Momentary Influences on Self-Regulation in Two Populations With Health Risk Behaviors: Adults Who Smoke and Adults Who Are Overweight and Have Binge-Eating Disorder

Abstract

IntroductionSelf-regulation has been implicated in health risk behaviors and is a target of many health behavior interventions. Despite most prior research focusing on self-regulation as an individual-level trait, we hypothesize that self-regulation is a time-varying mechanism of health and risk behavior that may be influenced by momentary contexts to a substantial degree. Because most health behaviors (e.g., eating, drinking, smoking) occur in the context of everyday activities, digital technologies may help us better understand and influence these behaviors in real time. Using a momentary self-regulation measure, the current study (which was part of a larger multi-year research project on the science of behavior change) used ecological momentary assessment (EMA) to assess if self-regulation can be engaged and manipulated on a momentary basis in naturalistic, non-laboratory settings.MethodsThis one-arm, open-label exploratory study prospectively collected momentary data for 14 days from 104 participants who smoked regularly and 81 participants who were overweight and had binge-eating disorder. Four times per day, participants were queried about momentary self-regulation, emotional state, and social and environmental context; recent smoking and exposure to smoking cues (smoking sample only); and recent eating, binge eating, and exposure to binge-eating cues (binge-eating sample only). This study used a novel, momentary self-regulation measure comprised of four subscales: momentary perseverance, momentary sensation seeking, momentary self-judgment, and momentary mindfulness. Participants were also instructed to engage with Laddr, a mobile application that provides evidence-based health behavior change tools via an integrated platform. The association between momentary context and momentary self-regulation was explored via mixed-effects models. Exploratory assessments of whether recent Laddr use (defined as use within 12 h of momentary responses) modified the association between momentary context and momentary self-regulation were performed via mixed-effects models.ResultsParticipants (mean age 35.2; 78% female) in the smoking and binge-eating samples contributed a total of 3,233 and 3,481 momentary questionnaires, respectively. Momentary self-regulation subscales were associated with several momentary contexts, in the combined as well as smoking and binge-eating samples. For example, in the combined sample momentary perseverance was associated with location, positively associated with positive affect, and negatively associated with negative affect, stress, and tiredness. In the smoking sample, momentary perseverance was positively associated with momentary difficulty in accessing cigarettes, caffeine intake, and momentary restraint in smoking, and negatively associated with temptation and urge to smoke. In the binge-eating sample, momentary perseverance was positively associated with difficulty in accessing food and restraint in eating, and negatively associated with urge to binge eat. While recent Laddr use was not associated directly with momentary self-regulation subscales, it did modify several of the contextual associations, including challenging contexts.ConclusionsOverall, this study provides preliminary evidence that momentary self-regulation may vary in response to differing momentary contexts in samples from two exemplar populations with risk behaviors. In addition, the Laddr application may modify some of these relationships. These findings demonstrate the possibility of measuring momentary self-regulation in a trans-diagnostic way and assessing the effects of momentary, mobile interventions in context. Health behavior change interventions may consider measuring and targeting momentary self-regulation in addition to trait-level self-regulation to better understand and improve health risk behaviors. This work will be used to inform a later stage of research focused on assessing the transdiagnostic mediating effect of momentary self-regulation on medical regimen adherence and health outcomes.Clinical Trial RegistrationClinicalTrials.gov, Identifier: NCT03352713.

Dynamic Structural Equation Models with Binary and Ordinal Outcomes in Mplus

“…The motivating data are an intensive longitudinal study supported by the National Institutes of Health's Science of Behavior Change initiative (Eisenberg et al, 2018;Nielsen et al, 2018;Scherer et al, 2022). Complete details about data collection and the research design are reported on the study's dedicated page on ClinicalTrials.gov.…”

Intensive longitudinal mediation in Mplus.

“…The motivating data are an intensive longitudinal study supported by the National Institutes of Health’s Science of Behavior Change initiative (Eisenberg et al, 2018; Nielsen et al, 2018; Scherer et al, 2022). Specific details about data collection and the research design are reported on the study’s dedicated page on ClinicalTrials.gov 1 .…”

“…We opted to use simulated data rather than the raw empirical data for two reasons. First, participants may be identifiable even from de-identified data because some recruitment was conducted in the local area around the lead institution’s campus (e.g., Scherer et al, 2022). Our goal for the tutorial was to make the materials as open as possible—including the data—which is facilitated by using synthetic data (Raghunathan, 2021; Reiter, 2004).…”

Intensive Longitudinal Mediation in Mplus

“…The motivating data are an intensive longitudinal study supported by the National Institutes of Health's Science of Behavior Change initiative (Eisenberg et al, 2018;Nielsen et al, 2018;Scherer et al, 2022). Specific details about data collection and the research design are reported on the study's dedicated page on ClinicalTrials.gov.…”

“…We opted to use simulated data rather than the raw empirical data for two reasons. First, participants may be identifiable even from de-identified data because some recruitment was conducted in the local area around the lead institution's campus (e.g., Scherer et al, 2022). Our goal for the tutorial was to make the materials as open as possibleincluding the datawhich is facilitated by using synthetic data (Ragunathan, 2021;Reiter, 2004).…”