The daily association between affect and alcohol use: A meta-analysis of individual participant data

Distinguishing Emotion Regulation Success in Daily Life From Maladaptive Regulation and Dysregulation

ACADEMIC ABSTRACT

This paper aims to motivate research on emotion regulation success in naturalistic settings. We define emotion regulation success as achieving one's emotion regulation goal and differentiate it from related concepts (i.e., maladaptive regulation and dysregulation). As goals vary across individuals and situations, it is insufficient to conceptualize emotion regulation success as maximizing positive affect and minimizing negative affect. Instead, emotion regulation success can be measured through novel approaches targeting the achievement of emotion regulation goals. In addition to utilizing novel data analytic tools (e.g., response surface analyses), future research can make use of informant reports and observing ambulatory behavior or physiology. Considering emotion regulation goals when measuring daily emotion regulation success has the potential to answer key questions about personality, development, and mental health.

PUBLIC ABSTRACT

People differ in how they want to feel in daily situations (e.g., excited) and why they want to feel that way (e.g., to make others feel better), depending on factors such as culture or age. Although people manage their emotions to reach these goals, most research assessing emotion regulation success has not taken individual goals into account. When assessing if people successfully regulate their emotions, most research in daily life has been focused on whether people feel more positive or less negative. To help study emotion regulation success in a more thoughtful and inclusive way, we propose a new approach to conceptualizing emotion regulation success that incorporates individual differences in what motivates people to regulate and discuss future research directions and applications.

Affect trajectories surrounding binge eating episodes and heavy drinking episodes in adults with binge-spectrum eating disorders

OBJECTIVE

The comorbidity of binge eating and heavy drinking (BE + HD) is concerning due to high prevalence and associated consequences. Affective pathways may maintain BE + HD, yet more micro-level research is needed. This study used ecological momentary assessment (EMA) to examine between-person and day-level relationships between positive and negative affect and binge eating or heavy drinking episodes in BE + HD.

METHODS

Participants (N = 53) were adults with binge-spectrum eating disorders who completed between 7 and 14 days of EMA prior to a treatment for binge eating.

RESULTS

Anxiety was highest on days with both binge eating and heavy drinking, while excitement and confidence were highest on days with only heavy drinking episodes for BE + HD. Global negative affect was relatively stable surrounding binge eating episodes. Guilt significantly increased prior to binge eating, and sadness significantly decreased following binge eating. Global positive affect significantly decreased prior to and stopped decreasing following heavy drinking episodes.

DISCUSSION

Results support binge eating being negatively reinforced by specific aspects of negative affect, while heavy drinking may be positively reinforced by global positive affect for individuals with BE + HD. Clinicians should incorporate interventions that focus on specific negative affect dimensions and that promote alternative rewarding activities besides heavy drinking.

Brief emotion regulation strategies to reduce alcohol craving: Mediating role of state difficulties in emotion regulation

OBJECTIVE

This study experimentally compared the effects of emotion regulation (ER) strategies on alcohol craving and examined the mediating effect of state difficulties in emotion regulation (S-DER) on the relationship between negative/positive emotion and alcohol craving.

METHOD

417 participants (76.74% women, Mage = 20.76 years) endorsing past-month heavy/binge drinking were randomly assigned to one of four ER conditions (positive reappraisal, distancing, distraction, and acceptance). Participants completed state assessments, including negative/positive emotion, S-DER, and alcohol craving, prior to (T0) and after (T1) engaging in a negative emotion induction task. Subsequently, participants completed an ER strategy task based on their assigned ER strategy condition and completed a third state assessment (T2).

RESULTS

Time had a significant quadratic effect on alcohol craving, such that craving increased from T0 to T1 and decreased from T1 to T2. There was no significant effect of ER strategy condition on craving. Change in S-DER mediated the relationship between the change in negative/positive emotion and the change in craving, with emotional modulation and emotional acceptance facets of S-DER dominating the mediating effect during negative emotion induction and ER strategy induction, respectively.

CONCLUSIONS

Results suggest interventions targeting S-DER's emotional modulation and acceptance facets could reduce acute craving when experiencing undesired emotions.

Effects of social drinking context on subjective effects, affect, and next-day appraisals in the natural environment

BACKGROUND

Drinking commonly occurs in social settings and may bolster social reinforcement. Laboratory studies suggest that subjective effects and mood are mechanisms through which the social context influences alcohol consumption. Ecological momentary assessment (EMA) may be useful for extending these findings to the natural environment. This pre-registered secondary analysis of EMA data investigated the influence of the social environment on: (1) stimulating and sedating subjective effects of alcohol, (2) contentedness and negative affect, and (3) next-day evaluations of the drinking occasion.

METHODS

Nontreatment seeking adults reporting past-month heavy drinking (N = 131; Mage = 28.09; 42% female) completed 7 days of EMA (in the morning, at random, and following drinking prompts), which included questions on their social context (drinking in the presence of known others or alone), contemporaneous stimulating and sedating effects, contentedness and negative affect, alcohol consumption, and next-day evaluations of a prior day's drinking event (how satisfying/pleasant was drinking). We used multi-level models in SAS 9.4 M7 software to examine relations among the variables.

RESULTS

Contemporaneous subjective effects (stimulating or sedating), negative affect, and contentedness did not significantly depend on the social context. For next-day evaluations of pleasure/satisfaction from drinking, context effects were dependent on consumption totals. As the total number of standard drinks consumed increased, recollections of pleasure/satisfaction were higher when drinking had occurred with others, relative to alone. At lower consumption totals, next-day evaluations did not appear to depend on social context.

CONCLUSIONS

When reported contemporaneously, subjective effects and affect do not appear dependent upon the presence of known others. However, heavier drinking events, relative to lighter drinking events, are appraised more favorably the following day when occurring within social contexts.

Mixed method examination of alcohol and suicidality among actively suicidal adults who engage in heavy episodic drinking

Suicide is a serious public health problem in the United States. Alcohol use has been substantially documented as a risk factors for suicide, yet it is unclear how alcohol is associated with suicidal ideation (SI) and behavior (SIB) at the event level. We examined the association between alcohol use and SI using a mixed methods approach that included daily assessments from 13 adults who engage in heavy episodic drinking with current SI and qualitative interviews among 12 of those adults. Participants were recruited on social media. Separate mixed effects logistic regression models indicated that individuals' alcohol use on a given day was associated with SI (OR = 1.37), and suicidal urges (OR = 1.41). Adjusting for repeated measures, the expected marginal mean for intensity of SI (EMM = 3.33) and urges (EMM = 2.94) were higher on days with reported drinking behavior than days without reported drinking (EMM = 2.68 and EMM = 2.62 respectively). Qualitative data indicated that the association between alcohol use and SIB is more complex than a single directionality. Instead, the association can be unidirectional, bidirectional, and/or dependent on factors including mental health and amount of alcohol consumed. Overall, these findings emphasize a need for integrated alcohol and SIB interventions while providing insight on possible daily, just-in-time adaptations.

The influence of negative urgency and mood inductions on alcohol cognitions

BACKGROUND

Negative urgency (NU), the tendency to act rashly during negative emotional states, is a robust risk factor for alcohol misuse that is posited to function in part through alcohol-related cognitions. Nonetheless, relatively little research has examined mood-based fluctuations in such cognitions, which could help to explain how the trait of NU translates to impulsive alcohol-related behaviors. We examined how NU impacted several alcohol cognitions (positive/negative alcohol expectancies, positive/negative alcohol valuations, and alcohol craving for positive/negative emotional reinforcement) before and after negative, neutral, or positive mood inductions. We hypothesized that NU would predict greater and more favorable endorsement of alcohol and its effects following negative (vs. positive or neutral) mood induction.

METHODS

Participants (N = 428) were southern-midwestern college students recruited for an online experiment. Following the provision of consent, participants rated NU and preinduction alcohol cognitions, and were then randomly assigned to one of three (negative, neutral, or positive) mood inductions; subsequently, postinduction alcohol-cognition ratings were immediately obtained. We conducted six robust multilevel linear models (one per DV) examining NU's influence on within-person changes in alcohol cognitions across each mood induction.

RESULTS

No three-way interactions were identified and only one two-way interaction involving NU was identified. There were main effects across mood induction conditions and time points for NU predicting greater endorsement of positive and negative alcohol outcome expectancies, and greater alcohol craving for positive and negative emotional reinforcement.

CONCLUSIONS

Greater NU predicts greater perceived likelihood of alcohol's effects, alongside greater desire for mood improvement from alcohol. The absence of three-way interactive effects indicates NU's influence on mood-dependent fluctuations in alcohol cognitions may manifest over longer timescales (e.g., months and years), involve alternative cognitive processes (e.g., drinking motives and implicit alcohol cognitions), and apply more broadly to desires for mood improvement than purely negative emotional reinforcement.

Subjective stress and any drinking during alcohol treatment: Disentangling within and between person autoregressive effects

Alcohol use has been shown to increase stress, and there is some evidence that stress predicts subsequent alcohol use during treatment for alcohol use disorder (AUD), particularly among females who are more likely to report coping-motivated drinking. Gaining a better understanding of the processes by which stress and alcohol use are linked during treatment could potentially inform AUD treatment planning. The current study aimed to characterize the association between stress and drinking during the course of AUD treatment and whether there were sex differences in these associations. Secondary data analyses of the COMBINE study (N = 1375; 69% male, 76.3% non-Hispanic and white, average age of 44.4 years) were conducted to examine self-reported perceived stress and alcohol consumption across 16 weeks of treatment for AUD using a Bayesian random-intercept cross-lagged panel model. There was stronger evidence for any alcohol use predicting greater than typical stress in subsequent weeks and less strong evidence for stress increasing the subsequent probability of alcohol use, particularly among males. For females, greater stress predicted subsequent drinking earlier in the treatment period, and a lower probability of subsequent drinking in the last week of treatment. Interventions might specifically focus on targeting reductions in stress following drinking occasions.

Age-related patterns in high-risk alcohol and cannabis use and their associations with positive and negative affect in young adulthood

OBJECTIVE

We examined age-varying associations between young adult simultaneous alcohol and marijuana/cannabis use (SAM) and heavy episodic drinking (HED) and positive and negative affect to inform harm reduction efforts.

METHODS

Young adults reporting past-year alcohol use (n = 556; ages 19-25) were recruited in a state where alcohol and nonmedical cannabis use was legal for those 21 +. Participants provided 24 repeated monthly assessments. Among those reporting past-month cannabis use on at least one survey, logistic time-varying effect models estimated (1) the age-varying prevalence of and associations between past-month SAM and HED and (2) age-varying unique associations of affect with SAM and HED.

RESULTS

There was a positive age-varying association between HED and SAM over time that was highest at age 19 (OR = 7.56), decreased until age 20.7 (OR = 3.39), increased until age 23.0 (OR = 4.85), and decreased until the association became non-significant by age 25. Negative affect was positively associated with SAM from ages 20.7 to 23.0, peaking at age 21.8 (OR = 1.36). Positive affect was positively associated with HED from ages 19.4 to 20.4 (peak OR = 1.25) and ages 22.5 to 24.5 (peak OR = 1.38). In contrast, positive affect was not uniquely associated with SAM nor negative affect with HED across ages 19-25.

CONCLUSIONS

While HED and SAM were positively associated throughout young adulthood and interventions could target them in tandem, their associations with affect suggest differential etiologic processes. Preventive intervention and harm reduction efforts should attend to psychological context in which these behaviors occur.

An experimental investigation into the effect of negative affect on the behavioral economic demand for alcohol

Influential theoretical models hypothesize that alcohol use is an especially potent reinforcer when used as a strategy to cope with negative affect. Although the evidence for this idea in observational data is weak, some experimental evidence suggests that the behavioral economic demand for alcohol increases immediately following a negative emotional event. Because existing studies testing the effect of negative mood inductions on the demand for alcohol have several methodological limitations and do not take inter- and intraindividual variability into account, we developed an improved experimental design to increase our confidence in any potential within-person effect of negative mood inductions on alcohol demand as well as to test whether this effect exhibits systematic inter- and intraindividual variability. We hypothesize that people will show a higher demand for alcohol following negative compared to neutral mood inductions and that this effect is stronger in heavy compared to light drinkers as well as stronger on days characterized by higher coping motives and negative urgency. Three hundred twenty college students will complete the alcohol purchase task (APT) after being subjected to 100 mood inductions (six negative, six neutral) on 20 separate days. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Daily level examination indicates that positive affect, but not negative affect, is associated with alcohol and marijuana use among adolescents and young adults

OBJECTIVE

Although there is growing evidence that alcohol use at the daily level is associated with positive but not negative affect, results are mixed when examining marijuana use and simultaneous alcohol and marijuana (SAM) use (i.e., use with overlapping effects). In addition, less is known about these daily level associations among diverse samples of adolescents and young adults. The present study will address these gaps.

METHOD

Participants (N = 1,006; 57.3% female, 43.1% White, Non-Hispanic, ages 15-25; Mage = 20.0) were part of a study on substance use that consisted of a 3-week ecological momentary assessment (EMA) burst design (eight surveys per week, up to 2×/day) that was repeated quarterly over a 12-month period.

RESULTS

Within-person results indicated that on days with elevated positive affect, participants reported consuming more drinks, whereas positive affect was not significant for hours high from marijuana. In addition, on days with elevated negative affect, participants reported fewer hours high from marijuana. No association was found between negative affect and number of drinks. Finally, within-person results indicated that on alcohol or marijuana days with elevated positive affect, individuals were more likely to report SAM use. There was no association found between negative affect and SAM use.

CONCLUSIONS

Results have implications such that in-the-moment interventions for alcohol and SAM use may be more salient when individuals have higher positive affect than average, whereas such interventions may be more relevant for marijuana use when negative affect is lower compared to average levels. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

The role of impulsivity in the relationship between affect and alcohol consumption in young adults

BACKGROUND

Theoretical models of alcohol use posit that individuals consume alcohol to ameliorate negative affect or to heighten positive affect. It is important, however, to consider the influence of factors that may determine an individual's tendency to consume excessive amounts of alcohol under positive and negative circumstances. Thus, the current study examined a large sample of young adults to clarify whether positive and negative affect predict total alcohol consumption on drinking days and whether facets of impulsivity moderate these relationships.

METHODS

Six-hundred ninety-three young adults (Mage  = 19.71 years, SD = 2.04; female = 62.9%) completed the Behavioral Inhibition System/Behavioral Activation System (BIS/BAS) scales at baseline followed by daily measures of positive and negative affect and self-reported alcohol use for 13 days. Generalized linear mixed models were specified to assess the role of pre-consumption affect on total drinks consumed across drinking days and to determine the moderating effect of each BIS/BAS subscale.

RESULTS

Participants were significantly more likely to drink in greater quantities on occasions preceded by higher positive affect but not negative affect. While fun-seeking positively predicted total drinks consumed, there were no significant interaction effects between the BIS/BAS subscales and affect on total drinks consumed.

CONCLUSIONS

These findings challenge existing affect regulation models and have implications for ecological momentary interventions aimed at addressing hazardous drinking behaviors.

Disaggregating within- and between-person effects of affect on drinking behavior in a clinical sample with alcohol use disorder

OBJECTIVE

The goal of the current study was to better understand affect-drinking relations among those diagnosed with an alcohol use disorder (AUD), as recent meta-analytic work suggests that daily negative affect may not universally predict subsequent alcohol consumption in those nondependent on alcohol. Specifically, we investigated the between- and within-person effects of positive and negative affects on drinking.

METHOD

Participants (n = 92) who met AUD diagnostic criteria completed a 90-day daily assessment of drinking behavior and positive and negative affects.

RESULTS

Time-lagged multilevel modeling revealed that within-person elevations in negative affect predicted increased odds and quantity of drinking later in the day. Relations between positive affect and drinking were nonsignificant.

CONCLUSIONS

These findings are in contrast to recent meta-analytic findings and highlight the complexity of affect-drinking relations among those diagnosed with AUD. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Event-level positive and negative reinforcement risk factors for alcohol use: Moderation by individual-level alcohol consequences and post-traumatic stress disorder symptom severity

BACKGROUND

The multistage model of drug addiction posits that risk processes contributing to alcohol use change as individuals develop alcohol use disorders. However, few studies have tested this theory outside of the lab or at the event level. We assessed whether event-level associations between positive reinforcement (e.g., positive affect, sociability expectancies) and negative reinforcement risk factors (e.g., negative affect, tension reduction expectancies) and same-/next-day alcohol consumption varied as a function of an individual's level of alcohol consequences. Given elevated alcohol use consequences among individuals with post-traumatic stress disorder (PTSD) and disruptions in reward processing that affect this population, we also tested whether these processes differed based on the presence and severity of PTSD.

METHODS

We used data from a 30-day ecological momentary assessment study with 174 undergraduate women who regularly engaged in heavy episodic drinking. A majority (78%) of the sample had experienced sexual assault and 44% had current PTSD. Analyses used Bayesian multilevel structural equation modeling with diffuse (non-informative) priors. We used markov chain monte carlo (MCMC) algorithms to generate a series of 10,000 random draws from the multivariate posterior distribution of our sample for each model.

RESULTS

Results partially supported the multistage model. Event-level negative reinforcement risk factors only predicted more alcohol consumption among individuals who experienced more alcohol consequences. Findings for positive reinforcement risk factors were partially consistent with hypotheses. Overall, findings appear to operate similarly across PTSD symptom severity.

CONCLUSIONS

Results suggest that interventions for heavy episodic drinking could benefit from attending to an individual's level of alcohol consequences. For example, preventive interventions for individuals who tend to experience few consequences may benefit more from addressing positive reinforcement risk factors, while treatment interventions for those who experience more consequences may benefit from attending to both positive and negative reinforcement.

Within-person associations between psychological and contextual factors and lapse incidence in smokers attempting to quit: A systematic review and meta-analysis of ecological momentary assessment studies

AIMS

When attempting to stop smoking, discrete smoking events ('lapses') are strongly associated with a return to regular smoking ('relapse'). No study has yet pooled the psychological and contextual antecedents of lapse incidence, captured in ecological momentary assessment (EMA) studies. This systematic review and meta-analysis aimed to synthesize within-person psychological and contextual predictor-lapse associations in smokers attempting to quit.

METHODS

We searched Ovid MEDLINE, Embase, PsycINFO and Web of Science. A narrative synthesis and multi-level, random-effects meta-analyses were conducted, focusing on studies of adult, non-clinical populations attempting to stop smoking, with no restrictions on setting. Outcomes were the association between a psychological (e.g. stress, cravings) or contextual (e.g. cigarette availability) antecedent and smoking lapse incidence; definitions of 'lapse' and 'relapse'; the theoretical underpinning of EMA study designs; and the proportion of studies with pre-registered study protocols/analysis plans and open data.

RESULTS

We included 61 studies, with 19 studies contributing ≥ 1 effect size(s) to the meta-analyses. We found positive relationships between lapse incidence and 'environmental and social cues' [k = 12, odds ratio (OR) = 4.53, 95% confidence interval (CI) = 2.02, 10.16, P = 0.001] and 'cravings' (k = 10, OR = 1.71, 95% CI = 1.34, 2.18, P < 0.001). 'Negative feeling states' was not significantly associated with lapse incidence (k = 16, OR = 1.10, 95% CI = 0.98, 1.24, P = 0.12). In the narrative synthesis, negative relationships with lapse incidence were found for 'behavioural regulation', 'motivation not to smoke' and 'beliefs about capabilities'; positive relationships with lapse incidence were found for 'positive feeling states' and 'positive outcome expectancies'. Although lapse definitions were comparable, relapse definitions varied widely across studies. Few studies explicitly drew upon psychological theory to inform EMA study designs. One of the included studies drew upon Open Science principles.

CONCLUSIONS

In smokers attempting to stop, environmental and social cues and cravings appear to be key within-person antecedents of smoking lapse incidence. Due to low study quality, the confidence in these estimates is reduced.