Seventeen percent of 12th graders reported past 2-week binge alcohol use in 2020 , defined as five or more drinks on the same occasion on at least 1 day in the past 30 days in males . Adolescents tend to drink more alcohol per occasion, yet less frequently than adults . Binge drinking may have serious consequences on adolescent health, including overdose, fatal injuries, and motor vehicle accidents, and long term impacts on families, other students, and the general community . In addition, binge drinking has potential detrimental impacts on brain development, cognition, mood, and school performance . One in five teens suffer from depression, over 30% suffer anxiety symptoms, and up to half of those endorsing internalizing symptoms also endorse comorbid substance use, including alcohol . Understanding youth binge drinking and its relation with co-morbid internalizing symptoms is therefore a critical problem that may affect adolescents throughout development. Persistent binge drinking, depression, and anxiety are associated with a variety of poor health outcomes that ultimately affect both quality and quantity of life . Rates of adolescents with depressive and/or anxiety symptoms in the United States have sharply increased since 2012 , and have coincided with decreases in alcohol consumption . As such, understanding the evolving relation between adolescent internalizing symptoms and developing binge drinking behaviors remains crucial in determining developmentally informed targets for prevention and intervention of substance-related health risks among youth. Theory has suggested bidirectional links between depression/anxiety and alcohol behaviors throughout adolescence .
For instance,cannabis grow equipment the self-medication model for negative affect and alcohol use proposes that because both anxiety and depressed mood produce aversive negative mood states, adolescents may develop coping motives for alcohol use via attempts to reduce negative affect symptoms through drinking. Over the long term, this behavior may lead to development of increasingly heavier use and delayed-onset alcohol use disorder by means of negative reinforcement. This model has been conceptually shared by several explanatory models for the development of alcoholism and has been supported by evidence including patients self-reporting drinking as a way of coping with their anxiety , especially in those suffering from phobias . On the other hand, substance-induced negative affect models propose that anxiety and depression develop because of persistent, heavy alcohol use . Alcohol misuse can lead to several work, school, and relationship-related difficulties, and internalizing symptoms can result from difficulties in each of these areas. The development of alcohol use disorder occurs over the course of many recurring episodes of excessive and frequent drinking, and withdrawal may cause neural changes that lead to and/or exacerbate negative mood states. Over time, repeated recurring episodes may result in increased neural adaptation that may make a person who drinks alcohol more vulnerable to developing internalizing symptoms . A number of clinical studies have demonstrated that people who drink alcohol heavily that have recently stopped drinking experience an increase in anxiety, panic, and/or low mood, as well as symptoms of autonomic hyperactivity during an extended withdrawal period . However, there have also been a number of recent studies that suggest that it is possible there are no significant associations between anxiety, depression, and binge drinking . Despite supporting theory, evidence demonstrating links between internalizing symptoms and alcohol use has been mixed, with several systematic reviews demonstrating modest associations between depression and alcohol use and minimal-to-no relation between anxiety and use .
For instance, Hussong and colleagues provided a systematic review of longitudinal studies testing the relation between negative affect symptoms and substance use controlling for externalizing factors. They found that while there is some consistent evidence of a link between depression and substance use, only 5 out of 61 studies reviewed found a positive unique association between anxiety and use, 6 found a negative association, and the remaining 52 found no relation. A more recent meta-analysis examined 97 associations across 51 studies testing the link between anxiety and alcohol use disorders. They found inconsistent evidence of this link for binge drinking or drinking frequency/quantity and no clear association between generalized anxiety and alcohol use disorder . In light of this mixed evidence, methodological considerations have been noted that may clarify the relation between internalizing symptoms and binge drinking in adolescence . First, because this relation may be confounded by other between-person factors characterizing risk, within-person designs better accounting for these factors that may reduce bias in estimating the link between internalizing symptoms and binge drinking . For instance, the common-factor model of anxiety and alcohol use disorders hypothesizes that no direct relations exist between these two conditions, and may instead be explained by confounding variables . Studies that have modeled such variables explicitly have shown that the internalizing-alcohol use link may be explained by environmental contexts [e.g., childhood and family factors, prior substance dependence, comorbid depression, and peer affiliations; ], trait-level sensitivity to anxiety , and genetic contributors . While these studies have controlled for common factors at the between-person level, alternative approaches can distinguish between- from within-person effects by relegating the influence of these factors to random-effects components of a statistical model . Utilizing such methods to explicate between- versus within-person effects may help partition the influence of these common factors from the link between internalizing symptoms and binge behaviors at the individual level.
Relatedly, few have assessed bidirectional links between internalizing and binge drinking factors . Because substance use may lead to increased risk of internalizing symptoms, assessing these associations bi-directionally will provide tests of self-medication and substance-induced negative affect models simultaneously . Finally, few studies to-date have addressed the concern of power in detecting these effects, which meta-analyses have suggested are relatively small in magnitude . Well-designed studies utilizing appropriate statistical methods may be better able to detect these effects if they are present, and conversely,cannabis grow racks increase confidence in null results when these effects are not found. The goal of the present study was to examine the extent to which adolescent and young adult depression and anxiety predict binge drinking using a within-person analytic approach. Aims were pursued in the large, diverse, prospectively followed National Consortium on Alcohol and Neurodevelopment in Adolescence cohort from age 17 to age 21. Similar to controlling for between-person common factors , utilization of within person designs can account for between-person factors by allowing statistical models to partition between-person variation from person-level effects. Further, analyses focused on temporally lagged effects, allowing these factors to be appropriately sequenced in time and were supplemented by post-hoc power analyses to increase confidence in the presence or absence of effects. As such, the NCANDA cohort and analyses conducted in this sample may be ideal for clarifying competing hypotheses regarding the association between internalizing symptoms and binge drinking. Given extant evidence favors the self-medication model of binge drinking risk, we hypothesized that adolescent depression, and to a lesser degree anxiety, would predict adolescent binge drinking over a 5-year period for participants from the NCANDA study. Observing no within-person links between these constructs may suggest common factors observed at the between-person level may instead be driving these associations. Data were from the nationally representative National Consortium on Alcohol and Neurodevelopment in Adolescence cohort. Participants were recruited between 12 and 21 years of age at project entry from 5 site locations in 2013–2014. NCANDA is following these individuals through adolescence and into young adulthood . After 2548 participants were screened, 1110 were excluded based on criteria that included MRI contraindications, physical limitations, lack of parental consent, substance use disorders, medication use, prenatal exposure to substances and learning disorders at baseline . To test NCANDA’s primary aims and ensure our sample was optimized to detect changes over time that pertain to one’s alcohol intake, we excluded youth with a range of other factors or conditions that could obscure our ability to do so, including those with prenatal exposure to substances. Recruitment was designed to over sample individuals at higher risk for substance use issues based on endorsement of externalizing symptoms, internalizing symptoms, and family history of alcohol or substance use disorders . The majority of the sample at baseline had limited or no exposure to alcohol or other drugs as determined by the age and sex-based guidelines from the National Institute on Alcohol Abuse and Alcoholism , indicative of misuse based on the Center for Disease Control surveillance. The study’s cohort sequential design recruited adolescents in three groups , facilitating investigation of a wide developmental span due to between-subject variance in starting age. Before entering the study, most participants had not participated in binge drinking .
Participants at risk for increased drinking were identified based on screening for early experimentation with alcohol, positive family history for substance use disorder, and externalizing/internalizing symptoms; these participants were over-recruited and consisted of 50% of participants at study entry . An accelerated longitudinal design allows recruitment of all ages in the cohort starting during the baseline year; 15% of the cohort were selected for enrichment of alcohol and drug use based on NIAAA guidelines for normative drinking in community sample, which was possible due to later age of recruitment at baseline. Of the 831 enrolled participants, 139 were people who drink alcohol and 692 were people who do not drink alcohol at study entry. People who do not drink alcohol were defined as those with fewer than 1–4 drinks once a year or 1–2 drinks once a month . People who drink alcohol were those that exceeded these thresholds . Adult participants provided voluntary informed consent, while minors provided assent in addition to the informed consent of a parent or legal guardian.Alcohol and other substance use history was assessed annually with the Customary Drinking and Drug Use Record to follow use of alcohol, tobacco, cannabis, illicit drugs, and misuse of prescription medications. The CDDR is an interviewer-administered questionnaire, designed for use with adolescents and young adults, that probes recent and lifetime use of alcohol, tobacco, cannabis, illicit drugs, and misuse of prescription medications. It has been found to be internally consistent and reliable over time and across interviewers, in addition to being able to differentiate between abusing and non-abusing adolescents, and with excellent diagnostic specificity compared to other standard instruments. Past year binge drinking was assessed using the item “during the past year, how many times have you consumed 4+ / 5+ drinks within an occasion? ”. Endorsement of binge drinking in the sample increased from baseline to year 5 of data collection . People who drink alcohol endorsed using higher amounts of alcohol, cigarette, and cannabis use compared to people who do not drink alcohol, and using higher amounts of other drugs . Socioeconomic status was assessed with a modified version of the MacArthur Sociodemographic Questionnaire . This reflected parental family income except if the youth was living independently, in which case it reflected the youth’s own socioeconomic status. Twenty percent of parents endorsed education below a college degree, twenty-seven percent with at least one parent completing college, and fifty-three percent with at least one parent with education beyond a college degree for the full sample. Annual family income ranged from below $12,000 to greater than $200,000. A total of 18% of the sample reported income below $50,000 per year. While the median income in the United States at the time of study entry was $52,250, median incomes ranged from $50,988 to $90,786 across NCANDA data collection sites. Reliability across sites and training for assessments was ensured through the development of training manuals, developed by doctoral-level senior staff members, mock and practice sessions, observations, and annual visits to check for interviewer drift and confirmation of training of new staff members . NCANDA uses a cohort sequential design , in which participants spanned a large range of ages at baseline then were assessed annually thereafter. Because participants were enrolled across multiple ages at baseline and provided data in up to five subsequent waves, data resulted in a pattern of planned missingness that can be considered completely at random . We therefore used full information maximum likelihood as an estimator in our structural model to accommodate this design . Missingness was 21.7%, 23.8%, 25.9%, 29.1%, and 29.0% on outcome variables across waves 2 through 6, respectively.