Using Smartphones to Provide Recovery Support Services

Nationwide cost estimates of 60 major illnesses place alcohol use disorders as the 2nd and drug use disorders as the 7th most costly health problems. Both are related to higher rates of HIV risk behaviors associated with the human immunodeficiency virus (HIV) and together they represent 3 of the top 10 modifiable behavioral causes of mortality in the US. Studies indicate that effective strategies for managing substance use disorders (SUD) must address their chronic and cyclical nature, of which HIV risk is a part. For many, substance use creates circumstances that increase HIV risk (e.g., needle use, trading sex) and for others, substance use is a coping mechanism for the negative impact of lifestyle choices (e.g. illegal activity, losing custody of their children, trading sex). Post-discharge monitoring, feedback and early re-intervention have become standard practice when managing numerous chronic conditions.10 Mobile technology has the potential to radically improve the odds of sustained recovery by providing tools for ongoing monitoring, assessment and access to recovery support interventions anytime and anywhere. This is particularly important for the 90% of those needing treatment that don't receive it. Therefore applying smartphone technology to provide recovery management anytime anywhere is potentially significant.

Using their smartphone-based suite of relapse-prevention interventions , Gustafson and colleagues11 conducted an RCT with a sample of 350 adults with alcohol use disorders discharged from 2 residential substance abuse treatment programs and followed them for 12 months. Patients self-initiated access to the interventions 24/7; the type of interventions ranged from being professionally supported EMIs like "ask an expert" which allowed participants to receive personal responses to their questions from addiction experts to a Panic Button which triggered automated reminders to the participant and, alerts to key people (e.g., counselor, sponsor, family member) who may reach out to the participant via phone or in person. Their system was designed to support recovery by promoting autonomous motivation, coping competence, and relatedness.12 Relative to participants assigned to recovery support services as usual, those receiving the support system reported significantly fewer heavy drinking days and were more likely to be abstinent from alcohol at the end of the study.13-15 In addition, this study demonstrated that providing recovery support via smart phones is feasible for this population.

Despite these encouraging outcomes, the full potential of providing 24/7 access to recovery support via smartphones may not yet have been realized. For example, while participants rated themselves on protective and risk factors weekly, opportunities to intervene "in the moments of need" were lost if the individuals did not self-initiate use of the application, ratings were subject to recall bias, and "teachable moments" in which participants could have cognitively linked the risk factors to their desire or actual use were unexplored. More frequent monitoring of current circumstances (vs. past generalized week) using methods like Ecological Momentary Assessment (EMA) may be better suited to this context as they can improve one's level of self-monitoring, provide additional opportunities to intervene in the moment of risk, minimize recall bias and provide participants with the opportunity to learn more about the relationship between current circumstances and their use.16,17 The primary goal of the proposed trial is to examine the effect of combining more frequent self-monitoring via EMAs and automated interventions via Ecological Momentary Interventions (EMIs) provided by smartphone, on days of abstinence from drugs and alcohol and HIV risk behaviors over 6 months following treatment discharge.

We will recruit a total of 400 participants in 4 cohorts of 100 people from sequential discharges (both planned or unplanned) over 2 months starting in months 7, 19, 31, 43 and follow them for 6 months post discharge. Using urn randomization, we will randomly assign participants at discharge in a 2 x 2 factorial design to receive EMAs only, EMIs only, combined EMA+EMI or control (no EMA or EMI). Post-randomization, participants assigned to the EMA, EMI and combined EMA/EMI groups will receive a smartphone for 6 months with 450 call minutes, 500 text messages and unlimited data plan monthly. Those in the EMA only and EMA/EMI combined group will be asked at 6 randomly selected times each day (183 days total) to record their past-30-minutes involvement in HIV risk behaviors and substance use, and their exposure to internal and external risk and protective factors and rate the extent to which these factors support their recovery, make them want to use drugs and/or alcohol. Those in the EMI only and EMA/EMI combined group will have continuous access to a suite of smartphone based support services (EMIs). To minimize interference with treatment, phones will be issued and trainings will be done after discharge. To minimize potential contamination across conditions, trainings for each condition will occur on separate days. During the month following treatment discharge \& training, participants in the three experimental groups will return to the research office 2 times a week for research staff to check participant's proficiency operating the phone and collect urine screens. To control for the effects of attention, the control group will also return to the research office 2 times a week for urine screens and a short survey on non-phone related topics. Research office visits and trainings will be done by Protocol Monitors (not interviewers). Research Interviewers blind to condition will conduct the discharge, 3 and 6 months post discharge interviews and urine tests. EMA and EMI utilization data will be electronically recorded immediately via a mobile/web application and used to track implementation.

The primary hypotheses are

H1 Random assignment to a) EMA (vs. not), b) EMI (vs. not), and c) their interaction will be associated with more days of abstinence from drugs and alcohol over the 6 months post discharge.

H2 Random assignment to a) EMA, b) EMI, and c) their interaction will be a associated with fewer HIV risk behaviors over the 6 months post discharge.

H3 Days abstinent at 3 months post discharge will mediate the effects of a) EMA, b) EMI and c) their interaction on HIV Risk behavior at 6 months post discharge.