Search

Alcohol use is one of the leading risk factors for death and disability (GBD 2016 Alcohol Collaborators, 2018). Loss of control over alcohol consumption has been linked to decreased cognitive control (Beck et al., 2012; Koob & Volkow, 2010; Stavro et al., 2013; Gan et al Biol Psychiatry 2014)—a construct that includes, for example, the ability to inhibit unhealthy impulses, the ability to avoid excessive risk, and the ability to focus on one’s goals when distracted (Diamond, 2013; Powell et al., 2017). So far, this research has mostly focused on interindividual differences in cognitive control and alcohol consumption by studying the relationship cross-sectionally in the laboratory setting. Recent studies indicate, however, that cognitive control fluctuates significantly within individuals and that these fluctuations can predict changes in health behaviors. In the domain of eating, for example, Powell et al. (2017) showed that within-subject changes in inhibitory control can predict daily changes in snacking behavior. In patients with substance use disorder, Konova et al. (2020) showed that fluctuations in risky decision making can predict prospective opioid use. Given these findings and that changes in control over alcohol consumption cannot sensibly be tracked inside the laboratory (Ekhtiari et al., 2017; Sobell et al., 2000), it is, thus, crucial to understand how changes in cognitive control within individuals relate to changes in their real-life alcohol consumption. The overall goal of this study is to determine whether experimental measures of cognitive control and decision making relate to real-life changes in drinking behavior (Heinz et al., 2020). The key variables of interest are repeatedly collected on a smartphone, which enables us to address our main question of how within-subject changes in cognitive control and decision-making relate to variability in drinking behavior as assessed in real life. As an additional goal, we will also assess whether cognitive control protects against risk factors that are known to increase drinking—such as stress (Spanagel et al., 2014). To measure drinking behavior, participants indicate their alcohol consumption once every two days over a period of one year using Ecological Momentary Assessment (EMA) on the smartphone. Over the same period, participants also complete four behavioral tasks on the smartphone that measure cognitive control and decision making—two times at baseline and then once every month (14 times in total). We will assess the relationship between task variables and alcohol consumption separately in two datasets: The baseline data (here we will focus on between-subject effects) and the longitudinal data (here we will focus on both between and within-subject effects). All assessments will be done in two samples, one initial sample (N = 300) to test a priori hypotheses and explore the data and a second, independent sample (300 < N < 600) to replicate our findings (see Table 1). Additional objectives will be addressed: firstly, we will assess the validity of the behavioral tasks by comparing them to independent measures obtained in the lab collected in the same individuals as part an overarching project (the collaborative research center on “Loosing and regaining control over drug intake”; Heinz et al., 2020). Secondly, we will also assess the tasks’ internal consistency and test-retest reliability both in baseline and longitudinal datasets. Table 1 Sample sizes Sample Initial sample Replication sample Dataset Baseline (2 x 4 tasks) N = 300 300 < N < 600 Longitudinal (12 x 4 tasks) N = 300 300 < N < 600 In sum, this pre-registration has three objectives: 1) To assess the relationship between cognitive control and drinking at baseline (between-subject effects) 2) To assess the relationship between cognitive control and drinking longitudinally (between- and within-subject effects). 3) To assess the validity and reliability of the behavioral tasks. Analysis Plan Sample sizes, stopping rule, and state of data collection The initial sample will consist of 300 participants. The replication sample will consist of up to 600 participants (because the overarching project aims to collect up to N=900, Heinz et al., 2020; see Table 1). Should this sample size not be reached, we will analyze the replication sample on 30.06.2023 (end of the first funding period of the overarching project). More data might be collected beyond 30.06.2023 should the funding of the overarching project be extended for a second funding period. For the initial sample, baseline data and partial longitudinal data from 290 participants have already been collected. For the replication sample, no data has been collected. We have not yet begun to analyze any of the task data. Hypotheses Note that all analyses will be run on two independent samples: The initial sample and on the replication sample. Baseline hypotheses will be tested using Model 1 and longitudinal hypotheses will be tested using Model 2 (for model specification, see significance tests section). For the replication sample, additional hypotheses—based on analyses of the initial sample—might be added to this preregistration at a later stage. Hypotheses: baseline dataset C1. Baseline differences in inhibitory control relate negatively to drinking. C2. Baseline differences in risk aversion for gains relate negatively to drinking. C3. Baseline differences in risk aversion for losses relate positively to drinking. C4. Baseline differences in loss aversion relate negatively to drinking. C5. Baseline differences in sampling biases relate negatively to drinking. C6. Baseline differences in working memory relate negatively to drinking. C7. Baseline differences in working memory moderate stress-related increase in drinking behavior such that positive relation between stress and drinking is stronger when working memory is low. Hypotheses: longitudinal dataset L1. Changes in inhibitory control relate negatively to real-life changes in drinking. L2. Changes in risk aversion for gains relate negatively to real-life changes in drinking. L3. Changes in risk aversion for losses relate positively to real-life changes in drinking. L4. Changes in loss aversion relate negatively to real-life changes in drinking. L5. Changes in sampling biases relate negatively to real-life changes in drinking. L6. Between-subject differences in working memory moderate the effects of changes in inhibitory control and risk aversion on changes in drinking. Specifically, the expected positive and negative relationships between changes in inhibitory control as well as in risk aversion (see L1, L2, and L3) and drinking are stronger when working memory is low. L7. Between-subject differences in working memory moderate the effects of stress on drinking such that the positive relationship between stress and drinking is stronger when working memory is low. Predictor variables Working memory will be operationalized as working memory capacity (WMC) measured with a working memory task (McNab et al, 2015; see Table 2). Inhibitory control will be operationalized as stop signal reaction times (SSRTs) measured with a response inhibition (stop signal) task (Smittenaar et al., 2015). Variables related to risk aversion for gains/losses (PCG/PCL) will be operationalized as the percentage certain choices participants take in a risk-taking task (on gain/loss trials; Rutledge et al., 2014). Another specific aspect of risk aversion will be operationalized by the information sampling bias (SB), defined as how much participants oversample information (relative to optimal) in an information sampling task (Hunt et al., 2016). A variable related to loss aversion (PCM) will be operationalized as the percentage certain outcomes participants choose on mixed trials of the risk-taking task. Stress will be operationalized daily stress in the month following the behavioral tasks, as measured by the individual-stress-questionnaire (collected at the same time as the drinking measures; Perrez et al., 2000). Behavioral tasks will be completed by participants once every four weeks. Note, however, that the duration between tasks performances may vary according to sampling rule. This is defined as the following: Before the first longitudinal session, participants chose on which weekday and during which time they are least likely to drink. Every four weeks they are then alerted to complete the tasks at the predefined time. If at that time, they are intoxicated, hungover, or miss the alert, they will be triggered again at the pre-defined time on the next day. In this case, a button will also become available allowing participants to start the tasks without having to wait for the alarm. Should participants delay the tasks the next session will nevertheless start four weeks after the initial (missed) trigger. Aggregation. In the baseline analyses, we will take the mean of the outcomes from the two baseline measures of each behavioral task. We will, in addition, repeat analyses separately for each measurement. According to Neuhaus and Kalbfleisch (1998), variables in longitudinal datasets can have distinct between-subject and within-subject effects, which are not distinctly captured by standard mixed models. As our hypotheses involving WMC only predict effects based on between-subject differences, we will, therefore, take the by-participant mean WMC in the longitudinal data. Centering. In the between-subject analyses, all seven predictors (SSRT, PCG, PCL, PCM, SB, stress, WMC) will be centered around the grand mean. In the within-subject analysis, six predictors (SSRT, PCG, PCL, PCM, SB, stress) will be centered around the participant mean. WMC (which is aggregated by participant; see aggregation), will instead be centered around the grand mean. Outliers and exclusions. Across all behavioral tasks, trials with unrealistically low reaction times (

Psychological stress and physical activity are interrelated, constituting a relevant association to human health, especially in children. However, the association’s nature remains elusive, i.e., why psychological stress predicts both decreased and increased physical activity. To test whether effects vary as a function of the level of analyses, we derived intensive longitudinal data via accelerometers and stress questionnaires from 74 children across 7 days as they went about their daily routines (\(\it n\) = 513 assessments). Multilevel modelling analyses revealed that between children, higher psychological stress predicted decreased physical activity (standardized beta coefficient = −0.14; \(\it p\) = 0.046). Concurrently, within those children, higher psychological stress predicted increased physical activity across days (standardized beta coefficient = 0.09; \(\it p\) = 0.015). Translated to practice, children who experienced more stress than others moved less, but children were more active on days when they experienced heightened stress. This suggests that the analyses level is crucial to the understanding of the association between psychological stress and physical activity and should be considered to receive unequivocal results. If replicated, e.g., including high-frequency sampling and experimental manipulation in everyday life for in-depth insights on underlying mechanisms and causality, our findings may be translated to individually tailored (digital) prevention and intervention strategies which target children's distress-feelings despite impairing their heightened physical activity in stressful situations and identify tipping points of chronic stress phases. Therefore, we especially call for more intensive longitudinal data approaches to tackle thus far neglected within-subject issues in the field of physical activity, sport and exercise research.

Supplemental material, sj-docx-1-heb-10.1177_10901981221090153 for Urban-Rural Differences in Children’s and Adolescent’s Physical Activity and Screen-Time Trends Across 15 Years by Carina Nigg, Christoph Weber, Jasper Schipperijn, Markus Reichert, Doris Oriwol, Annette Worth, Alexander Woll and Claudia Niessner in Health Education & Behavior

Importance Alcohol consumption (AC) leads to death and disability worldwide. Ongoing discussions on potential negative effects of the COVID-19 pandemic on AC need to be informed by real-world evidence. Objective To examine whether lockdown measures are associated with AC and consumption-related temporal and psychological within-person mechanisms. Design, Setting, and Participants This quantitative, intensive, longitudinal cohort study recruited 1743 participants from 3 sites from February 20, 2020, to February 28, 2021. Data were provided before and within the second lockdown of the COVID-19 pandemic in Germany: before lockdown (October 2 to November 1, 2020); light lockdown (November 2 to December 15, 2020); and hard lockdown (December 16, 2020, to February 28, 2021). Main Outcomes and Measures Daily ratings of AC (main outcome) captured during 3 lockdown phases (main variable) and temporal (weekends and holidays) and psychological (social isolation and drinking intention) correlates. Results Of the 1743 screened participants, 189 (119 [63.0%] male; median [IQR] age, 37 [27.5-52.0] years) with at least 2 alcohol use disorder (AUD) criteria according to the Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) yet without the need for medically supervised alcohol withdrawal were included. These individuals provided 14 694 smartphone ratings from October 2020 through February 2021. Multilevel modeling revealed significantly higher AC (grams of alcohol per day) on weekend days vs weekdays (β = 11.39; 95% CI, 10.00-12.77; P < .001). Alcohol consumption was above the overall average on Christmas (β = 26.82; 95% CI, 21.87-31.77; P < .001) and New Year’s Eve (β = 66.88; 95% CI, 59.22-74.54; P < .001). During the hard lockdown, perceived social isolation was significantly higher (β = 0.12; 95% CI, 0.06-0.15; P < .001), but AC was significantly lower (β = −5.45; 95% CI, −8.00 to −2.90; P = .001). Independent of lockdown, intention to drink less alcohol was associated with lower AC (β = −11.10; 95% CI, −13.63 to −8.58; P < .001). Notably, differences in AC between weekend and weekdays decreased both during the hard lockdown (β = −6.14; 95% CI, −9.96 to −2.31; P = .002) and in participants with severe AUD (β = −6.26; 95% CI, −10.18 to −2.34; P = .002). Conclusions and Relevance This 5-month cohort study found no immediate negative associations of lockdown measures with overall AC. Rather, weekend-weekday and holiday AC patterns exceeded lockdown effects. Differences in AC between weekend days and weekdays evinced that weekend drinking cycles decreased as a function of AUD severity and lockdown measures, indicating a potential mechanism of losing and regaining control. This finding suggests that temporal patterns and drinking intention constitute promising targets for prevention and intervention, even in high-risk individuals., Zweitveröffentlichungen der Universität Potsdam : Humanwissenschaftliche Reihe; 805