Brian Suffoletto, MD, MS, Assistant Professor of Emergency Medicine, Department of Emergency Medicine, University of Pittsburgh
Inbal Nahum-Shani, PhD, Research Associate Professor, Institute for Social Research, Co-Director, Data-Science for Dynamic Decision-Making Lab, University of Michigan
J. Douglas Coatsworth, PhD, Colorado State University
Title: mHealth and the Science of Intervention: Conceptual and Clinical Opportunities and Challenges
Session Introduction: The goal of this symposium session is to present mHealth conceptual and clinical applications that address public health issues within the context of participants’ natural settings and in real-time. The symposium will highlight the unique methods, technological issues, and creative analytic approaches. Case studies will be used to stimulate questions and discussion regarding the advantages, opportunities, and challenges of mHealth applications and the critical need for theories and scientific models that inform mHealth research design.
The first presentation, “Just-In-Time Adaptive Interventions: Scientific Motivation and Key Components” will review the key elements of Just-In-Time Adaptive Interventions and propose scientific questions concerning the development and future directions for research.
The second presentation, “Using Digital Communication to Support Situational Self-Regulation in Young Adults” will examine in detail the effects of mHealth delivered drinking limit goal commitment on event-level alcohol consumption.
The third presentation, “Experiments with mHealth interventions across populations and settings: Opportunities & challenges” will highlight the adaptation process of in-person substance use interventions as well as results of multiple trials and discuss opportunities and challenges.
Discussant: An experienced prevention scientist will provide critical reflections on the presentations and will stimulate audience participation.
1st Paper within Organized Paper Symposium
Title: Just-In-Time Adaptive Interventions: Scientific Motivation and Key Components
Author: Inbal (Billie) Nahum-Shani
The use of smartphones to prevent and support the treatment of addictions is increasing rapidly. Mobile and wearable devices facilitate the delivery of Just-In-Time Adaptive Interventions (JITAIs)—an intervention design in which the timing and type of support are adapted over time to an individual’s state and context. JITAIs are motivated to address conditions that change rapidly, unexpectedly and ‘in the wild’ (i.e., outside of standard treatment settings). Despite the increasing use of JITAIs, a major gap exists between the growing technological capabilities for delivering JITAIs and research on the development and evaluation of these interventions. Scientific models typically used to inform intervention development often fail to provide the dynamic theory necessary for guiding when and how to intervene so as to help individuals address risk factors (e.g., stress) or capitalize on protective factors (e.g., social support) as they occur in real-life.
Here, we will review the key elements of JITAIs, using an example of a stress-management JITAI – Sense2stop—to guide the discussion. Currently under development, Sense2stop is designed to help individuals attempting to quit smoking regulate real-time experiences of stress in order to prevent stress-induced lapses. Open scientific questions concerning the development of this JITAI will be reviewed, and directions for future research aiming to advance the science of JITAIs for supporting stress management will be discussed.
It is critical that researchers develop sophisticated and nuanced scientific models capable of guiding the construction of JITAIs. Particular attention has to be given to better understanding how to promote engagement in interventions that are delivered ‘in the wild’ given that in these settings multiple demands compete for an individual’s time, attention and effort.
2nd Paper within Organized Paper Symposium
Title: Using Digital Communication to Support Situational Self-Regulation in Young Adults
Author: Brian Suffoletto
Introduction: Successful behavior change involves committing to goals, actively striving to reach them, and planning how to overcome potential obstacles. These self-regulation processes are particularly difficult in young adults. Our group has spent the past ten years developing text message interventions that support situational self-regulation processes to limit alcohol consumption in non-treatment seeking young adults. We will review past trial findings and describe a model for situational self-regulation of alcohol consumption in young adults.
Methods: We will examine in detail the effects of drinking limit goal commitment on event-level alcohol consumption. In this study, we included 482 young adults (aged 18-25) enrolled in 3 trials who were exposed to text message interventions including pre-drinking goal commitment (GC) prompts to limit number of drinks consumed, tailored feedback on goal commitment and protective behavioral strategies. We analyzed ecological momentary assessment (EMA) data from 1600 unique occasions where event-level GC was prompted and subsequent alcohol consumption was collected. We used logistic models accounting for clustering within individuals and autocorrelation, and adjusted for sex, race, alcohol use severity and trial to determine the strength of association between GC and binge drinking (i.e. 4/5 or more drinks for women/men).
Results: Drinking limit goals (GC+) were made 57.6% of the time they were prompted. GC+ was associated with lower odds of same day binge drinking [Odds Ratio(OR)=0.61; 95% CI 0.44, 0.84]. Black race and higher alcohol use severity were associated with higher odds of binge drinking. Compared to interventions using static GC prompts, adaptive GC prompts were associated with higher rates of commitment and lower odds of binge drinking [OR=0.27; 95% CI 0.17, 0.42].
Conclusions: Providing adaptive goal commitment prompts via digital communication technology can support situational self-regulation in young adults. We discuss next steps and key challenges.
3rd Paper within Organized Paper Symposium
Title: Experiments with mHealth interventions across populations and settings: Opportunities & challenges
Author: Michael Mason
Introduction: While substantial scientific progress has been made regarding substance use interventions, the limits of human-delivered approaches for public health issues such as substance use are unmistakable. There simply are not enough evidence-based interventions delivered with high fidelity to meet the need to prevent adolescent and young adult substance use. To address both the need for scaling up interventions and combining effective programs to reach underserved populations, mobile interventions, such as text messaging, have emerged with the promise of bridging some of the critical gaps in the field. Beyond the cost effectiveness of mHealth approaches compared to traditional interventions, advantages include increased access, privacy, technological/platform familiarity, personalization, and these interventions are delivered with 100% fidelity. In response to these issues, we developed and tested a text-based intervention and conducted five randomized clinical trials with over 400 adolescents and young adults. Development, evolution, and results of our intervention, Peer Network Counseling-txt (PNC-txt), will be presented.
Methods: Building upon our empirical work that described the social (close friends’ risk and protective behaviors) and the spatial (activity space- routine locations) effects on youth substance use, we developed PNC-txt. PNC-txt applies Motivational Interviewing principles but uniquely targets close peer relationships as a mechanism of change. PNC-txt focuses on place-based peer relations, which raises awareness of both peers and place as protective and risk-enhancing. Beginning with a proof of concept RCT addressing young adult hazardous drinking, we conducted RCTs targeting adolescent tobacco use, young adult cannabis use, and adolescent opioid misuse that includes a parent complimentary intervention. Settings vary from adolescent medicine clinics, federally qualified health care settings, to college campuses.
Results: Attitudinal, behavioral, and biochemical results are summarized along with participant engagement, satisfaction and feedback. A current RCT is briefly described as a model of our ongoing experimentation and refinement of PNC-txt.
Conclusions: Continued investigations into how to address inherent platform challenges, leverage unique opportunities, and how to integrate mHealth interventions into larger systems of care are warranted.