMA. Chi-square Automatic Interaction Detector (CHAID) was used to classify EMAs into six levels of risk and then validated with a hierarchical linear model (HLM).

RESULTS

The major predictors of substance use in the next five completed EMAs were substance use pattern over the current and prior five EMAs (no recent/current use, either recent or current use [but not both], continued use [both recent and current]), negative affect (feelings), and craving (rating). Negative affect was important for EMAs with no current or recent use reported; craving was important for EMAs with either recent or current use; and neither mattered for EMAs with continued use. The CHAID gradated EMA risk from 0.7% to 36.6% of the next five completed EMAs with substance use reported. It also gradated risk of "any" use in the next five completed EMAs from 3% to 82%.

CONCLUSIONS

This study demonstrated the potential of using smartphone-based EMAs to monitor and provide feedback for relapse prevention in future studies.

Using ecological momentary assessments to predict relapse after adult substance use treatment

BACKGROUND

A key component of relapse prevention is to self-monitor the internal (feelings or cravings) and external (people, places, activities) factors associated with relapse. Smartphones can deliver ecological momentary assessments (EMA) to help individuals self-monitor. The purpose of this exploratory study was to develop a model for predicting an individual's risk of future substance use after each EMA and validate it using a multi-level model controlling for repeated measures on persons.

METHODS

Data are from 21,897 observations from 43 adults following their initial episode of substance use treatment in Chicago from 2015 to 2016. Participants were provided smartphones for six months and asked to complete two to three minute EMAs at five random times per day (81% completion). In any given EMA, 2.7% reported substance use and 8% reported any use in the next five completed EMA. Chi-square Automatic Interaction Detector (CHAID) was used to classify EMAs into six levels of risk and then validated with a hierarchical linear model (HLM).

RESULTS

The major predictors of substance use in the next five completed EMAs were substance use pattern over the current and prior five EMAs (no recent/current use, either recent or current use [but not both], continued use [both recent and current]), negative affect (feelings), and craving (rating). Negative affect was important for EMAs with no current or recent use reported; craving was important for EMAs with either recent or current use; and neither mattered for EMAs with continued use. The CHAID gradated EMA risk from 0.7% to 36.6% of the next five completed EMAs with substance use reported. It also gradated risk of "any" use in the next five completed EMAs from 3% to 82%.

CONCLUSIONS

This study demonstrated the potential of using smartphone-based EMAs to monitor and provide feedback for relapse prevention in future studies.

Predictive modeling in e-mental health: A common language framework

Recent developments in mobile technology, sensor devices, and artificial intelligence have created new opportunities for mental health care research. Enabled by large datasets collected in e-mental health research and practice, clinical researchers and members of the data mining community increasingly join forces to build predictive models for health monitoring, treatment selection, and treatment personalization. This paper aims to bridge the historical and conceptual gaps between the distant research domains involved in this new collaborative research by providing a conceptual model of common research goals. We first provide a brief overview of the data mining field and methods used for predictive modeling. Next, we propose to characterize predictive modeling research in mental health care on three dimensions: 1) time, relative to treatment (i.e., from screening to post-treatment relapse monitoring), 2) types of available data (e.g., questionnaire data, ecological momentary assessments, smartphone sensor data), and 3) type of clinical decision (i.e., whether data are used for screening purposes, treatment selection or treatment personalization). Building on these three dimensions, we introduce a framework that identifies four model types that can be used to classify existing and future research and applications. To illustrate this, we use the framework to classify and discuss published predictive modeling mental health research. Finally, in the discussion, we reflect on the next steps that are required to drive forward this promising new interdisciplinary field.

Biosignature Discovery for Substance Use Disorders Using Statistical Learning

There are limited biomarkers for substance use disorders (SUDs). Traditional statistical approaches are identifying simple biomarkers in large samples, but clinical use cases are still being established. High-throughput clinical, imaging, and 'omic' technologies are generating data from SUD studies and may lead to more sophisticated and clinically useful models. However, analytic strategies suited for high-dimensional data are not regularly used. We review strategies for identifying biomarkers and biosignatures from high-dimensional data types. Focusing on penalized regression and Bayesian approaches, we address how to leverage evidence from existing studies and knowledge bases, using nicotine metabolism as an example. We posit that big data and machine learning approaches will considerably advance SUD biomarker discovery. However, translation to clinical practice, will require integrated scientific efforts.

Self-guided Change: The most common form of long-term, maintained health behavior change

Millions of people change risky, health-related behaviors and maintain those changes. However, they often take years to change, and their unhealthy behaviors may harm themselves and others and constitute a significant cost to society. A review-similar in nature to a scoping review-was done of the literature related to long-term health behavior change in six areas: alcohol, cocaine and heroin misuse, gambling, smoking, and overeating. Based on the limited research available, reasons for change and strategies for changing and for maintaining change were also reviewed. Fifty years of research clearly indicate that as people age, in the case of alcohol, heroin and cocaine misuse, smoking, and gambling, 80-90 percent moderate or stop their unhealthy behaviors. The one exception is overeating; only 20 percent maintain their weight loss. Most of these changes, when they occur, appear to be the result of self-guided change. More ways to accelerate self-guided, health-related behavior change need to be developed and disseminated.

To identify predictors of relapse in cases of alcohol dependence syndrome in relation to life events

BACKGROUND AND OBJECTIVES

Relapse is a complex and dynamic phenomenon that appears to be determined by biological, psychological, and social factors and an interaction among these. This study examined the association between demographic variables, clinical parameters, and psychosocial factors that predict the vulnerability to relapse in cases of alcohol dependence syndrome.

MATERIALS AND METHODS

Structured assessments of clinical/demographic parameters, relapse precipitants, life events, and dysfunction were carried out among patients with alcohol dependence syndrome (n = 100) who had relapsed and compared with those (n = 100) who had managed to remain abstinent.

RESULTS

Patients who had relapsed were found to have significantly more positive family history of substance use, past history of alcohol-related comorbidity, experienced a higher number of undesirable life events, and higher negative mood states and social anxiety and dysfunction in social, vocational, personal, family, and cognitive spheres compared to patients who had remained abstinent.

CONCLUSIONS

Relapse in alcohol dependents is an interaction of many factors, and multiple layers of assessment may be required to predict relapse. This study provided further evidence in support of the importance of certain clinical/psychosocial factors in relapse in substance dependence. It provides the basis for investigating the correlates of relapse in a wide range of behavioral and substance use problems.

e-Addictology: An Overview of New Technologies for Assessing and Intervening in Addictive Behaviors

BACKGROUND

New technologies can profoundly change the way we understand psychiatric pathologies and addictive disorders. New concepts are emerging with the development of more accurate means of collecting live data, computerized questionnaires, and the use of passive data. Digital phenotyping, a paradigmatic example, refers to the use of computerized measurement tools to capture the characteristics of different psychiatric disorders. Similarly, machine learning-a form of artificial intelligence-can improve the classification of patients based on patterns that clinicians have not always considered in the past. Remote or automated interventions (web-based or smartphone-based apps), as well as virtual reality and neurofeedback, are already available or under development.

OBJECTIVE

These recent changes have the potential to disrupt practices, as well as practitioners' beliefs, ethics and representations, and may even call into question their professional culture. However, the impact of new technologies on health professionals' practice in addictive disorder care has yet to be determined. In the present paper, we therefore present an overview of new technology in the field of addiction medicine.

METHOD

Using the keywords [e-health], [m-health], [computer], [mobile], [smartphone], [wearable], [digital], [machine learning], [ecological momentary assessment], [biofeedback] and [virtual reality], we searched the PubMed database for the most representative articles in the field of assessment and interventions in substance use disorders.

RESULTS

We screened 595 abstracts and analyzed 92 articles, dividing them into seven categories: e-health program and web-based interventions, machine learning, computerized adaptive testing, wearable devices and digital phenotyping, ecological momentary assessment, biofeedback, and virtual reality.

CONCLUSION

This overview shows that new technologies can improve assessment and interventions in the field of addictive disorders. The precise role of connected devices, artificial intelligence and remote monitoring remains to be defined. If they are to be used effectively, these tools must be explained and adapted to the different profiles of physicians and patients. The involvement of patients, caregivers and other health professionals is essential to their design and assessment.

Using smartphones to decrease substance use via self-monitoring and recovery support: study protocol for a randomized control trial

Background

Alcohol abuse, other substance use disorders, and risk behaviors associated with the human immunodeficiency virus (HIV) represent three of the top 10 modifiable causes of mortality in the US. Despite evidence that continuing care is effective in sustaining recovery from substance use disorders and associated behaviors, patients rarely receive it. Smartphone applications (apps) have been effective in delivering continuing care to patients almost anywhere and anytime. This study tests the effectiveness of two components of such apps: ongoing self-monitoring through Ecological Momentary Assessments (EMAs) and immediate recovery support through Ecological Momentary Interventions (EMIs).

Methods/design

The target population, adults enrolled in substance use disorder treatment (n = 400), are being recruited from treatment centers in Chicago and randomly assigned to one of four conditions upon discharge in a 2 × 2 factorial design. Participants receive (1) EMAs only, (2) EMIs only, (3) combined EMAs + EMIs, or (4) a control condition without EMA or EMI for 6 months. People in the experimental conditions receive smartphones with the apps (EMA and/or EMI) specific to their condition. Phones alert participants in the EMA and EMA + EMI conditions at five random times per day and present participants with questions about people, places, activities, and feelings that they experienced in the past 30 min and whether these factors make them want to use substances, support their recovery, or have no impact. Those in the EMI and EMA + EMI conditions have continual access to a suite of support services. In the EMA + EMI condition, participants are prompted to use the EMI(s) when responses to the EMA(s) indicate risk. All groups have access to recovery support as usual. The primary outcome is days of abstinence from alcohol and other drugs. Secondary outcomes are number of HIV risk behaviors and whether abstinence mediates the effects of EMA, EMI, or EMA + EMI on HIV risk behaviors.

Discussion

This project will enable the field to learn more about the effects of EMAs and EMIs on substance use disorders and HIV risk behaviors, an understanding that could potentially make treatment and recovery more effective and more widely accessible.

Trial registration

ClinicalTrials.gov, ID: NCT02132481. Registered on 5 May 2014.

Electronic supplementary material

The online version of this article (doi:10.1186/s13063-017-2096-z) contains supplementary material, which is available to authorized users.

iSurvive: An Interpretable, Event-time Prediction Model for mHealth

An important mobile health (mHealth) task is the use of multimodal data, such as sensor streams and self-report, to construct interpretable time-to-event predictions of, for example, lapse to alcohol or illicit drug use. Interpretability of the prediction model is important for acceptance and adoption by domain scientists, enabling model outputs and parameters to inform theory and guide intervention design. Temporal latent state models are therefore attractive, and so we adopt the continuous time hidden Markov model (CT-HMM) due to its ability to describe irregular arrival times of event data. Standard CT-HMMs, however, are not specialized for predicting the time to a future event, the key variable for mHealth interventions. Also, standard emission models lack a sufficiently rich structure to describe multimodal data and incorporate domain knowledge. We present iSurvive, an extension of classical survival analysis to a CT-HMM. We present a parameter learning method for GLM emissions and survival model fitting, and present promising results on both synthetic data and an mHealth drug use dataset.

Relapse prediction: A meteorology-inspired mobile model

Predicting relapses to binge drinking in non-dependent drinkers may now be possible with smartphones. Smartphones have been shown to help individuals reduce their drinking and may help binge drinkers accelerate that process. Predicting the weather has improved greatly over the past 50 years, but predicting a binge drinking episode may be less difficult. It is hypothesized that the number of factors with high predictive value for any particular individual may not be large. Collecting data over time, a smartphone should be able to learn which combination of factors has a high probability of leading to an episode of binge drinking.

A Smartphone Application Supporting Recovery from Heroin Addiction: Perspectives of Patients and Providers in China, Taiwan, and the USA

Smartphone-based interventions are increasingly used to support self-monitoring, self-management, and treatment and medication compliance in order to improve overall functioning and well-being. In attempting to develop a smartphone application (S-Health) that assists heroin-dependent patients in recovery, a series of focus groups (72 patients, 22 providers) were conducted in China, Taiwan, and the USA to obtain their perspectives on its acceptance and potential adoption. Data were analyzed according to the Diffusion of Innovation (DOI) theory of characteristics important to the adoption of innovation. Important to Relative Advantage, USA participants cited S-Health's potential ability to overcome logistical barriers, while those in China and Taiwan valued its potential to supplement currently limited services. In terms of Compatibility, participants across sites reported recovery needs and goals that such an application could be helpful in supporting; however, its utility during strong craving was questioned in China and Taiwan. Important factors relevant to Complexity included concerns about smartphone access and familiarity, individualization of content, and particularly in China and Taiwan, participants wanted assurance of privacy and security. The study results suggest a general acceptance, but also indicate cultural variations in access to therapeutic and other social support systems, legal repercussions of substance use, societal perceptions of addiction, and the role of family and other social support in recovery. Taking these factors into consideration is likely to increase diffusion as well as effectiveness of these smartphone-based interventions.

Technological aids for improving longitudinal research on substance use disorders

Background

There is a broad consensus that addictive behaviors tend to be chronic and relapsing. But for field studies of substance users, successfully tracking, locating, and following up with a representative sample of subjects is a challenge.

Methods

The purpose of this paper is to provide a general overview of how current technological aids can support and improve the quality of longitudinal research on substance use disorders. The review is grouped into four domains: (1) tracking and locating, (2) prompting/engaging, (3) incentivizing, and (4) collecting data.

Results & conclusions

Although the technologies described in this review will be modified or replaced over time, our findings suggest that incorporating some or all of these currently available approaches may improve research efficiency, follow-up rates, and data quality.

New Tools for New Research in Psychiatry: A Scalable and Customizable Platform to Empower Data Driven Smartphone Research

BACKGROUND

A longstanding barrier to progress in psychiatry, both in clinical settings and research trials, has been the persistent difficulty of accurately and reliably quantifying disease phenotypes. Mobile phone technology combined with data science has the potential to offer medicine a wealth of additional information on disease phenotypes, but the large majority of existing smartphone apps are not intended for use as biomedical research platforms and, as such, do not generate research-quality data.

OBJECTIVE

Our aim is not the creation of yet another app per se but rather the establishment of a platform to collect research-quality smartphone raw sensor and usage pattern data. Our ultimate goal is to develop statistical, mathematical, and computational methodology to enable us and others to extract biomedical and clinical insights from smartphone data.

METHODS

We report on the development and early testing of Beiwe, a research platform featuring a study portal, smartphone app, database, and data modeling and analysis tools designed and developed specifically for transparent, customizable, and reproducible biomedical research use, in particular for the study of psychiatric and neurological disorders. We also outline a proposed study using the platform for patients with schizophrenia.

RESULTS

We demonstrate the passive data capabilities of the Beiwe platform and early results of its analytical capabilities.

CONCLUSIONS

Smartphone sensors and phone usage patterns, when coupled with appropriate statistical learning tools, are able to capture various social and behavioral manifestations of illnesses, in naturalistic settings, as lived and experienced by patients. The ubiquity of smartphones makes this type of moment-by-moment quantification of disease phenotypes highly scalable and, when integrated within a transparent research platform, presents tremendous opportunities for research, discovery, and patient health.

Smartphone applications to reduce alcohol consumption and help patients with alcohol use disorder: a state-of-the-art review

Hazardous drinking and alcohol use disorder (AUD) are substantial contributors to USA and global morbidity and mortality. Patient self-management and continuing care are needed to combat these public health threats. However, services are rarely provided to patients outside of clinic settings or following brief intervention. Smartphone applications (“apps”) may help narrow the divide between traditional health care and patient needs. The purpose of this review is to identify and summarize smartphone apps to reduce alcohol consumption or treat AUD that have been evaluated for feasibility, acceptability, and/or efficacy. We searched two research databases for peer-reviewed journal articles published in English that evaluated smartphone apps to decrease alcohol consumption or treat AUD. We identified six apps. Two of these apps (A-CHESS and LBMI-A) promoted self-reported reductions in alcohol use, two (Promillekoll and PartyPlanner) failed to promote self-reported reductions in alcohol use, and two (HealthCall-S and Chimpshop) require further evaluation and testing before any conclusions regarding efficacy can be made. In summary, few evaluations of smartphone apps to reduce alcohol consumption or treat AUD have been reported in the scientific literature. Although advances in smartphone technology hold promise for disseminating interventions among hazardous drinkers and individuals with AUD, more systematic evaluations are necessary to ensure that smartphone apps are clinically useful.

Treatment Burden and Treatment Fatigue as Barriers to Health

Effective management of chronic diseases involves sustained changes in health behavior, which often requires substantial effort and patient burden. As treatment burden is associated with reduced adherence across several chronic conditions, its assessment and treatment are important clinical priorities. The balance between patient demands and capacity (e.g., coping resources) may be indexed by patients' subjective experience of treatment fatigue. We present a modified workload-capacity model that incorporates evidence that treatment fatigue may 1) be caused by increased workload due to treatment burden (e.g., intensity, complications) and 2) undermine adherence. Emerging technology-based interventions may be well-suited to reduce treatment burden, prevent treatment fatigue, and increase treatment adherence.

Building new computational models to support health behavior change and maintenance: new opportunities in behavioral research

Adverse and suboptimal health behaviors and habits are responsible for approximately 40 % of preventable deaths, in addition to their unfavorable effects on quality of life and economics. Our current understanding of human behavior is largely based on static "snapshots" of human behavior, rather than ongoing, dynamic feedback loops of behavior in response to ever-changing biological, social, personal, and environmental states. This paper first discusses how new technologies (i.e., mobile sensors, smartphones, ubiquitous computing, and cloud-enabled processing/computing) and emerging systems modeling techniques enable the development of new, dynamic, and empirical models of human behavior that could facilitate just-in-time adaptive, scalable interventions. The paper then describes concrete steps to the creation of robust dynamic mathematical models of behavior including: (1) establishing "gold standard" measures, (2) the creation of a behavioral ontology for shared language and understanding tools that both enable dynamic theorizing across disciplines, (3) the development of data sharing resources, and (4) facilitating improved sharing of mathematical models and tools to support rapid aggregation of the models. We conclude with the discussion of what might be incorporated into a "knowledge commons," which could help to bring together these disparate activities into a unified system and structure for organizing knowledge about behavior.

Successful Organizational Strategies to Sustain Use of A-CHESS: A Mobile Intervention for Individuals With Alcohol Use Disorders

Background

Mobile health (mHealth) services are growing in importance in health care research with the advancement of wireless networks, tablets, and mobile phone technologies. These technologies offer a wide range of applications that cover the spectrum of health care delivery. Although preliminary experiments in mHealth demonstrate promising results, more robust real-world evidence is needed for widespread adoption and sustainment of these technologies.

Objective

Our aim was to identify the problems/challenges associated with sustained use of an mHealth addiction recovery support app and to determine strategies used by agencies that successfully sustained client use of A-CHESS.

Methods

Qualitative inquiry assessed staff perceptions about organizational attributes and strategies associated with sustained use of the mobile app, A-CHESS. A total of 73 interviews of clinicians and administrators were conducted. The initial interviews (n=36) occurred at the implementation of A-CHESS. Follow-up interviews (n=37) occurred approximately 12 and 24 months later. A coding scheme was developed and Multiuser NVivo was used to manage and analyze the blinded interview data.

Results

Successful strategies used by treatment providers to sustain A-CHESS included (1) strong leadership support, (2) use of client feedback reports to follow up on non-engaged clients, (3) identify passionate staff and incorporate A-CHESS discussions in weekly meetings, (4) develop A-CHESS guidelines related to client use, (5) establish internal work groups to engage clients, and (6) establish a financial strategy to sustain A-CHESS use. The study also identified attributes of A-CHESS that enhanced as well as inhibited its sustainability.

Conclusions

Mobile apps can play an important role in health care delivery. However, providers will need to develop strategies for engaging both staff and patients in ongoing use of the apps. They will also need to rework business processes to accommodate the changes in communication frequency and style, learn to use app data for decision making, and identify financing mechanisms for supporting these changes.

Trends in telemedicine use in addiction treatment

Introduction

Telemedicine use in addiction treatment and recovery services is limited. Yet, because it removes barriers of time and distance, telemedicine offers great potential for enhancing treatment and recovery for people with substance use disorders (SUDs). Telemedicine also offers clinicians ways to increase contact with SUD patients during and after treatment.

Case description

A project conducted from February 2013 to June 2014 investigated the adoption of telemedicine services among purchasers of addiction treatment in five states and one county. The project assessed purchasers’ interest in and perceived facilitators and barriers to implementing one or more of the following telemedicine modalities: telephone-based care, web-based screening, web-based treatment, videoconferencing, smartphone mobile applications (apps), and virtual worlds.

Discussion and evaluation

Purchasers expressed the most interest in implementing videoconferencing and smartphone mobile devices. The anticipated facilitators for implementing a telemedicine app included funding available to pay for the telemedicine service, local examples of success, influential champions at the payer and treatment agencies, and meeting a pressing need. The greatest barriers identified were: costs associated with implementation, lack of reimbursement for telemedicine services, providers’ unfamiliarity with technology, lack of implementation models, and confidentiality regulations. This paper discusses why the project participants selected or rejected different telemedicine modalities and the policy implications that purchasers and regulators of addiction treatment services should consider for expanding their use of telemedicine.

Conclusions

This analysis provides initial observations into how telemedicine is being implemented in addiction services in five states and one county. The project demonstrated that despite the considerable interest in telemedicine, implementation challenges exist. Future studies should broaden the sample analyzed and track technology implementation longitudinally to help the research and practitioner communities develop a greater understanding of technology implementation trends and practices.

Exploring the use patterns of a mobile health application for alcohol addiction before the initial lapse after detoxification

How patients used Addiction-Comprehensive Health Enhancement Support System (A-CHESS)1, a mobile health intervention, while quitting drinking is worthy exploring. This study is to explore A-CHESS use patterns prior to the initial lapse reported after discharge from inpatient detoxification programs. 142 patients with alcohol addiction from two treatment agencies in the U.S. were included. A comprehensive set of A-CHESS use measures were developed based on a three-level system use framework and three A-CHESS service categories. In latent profile analyses, three A-CHESS system use patterns-inactive, passive, and active users-were found. Compared to the passive users (with the highest chance of the initial lapse), the active users (with the lowest chance of such behavior) participated more in online social activities, used more sessions, viewed more pages, and used A-CHESS longer. However, the chances of the initial lapse between A-CHESS user profiles were not statistically different. Implications of this finding were provided.