Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Aug 1.
Published in final edited form as: J Abnorm Psychol. 2021 Aug;130(6):641–650. doi: 10.1037/abn0000684

Real-time reports of drinking to cope: Associations with subjective relief from alcohol and changes in negative affect

Andrea M Wycoff a, Ryan W Carpenter b, Johanna Hepp c, Thomas M Piasecki a, Timothy J Trull a
PMCID: PMC8480418  NIHMSID: NIHMS1716291  PMID: 34553959

Abstract

Many individuals report drinking alcohol to cope or relieve negative affective states, but existing evidence is inconsistent regarding whether individuals experience negatively reinforcing effects after drinking to cope (DTC). We used ecological momentary assessment to examine the effects of DTC during daily-life drinking episodes in a sample of current drinkers (N=110; 52 individuals with borderline personality disorder and 58 community individuals). Multilevel models were used to test whether momentary and episode-level endorsement of DTC-depression and DTC-anxiety motives would be related to increased subjective drinking-contingent relief and decreased depression and anxiety during drinking episodes. Momentary DTC-anxiety predicted greater subsequent drinking-contingent relief, and greater episode-level DTC-anxiety and DTC-depression predicted greater drinking-contingent relief during the episode. However, we did not find decreased depression and anxiety following endorsement of DTC-depression and DTC-anxiety. Instead, we found that greater episode-level DTC-depression was associated with increased depression. Thus, findings suggest that individuals’ negative affective states may not improve during DTC despite endorsing drinking-contingent relief. This discrepancy warrants further attention because subjective relief likely reinforces DTC, whereas awareness of one’s change (or lack of change) in affect may provide valuable counterevidence for whether alcohol use is an effective coping strategy.

Keywords: alcohol, drinking to cope, negative affect, negative reinforcement, ecological momentary assessment

A frequently endorsed motive for drinking alcohol is to reduce negative affect (NA), also called drinking to cope (DTC; Cooper, Frone, Russell, & Mudar, 1995; Cooper, Kuntsche, Levitt, Barber, & Wolf, 2016; Cox & Klinger, 1988). DTC has been a particular focus of study due to its associations with negative consequences such as engaging in risky behaviors while drinking, academic and occupational problems, and later development of alcohol dependence (e.g., Cooper et al., 1995; Kuntsche, Knibbe, Gmel, & Engals, 2005; Merrill, Wardell, & Read, 2014). Central to DTC is the idea that alcohol use reduces or relieves NA. Individuals presumably remember this relief and, the next time they experience NA, drink again in an effort to reduce it (i.e., negative reinforcement; Baker et al., 2004; Koob & Le Moal, 2008; Sher & Levenson, 1982). However, it remains unclear whether individuals who engage in DTC actually achieve this goal and experience relief from NA during or after doing so (Sher & Grekin, 2007). Whether DTC leads to relief from NA has implications for understanding negative reinforcement processes in alcohol use and can thereby inform intervention efforts that seek to help individuals to disrupt these processes and develop alternative coping strategies. Thus, we sought to examine whether DTC is associated with relief from NA in individuals’ daily lives.

Ecological momentary assessment (EMA; Stone & Shiffman, 1994) is well-suited to examine patterns of DTC and its immediate, momentary effects. EMA is characterized by repeated assessments over time in individuals’ daily lives, providing high ecological validity, minimizing retrospective bias, and allowing for the disaggregation of between- and within-person processes. Using EMA, Piasecki et al. (2014) found that community-recruited frequent drinkers with higher baseline DTC motives, assessed by the Drinking Motives Questionnaire-Revised (DMQ-R; Cooper, 1994), reported greater drinking-contingent relief (drinking “relieved an unpleasant feeling or symptom”) during daily-life drinking moments, but also reported drinking-contingent punishment (drinking “made me feel worse”). This study marked the first daily-life evidence that DTC motives predict momentary subjective relief from alcohol.

Gorka, Hedeker, Piasecki, and Mermelstein (2017) built upon Piasecki et al. (2014), using the Modified DMQ-R (Grant, Stewart, O’Connor, Blackwell, & Conrod, 2007) to distinguish between coping in terms of coping with depression (DTC-depression) and coping with anxiety (DTC-anxiety). Previous EMA work suggests that DTC-depression and DTC-anxiety show unique patterns of daily-life associations with alcohol use (e.g., Grant, Stewart, & Mohr, 2009), which Gorka et al. (2017) extended to examine whether DTC-depression and DTC-anxiety predicted momentary NA while drinking. Contrary to their hypotheses, however, individuals higher on DTC-anxiety, compared to those lower, experienced smaller decreases in NA when drinking. DTC-depression was not related to changes in NA when drinking.

It is not clear why Piasecki et al. (2014) and Gorka et al. (2017) found differing results. However, subjective relief (assessed in Piaecki et al., 2014) and self-reported NA (assessed in Gorka et al., 2017) may capture overlapping, but distinct, measures of drinking’s affective effects. Items assessing subjective relief and other effects are typically intended to be drinking-contingent, explicitly referring to alcohol use and requiring individuals to judge the effects their recent drinking had on their affect. Individuals high in DTC may be predisposed to report that drinking relieved an unpleasant feeling regardless of their actual experience. In contrast, self-reported NA ratings do not require individuals to make a judgment about how drinking made them feel. Although they examined expectancies and not drinking motives, Treloar, Piasecki, McCarthy, Sher, & Heath (2015) reported EMA results consistent with this idea, finding that individuals with greater tension-reduction expectancies reported both attenuated decreases in self-reported NA at first drink and greater drinking-contingent relief. These findings support the notion that subjective effects of drinking and self-reported NA may provide different information about alcohol’s effects on mood, and studies seeking to characterize whether DTC is associated with relief from NA may benefit from measuring both.

A limitation of these studies is that they measured coping motives only at baseline and treated them as trait-like. However, even individuals who are prone to engaging in DTC also drink for other reasons. Supporting this, recent evidence from studies that measured motives at the day level suggests that motives fluctuate within person and that drinking episodes show differential patterns of affective antecedents (Arbeau, Kuiken, & Wild, 2011; Dvorak, Pearson, & Day, 2014; O’Hara, Armeli, & Tennen, 2014a; Stevenson et al., 2019), social contexts (e.g., drinking alone versus with other people; O’Hara, Armeli, & Tennen, 2015; O’Hara et al., 2014b), quantities of alcohol use (Dvorak et al., 2014; O’Hara et al., 2014b), acute AUD symptoms (Dvorak et al., 2014), and next-day mood (Armeli, O’Hara, Ehrenberg, Sullivan, & Tennen, 2014), depending on whether specific motives are endorsed for that day or that drinking episode, particularly with respect to the coping motive. These findings are consistent with the motivational model, which implies that motives are proximally relevant for daily-life drinking behavior (Cooper et al., 2016). These studies support the measurement of within-person motives in examinations of daily-life drinking processes but did not examine associations of within-person motives with immediate effects of alcohol.

Two daily-diary studies sought to capture NA following DTC at the day level, finding that greater DTC on drinking days was positively associated with next-day NA (Armeli et al., 2014) and stress reactivity (Armeli, O’Hara, Covault, Scott, & Tennen, 2016). This suggests that DTC may not work to relieve NA, but instead exacerbates it, in apparent contrast to the negative reinforcement hypothesis at the core of DTC. However, examination of next-day effects of DTC misses the acute effects of alcohol that are experienced during and immediately after drinking. Acute and immediate relief from NA may be highly salient to individuals who engage in DTC and outweigh more delayed outcomes like next-day affect (Miltenberger, 2011). This would potentially explain the persistence of the DTC motive for future drinking events despite undesirable next-day effects. Thus, the question of whether within-person variation in DTC is associated with changes in momentary subjective relief and NA during daily-life drinking episodes has not yet been addressed.

The current study

A developing literature has examined fluctuations in individuals’ motivations to drink, but significant gaps remain in our understanding of how motivational models of alcohol use translate to individual drinking events and episodes in daily life. The aim of the current study was to examine whether DTC-depression and DTC-anxiety, assessed in the moment during daily-life drinking episodes, predict relief from NA, as assessed by both subjective relief and change in NA while drinking. We examined both indices based on past work that suggested they may provide different information about alcohol’s effects on affect (Gorka et al., 2017; Piasecki et al., 2014; Treloar et al., 2015). We used a sample of individuals with borderline personality disorder (BPD) and community individuals (COM) without BPD to capture variability in emotional states (APA, 2013; Linehan, 1993) and proneness to problematic drinking (e.g., Carpenter et al., 2017; Jahng et al., 2011; Lane, Carpenter, Sher, & Trull, 2016; Trull et al., 2000; 2018). Introducing variability in constructs of interest is methodologically beneficial as the timeframe that EMA captures may not adequately sample problematic drinking and its correlates in the general population (e.g., Lane & Hennes, 2019).

Our primary interest was the lagged momentary level of analysis, as we collected momentary motive endorsements each time participants reported drinking. However, as previous studies measured motives at the day level (i.e., once per day or once per drinking episode; Arbeau et al., 2011; Armeli et al., 2014; Dvorak et al., 2014; O’Hara et al., 2014a; O’Hara et al., 2015), we also examined the day level of analysis to facilitate comparison across studies. We expected day-level aggregates, henceforth referred to as episode-level,1 to show similar patterns as our momentary motive indices. Based on theoretical concepts of alcohol use being negatively reinforcing (Baker et al., 2004; Cooper et al., 1995), we hypothesized that lagged momentary- and episode-level endorsement of DTC-depression and DTC-anxiety would both be associated with increased endorsement of subjective drinking-contingent relief. The hypotheses for change in NA after drinking mirrored the hypothesis for subjective relief. We hypothesized that both momentary and episode-level endorsement of DTC-depression would be associated with decreased depression since the last prompt. Additionally, we hypothesized that both momentary and episode-level DTC-anxiety would be associated with decreased anxiety since the last prompt. Although these hypotheses for predicting momentary change in NA are in contrast to existing literature on the effects of DTC on NA (e.g., Armeli et al., 2014; Gorka et al., 2017), our study measured both DTC and affective outcomes in the moment, as opposed to at the day-level or at baseline. Therefore, we focused on testing theories of negative reinforcement at the momentary level and derived our hypotheses accordingly, rather than expecting that momentary findings would necessarily mirror previous empirical findings at the day level.

Based on evidence that negative reinforcement processes may be more relevant to patterns of alcohol use in individuals with AUD than those without (Cho et al., 2019; Koob & Volkow, 2010), we also modeled whether associations of interest differed between individuals with an alcohol use disorder (AUD) diagnosis and those without. Because we recruited individuals based on BPD diagnosis, there is an overlap of AUD and BPD in our sample. Therefore, we also included BPD moderator models for comparison to determine whether any observed AUD moderator effects might be attributable to the fact that many of those individuals also had a BPD diagnosis. We briefly present and discuss results from these AUD and BPD moderator models in the manuscript but see the Supplemental Material for full results and discussion related to this question.

Method

Participants

The sample included 116 individuals between the ages of 18 and 45 who reported consuming alcohol at least once per week. Participants were recruited for a study of alcohol use and emotional instability (Lane et al. 2016; Wycoff, Carpenter, Hepp, Lane, & Trull, 20202) and included 56 participants with BPD and 60 COM participants from the community. Individuals were excluded from participating if they reported current psychotic symptoms, had an intellectual disability, were pregnant or planning to become pregnant, or had experienced a significant past head trauma. Individuals were also excluded from participating if they were currently in or seeking treatment for substance use, reported past-year attempts to cut down or stop drinking, or experienced significant past-year physiological withdrawal symptoms, because paying such individuals for an observational study of alcohol use could be seen as undermining efforts to change.

Following an initial phone screening, potential participants were invited to the laboratory to complete in-person diagnostic interviews using the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I; First et al., 1995) and Structured Interview for DSM-IV Personality Disorders (Pfohl et al., 1994). Interrater reliabilities were computed for a subset of 20 participants and were excellent for the diagnosis of BPD (κ = 0.88) and for current AUD (κ = 1.00). Participants in the BPD group met criteria for BPD and were currently in outpatient mental health treatment. Participants in the COM group did not meet criteria for BPD nor endorse the affective instability criterion of BPD, though they could (and did) have other psychiatric diagnoses.

One participant was excluded from current analyses due to reporting drink totals within multiple episodes that were implausibly high (i.e., would have been expected to result in hospitalization or death; see Carpenter et al., 2017, for details), one participant was excluded due to not providing data on baseline drinking motives, of which one subscale is used as a covariate in the current analyses, two participants were excluded from current analyses as they did not report any drinking during the EMA period, and two additional participants were excluded as they only reported drinking during random prompts, which did not assess the variables that are used in the current analyses (i.e., momentary motives, subjective relief). The final sample therefore consisted of 110 participants, comprising 58 COM individuals and 52 BPD individuals. In the total sample, mean age was 26.48 (SD = 7.24, range = 18–45) and participants were 78.2% female. Twenty-four participants (21.8%) met criteria for a current AUD based on the SCID-I. Additional demographic and diagnostic information for the sample are presented in Table 1 stratified by AUD and BPD diagnostic status to provide detail on group differences and co-occurring diagnoses.

Table 1.

Demographic information and psychiatric diagnoses, presented by AUD and BPD diagnostic status.

AUD diagnostic status
 BPD diagnostic status
Absent (n = 86)
 Present (n = 24)
 Absent (n = 58)
 Present (n = 52)
n % n % n % n %
Ethnicity
 Caucasian 73 84.9 20 82.7 50 86.2 43 82.7
 African-American 4 4.7 2 8.3 3 5.2 3 5.8
 Asian-American 3 3.5 0 0.0 2 3.5 1 1.9
 Hispanic 2 2.3 1 4.2 2 3.5 1 1.9
 Other 4 4.7 1 4.2 1 1.7 4 7.7
Marital Status
 Single 56 65.9 19 79.2 37 64.9 38 73.1
 Married 18 21.2 4 16.7 16 28.1 6 11.5
 Divorced or separated 7 8.2 1 4.2 1 1.8 7 13.5
 Cohabitating 4 4.7 0 0.0 3 5.3 1 1.9
Annual Household Income
 $0 to $25,000 50 58.1 13 54.2 23 39.7 40 76.9
 $25,001 to $50,000 18 20.9 4 16.7 16 27.6 6 11.5
 $50,001 to $75,000 5 5.8 2 8.3 5 8.6 2 3.9
 $75,001 to $100,000 5 5.8 2 8.3 5 8.6 2 3.9
 Above $100,000 8 9.3 3 12.5 9 15.5 2 3.9
Current Psychiatric Disorders
 BPD 35 40.7 17 70.8 0 0.0 52 100.0
 Any PD other than BPD 16 18.6 11 45.8 1 1.7 26 50.0
 AUD 0 0.0 24 100.0 7 12.1 17 32.7
 Any SUD other than AUD 5 5.8 6 25.0 1 1.7 21 40.4
 Any Anxiety Disorder 32 37.2 13 54.2 12 20.7 33 63.5
 Any Mood Disorder 14 16.3 9 37.5 1 1.7 22 42.3
 Any Eating Disorder 3 3.5 0 0.0 0 0.0 10 19.2
Open in a new tab

Note. BPD = borderline personality disorder, SUD = substance use disorder, AUD = alcohol use disorder, PD = personality disorder. AUD and BPD diagnostic status columns refer to current AUD and BPD diagnoses.

Procedures

Approval for study procedures was granted by the Institutional Review Board of the University of Missouri (Protocol 1133597). First, participants visited the lab to complete demographic and self-report questionnaires and receive an electronic diary (ED; Palm Tungsten E2©) and instructions on the EMA procedures. Participants provided 21 days (M = 21.6 days, SD = 2.0) of EMA data during which they completed a combination of time-based, random, and event-based reports. The time-based report was a morning report upon waking. Then, participants received six random prompts within six stratified equal time intervals throughout the day. Event-based, user-initiated reports consisted of initial drink reports after consuming the first standard drink of alcohol of a day, cigarette use reports after smoking a cigarette, and self-harm reports after engaging in self-harm. After reporting alcohol use in any type of prompt,3 participants received a series of four follow-up prompts at 30-, 60-, 120-, and 180-minutes after the initial report to more densely sample the drinking episode. Additional follow-up prompts were scheduled 60 minutes after the last scheduled prompt any time additional drinking was reported. Participants were compensated $20 for the diagnostic screening, $10 for the initial lab session, $50 for 80% or better compliance for each week of EMA, and an additional $10 for a lab session during which they completed additional self-report questionnaires at the end of the EMA period. Random-prompt compliance for the present sample was 90.4% (SD = 8.0%). Drinking follow-up compliance for the present sample was 93.4% (SD = 9.1%).

Measures

Momentary drinking motives.

When participants reported drinking during self-initiated and follow-up drinking reports (in total, 1,713 observations), they were asked to rate whether they drank because it would make them “feel less guilty or depressed” and “feel more relaxed or calm,” on a 1 to 4 scale (1 = strongly disagree, 4 = strongly agree).4 Drinking to “feel less guilty or depressed” maps onto a DTC-depression item of the Modified DMQ-R (Grant et al., 2007), “because it helps me when I am feeling depressed,” and drinking to “feel more relaxed or calm” maps onto a DTC-anxiety item of the Modified DMQ-R, “to relax.” Thus, we used the guilty/depressed item in the current study to indicate DTC-depression, and the relaxed/calm item to indicate DTC-anxiety. Table 2 provides descriptive statistics for momentary DTC motives and correlations with central variables.

Table 2.

Means, SDs, and correlations for person-level mean of DTC, baseline DMQ-R coping, and person-level means of drinking-contingent relief, depression, and anxiety.

Correlations
Variable Mean SD 1 2 3 4 5
1. Person-level DTC-depression 1.72 0.73
2. Person-level DTC-anxiety 2.72 0.72 .62
3. Baseline DMQ-R coping 9.30 4.52 .72 .51
4. Person-level drinking-contingent relief 2.16 1.08 .72 .67 .55
5. Person-level depression 1.26 0.49 .62 .36 .49 .40
6. Person-level anxiety 1.23 0.41 .59 .39 .55 .45 .81
Open in a new tab

Note. DTC = drinking to cope. DMQ-R = Drinking Motives Questionnaire-Revised (Cooper, 1994). All bivariate correlations were significant at p < .001.

Momentary subjective relief from alcohol.

When participants reported drinking during self-initiated and follow-up drinking reports, they were also asked to rate how much the drink(s) “relieved unpleasant feelings or symptoms” (drinking-contingent relief) on a 1 to 5 scale (1 = not at all, 5 = extremely).5 See Table 2 for descriptive statistics.

Momentary NA.

At every prompt, participants rated how much they felt various negative affective states in the past 15 minutes on a 1 to 5 scale (1 = very slightly/not at all, 5 = extremely) on items taken from the Positive and Negative Affect Schedule-Extended version (PANAS-X; Watson & Clark, 1999). Sadness (5 items) and fear (6 items) scales were aggregated into mean scores for each person at each prompt. The items included in the fear scale were afraid, scared, frightened, nervous, jittery, and shaky. The items included in the sadness scale were sad, blue, downhearted, alone, and lonely. The sadness and fear scales are considered to measure the same basic affective states as depressed and anxious mood (Watson & Clark, 1999). Specifically, PANAS-X fear is highly correlated with measures of anxiety symptoms and with the tension-anxiety subscale of the Profile of Mood States (POMS; McNair, Lorr, & Droppleman, 1971), and PANAS-X sadness is highly correlated with measures of depression symptoms and with the depression-dejection subscale of the POMS (Watson & Clark, 1999). As such, we refer to the sadness and fear scales as depression and anxiety for the purpose of keeping terms consistent with “DTC-depression” and “DTC-anxiety.” Change scores were calculated for use as an outcome variable by subtracting the last prompt’s depression or anxiety rating from the current prompt’s depression or anxiety rating. See Table 2 for descriptive statistics. To justify using change scores, within-person reliabilities of change across moments in the study (RCs) were calculated for depression and anxiety, consistent with recommendations by Shrout & Lane (2012). RCs were adequate: RC for anxiety was .76, and RC for depression was .73.

Analytic method

Analyses retained only those observations during which drinking was reported in user-initiated initial drink and drinking follow-up reports, as momentary motives and subjective effects of drinking were only collected in these types of reports when participants reported that they had drank alcohol since the last prompt. Data were analyzed using three multilevel models in SAS PROC MIXED. We specified random intercepts for individuals and episodes within individuals. We further modeled random slopes for the effects of episode-level drinking motives and momentary drinking motives on outcomes. All random intercepts were significant and thus retained in the final models; the random slopes that were significant and retained are listed in their respective table notes.

Model 1 predicted the level of endorsement of momentary drinking-contingent relief. The predictors of interest were DTC-depression and DTC-anxiety motives at the previous moment (lagged) and at the episode level. Because we used lagged DTC motives to predict subjective relief at the next timepoint, we also controlled for concurrent-moment DTC motives in Model 1.6 Models 2 and 3 predicted momentary change in depression (Model 2) and anxiety (Model 3). The predictors of interest in both Models 2 and 3 were concurrent-moment and episode-level DTC-depression and DTC-anxiety motives. Change scores for depression and anxiety were calculated prior to excluding non-drinking prompts, meaning that change scores used in the analyses represented affect reported in the current drinking moment minus affect reported from the most recent previous prompt, regardless of whether that most recent prompt contained drinking. Because we predicted change scores, we also included the last prompt’s level of depression (Model 2) or anxiety (Model 3) as covariates to account for there being more room to “improve” if depression or anxiety started out high. Using change scores for depression and anxiety facilitated interpretations of results. For instance, a negative estimate for momentary DTC-depression predicting change in depression would indicate that higher current DTC-depression is associated with a reduction in depression from the previous to the current assessment. Whereas this was appropriate for the affect models because it established the temporal ordering of affect before and after DTC, we examined level of subjective relief as opposed to change in subjective relief because of the way the subjective effects were assessed. Specifically, subjective effects were only assessed if drinking was endorsed, meaning that subjective effects were not assessed at the previous assessment if drinking was not endorsed at the previous assessment. However, our use of lagged DTC motives to predict current-moment subjective relief served to establish the temporal ordering of subjective relief following endorsement of DTC motives.

To disaggregate momentary-, episode-, and person-level indicators based on these momentary motive variables, momentary responses were included as predictors and were centered on episode-means, episode-means were included as predictors and were centered on person-means, and person-means were included as covariates and were centered on the sample mean (Curran & Bauer, 2011). Thus, the momentary- and episode-level predictors are within-person components, whereas the person-level predictor is the between-person component. As we were interested in the momentary and episode-levels of analysis, the person-level variables were included only as adjustments and were not of primary interest. Additional covariates for all three models included BPD (present/absent), AUD (present/absent), any current mood disorder (present/absent), any current anxiety disorder (present/absent), baseline score on the DMQ-R coping subscale to establish whether DTC motives measured in the moment are related to momentary relief and NA above and beyond what can be explained by trait DTC (centered on the sample mean; see Table 2 for descriptive statistics), whether participants were alone or with other people at each moment, age (centered on the sample mean), gender, cumulative drinks so far in that drinking episode at that moment, whether it was a weekday or the weekend, day in the study, and hour after wake.

Results

The total number of drinking episodes analyzed was 716, the average number of drinking episodes analyzed per person was 6.51 (SD = 4.02, range = 1–18), and the average number of observations analyzed per drinking episode was 2.39 (SD = 1.59, range = 1–10).

Unconditional models predicting momentary DTC-depression- and DTC-anxiety provided empirical support for the use of three-level multilevel models, as there was sufficient variability in endorsement of each motive at all three levels (momentary ICCs ≥ .31, episode-level ICCs ≥ .17, and person-level ICCs ≥ .47).

Subjective relief (Model 1)

Table 3 presents results showing the effects of momentary motives on drinking-contingent relief (Model 1). In Model 1, lagged-moment endorsement of DTC-anxiety predicted higher endorsement of current-moment drinking-contingent relief. In addition, episode-level endorsements of DTC-depression and DTC-anxiety predicted higher endorsement of drinking-contingent relief. At the person level, both DTC-depression and DTC-anxiety were associated with greater endorsement of drinking-contingent relief.

Table 3.

Estimates, 95% CIs, and p-values for lagged momentary-, momentary-, and episode-level motives and covariates predicting momentary subjective relief from alcohol.

Relief (Model 1)
Effect Estimate 95% CI p
Intercept 1.86 [1.31, 2.40] <.001
Lag Mom DTC-dep −0.01 [−0.20, 0.18] .904
Lag Mom DTC-anx 0.12 [0.01, 0.22] .030
Mom DTC-dep 0.01 [−0.19, 0.21] .910
Mom DTC-anx 0.06 [−0.05, 0.17] .265
Ep DTC-dep 0.33 [0.04, 0.62] .025
Ep DTC-anx 0.30 [0.16, 0.45] <.001
Covariates
 Person DTC-dep 0.79 [0.40, 1.18] <.001
 Person DTC-anx 0.66 [0.35, 0.97] <.001
 DMQ-R Coping 0.00 [−0.06, 0.07] .924
 BPD (absent=ref) 0.02 [−0.46, 0.50] .927
 AUD (absent=ref) 0.02 [−0.42, 0.45] .940
 Current mood disorder −0.15 [−0.67, 0.37] .577
 Current anx disorder 0.16 [−0.23, 0.56] .418
 With people 0.02 [−0.16, 0.20] .818
 Age −0.02 [−0.05, 0.00] .098
 Gender (male=ref) −0.12 [−0.54, 0.30] .571
 Cumulative drinks 0.02 [−0.00, 0.04] .067
 Weekend −0.11 [−0.24, 0.03] .137
 Day in the study −0.01 [−0.02, 0.00] .108
 Hour after wake 0.03 [0.00, 0.05] .026
Open in a new tab

Note. CI = confidence interval, Mom = momentary, Ep = episode, DTC-dep = drinking to cope-depression, DTC-anx = drinking to cope-anxiety, BPD = borderline personality disorder, AUD = alcohol use disorder, anx = anxiety, ref = reference. Random slopes were specified for lagged momentary DTC-depression, concurrent momentary DTC-depression, and episode-level DTC-depression. Significant effects at p < .05 are bolded.

Change in NA (Models 2 and 3)

Table 4 presents the results of motives on change in depression (Model 2) and change in anxiety (Model 3) since the last prompt. In Model 2, episode-level DTC-depression predicted an increase in depression since the last prompt. Though consistent in sign, momentary DTC-depression was not significantly related to change in depression (p = .051). In Model 3, neither momentary nor episode-level DTC-anxiety significantly predicted change in anxiety since the last prompt. At the person level, DTC-depression was associated with increased depression (Model 2) as well as increased anxiety (Model 3) at any given drinking moment.

Table 4.

Estimates, 95% CIs, and p-values for momentary- and episode-level motives and covariates predicting momentary change in NA.

Depression (Model 2)
 Anxiety (Model 3)
Effect Estimate 95% CI p Estimate 95% CI p
Intercept 0.76 [0.62, 0.90] <.001 0.71 [0.58, 0.84] <.001
Mom DTC-dep 0.07 [−0.00, 0.14] .051 0.00 [−0.05, 0.06] .905
Mom DTC-anx 0.01 [−0.03, 0.05] .731 0.02 [−0.01, 0.05] .155
Ep DTC-dep 0.11 [0.05, 0.16] <.001 0.02 [−0.01, 0.04] .294
Ep DTC-anx 0.01 [−0.02, 0.05] .527 0.02 [−0.01, 0.04] .220
Covariates
 Person DTC-dep 0.19 [0.10, 0.29] <.001 0.14 [0.05, 0.23] .003
 Person DTC-anx −0.03 [−0.10, 0.04] .391 −0.00 [−0.07, 0.06] .915
 DMQ-R Coping 0.00 [−0.01, 0.02] .591 0.01 [−0.00, 0.02] .158
 Last-prompt NA −0.49 [−0.53, −0.45] <.001 −0.56 [−0.60, −0.52] <.001
 BPD (absent=ref) 0.06 [−0.06, 0.17] .327 0.08 [−0.03, 0.18] .136
 AUD (absent=ref) 0.04 [−0.07, 0.14] .478 −0.01 [−0.11, 0.09] .858
 Current mood disorder 0.01 [−0.11, 0.13] .848 −0.05 [−0.16, 0.06] .372
 Current anx disorder −0.02 [−0.11, 0.07] .664 −0.04 [−0.13, 0.05] .409
 With people −0.10 [−0.15, −0.05] <.001 0.00 [−0.03, 0.04] .933
 Age 0.00 [−0.00, 0.01] .630 −0.00 [−0.01, 0.00] .479
 Gender (male=ref) −0.02 [−0.12, 0.07] .632 −0.05 [−0.15, 0.04] .267
 Cumulative drinks 0.01 [−0.00, 0.01] .068 0.00 [−0.00, 0.01] .178
 Weekend −0.04 [−0.07, 0.00] .053 −0.02 [−0.04, 0.01] .235
 Day in the study −0.00 [−0.00, 0.00] .786 0.00 [−0.00, 0.00] .388
 Hour after wake −0.01 [−0.01, −0.00] .049 −0.00 [−0.01, 0.00] .513
Open in a new tab

Note. CI = confidence interval, NA = negative affect, Mom = momentary, Ep = episode, DTC-dep = drinking to cope-depression, DTC-anx = drinking to cope-anxiety, BPD = borderline personality disorder, AUD = alcohol use disorder, anx = anxiety, ref = reference. In Model 2, random slopes were specified for momentary and episode-level DTC-depression. In Model 3, a random slope was specified for momentary DTC-depression. Significant effects at p < .05 are bolded.

Moderation Analyses (Supplemental Models S2 through S7)

AUD diagnosis did not moderate any of the associations between lagged momentary or episode-level DTC motives and drinking-contingent relief (Supplemental Table S2). AUD diagnosis did, however, moderate the effect of episode-level DTC-depression and change in depression such that the positive association between the two was stronger for individuals with AUD (Supplemental Table S3; Supplemental Figure S1). In addition, AUD diagnosis moderated the effect of episode-level DTC-anxiety and change in anxiety such that episode-level DTC-anxiety predicted increased anxiety for individuals with AUD but not for individuals without AUD (Supplemental Table S3; Supplemental Figure S3). BPD diagnosis did not moderate any of the associations of interest between DTC motives and drinking-contingent relief, change in depression, or change in anxiety (Supplemental Tables S4 and S5). See the Supplemental Material for additional detail.

Discussion

The current study examined whether DTC-depression and DTC-anxiety, measured in the moment during participants’ daily-life drinking episodes, were associated with potentially negatively reinforcing outcomes. We used a sample of individuals with a wide range of emotion dysregulation to capture variability in DTC and its effects during the EMA assessment period. We tested a central component of negative reinforcement (Baker et al., 2004) and motivational (Cooper et al., 1995) theories of alcohol use: that individuals experience relief of NA in daily life following DTC. Consistent with previous practices (Grant et al., 2009), we distinguished between DTC-depression and DTC-anxiety to determine whether these two types of coping were differentially related to potentially reinforcing effects. Further, we measured DTC outcomes in terms of subjective relief from alcohol as well as self-reported depression and anxiety, with the two methods potentially providing different information about an individual’s experience.

First, we found that drinking moments with higher endorsement of DTC-anxiety predicted subsequent momentary subjective relief from unpleasant feelings or symptoms. We also found that drinking episodes with higher endorsement of both DTC-depression and DTC-anxiety predicted momentary subjective relief. This supports the negative reinforcement process inherent in the motivational model (Baker et al., 2004; Cooper et al., 1995) and is consistent with evidence that baseline endorsement of DTC is related to subjective drinking-contingent relief in daily life (Piasecki et al., 2014).

Second, we expected momentary depression and anxiety to decrease after drinking moments and during drinking episodes that were characterized by higher endorsement of DTC-depression and DTC-anxiety. However, we did not find that pattern. Instead, we found increased depression after drinking during episodes when DTC-depression was more highly endorsed than during other episodes, and no significant change in anxiety after drinking when DTC-anxiety was endorsed more highly at that moment or during that episode. Findings of increased depression and unchanged anxiety during drinking episodes characterized by DTC suggest that alcohol use may not be an effective means of coping with NA, at least not during drinking episodes for which DTC is endorsed. This is somewhat consistent with the finding of attenuated decreases in NA during daily-life drinking episodes among individuals who endorsed baseline DTC-anxiety motives (Gorka et al., 2017) and seems to contradict the hypothesis that DTC would be negatively reinforcing. Notably, momentary depression increased even more for individuals with AUD than those without during episodes when DTC-depression was highly endorsed. Similarly, momentary anxiety increased for individuals with AUD, but not for those without, during episodes when DTC-anxiety was highly endorsed. Further, despite the overlap of AUD and BPD in our sample, these effects appear to be specific to AUD diagnosis. See the Supplemental Material for additional discussion on this topic.

The discrepancy between subjective relief from alcohol versus affective change could reflect the difference in the way the two are measured. Change in affect directly compares current reported affect to affect reported prior to the most recent alcohol consumption, which establishes temporal precedence and does not rely on an individual’s belief or attribution of how their recent drink(s) made them feel. In contrast, subjective effects require individuals to make a judgment about how drinking made them feel and may reflect some combination of an expectation about the effects of alcohol and a pharmacological effect of alcohol. Consistent with this, Treloar and colleagues (2015) reported a similar discrepancy such that individuals with greater tension-reduction expectancies at baseline reported greater drinking-contingent relief but attenuated decreases in NA during daily-life drinking episodes. Findings from the current study extend this finding to the within-person level for drinking episodes characterized by DTC motives. It is also possible that participants’ endorsement of drinking-contingent relief might have captured a different set of experiences than depression and anxiety, such as pain, discomfort, restlessness, etc. This could point to a methodological discrepancy that could be examined in future studies to improve our understanding of in what ways individuals drink to cope with experiences other than negative affect. Ultimately, this discrepancy warrants further attention. Whereas subjective drinking-contingent relief likely reinforces DTC, awareness of how NA is or is not changing during DTC may provide valuable counterevidence for whether alcohol use is an effective coping strategy.

Finally, although between-person effects were not a focus of the present study, it is worth noting that there were several significant associations of person-level DTC motives with our momentary outcomes of interest. In particular, both DTC-depression and DTC-anxiety at the person level were associated with greater endorsement of subjective relief, indicating that individuals who reported greater DTC motives throughout the study also reported more relief from drinking. This is consistent with the momentary DTC-anxiety effect as well as the episode-level effects for both DTC motives, and provides some evidence for the importance of capturing DTC motives at the momentary-, episode-, and person-levels. In addition, person-level DTC-depression was related to increased depression and increased anxiety at any given drinking moment, indicating that individuals who reported greater DTC-depression motives throughout the study reported greater increases in depressed and anxious affect during drinking moments. This is consistent with the effect of episode-level DTC-depression predicting increased momentary depression. Thus, individuals who tended to endorse DTC also reported greater increases in depression and anxiety during drinking episodes overall, even in moments and episodes where they did not necessarily strongly endorse DTC. This suggests that DTC may be generally associated with broad emotion regulation difficulties.

Limitations

There are number of limitations to the current study. First, we only used one item per DTC-depression and DTC-anxiety. Future studies interested in characterizing alcohol use patterns based on momentary drinking motives would benefit from using multiple items per motive to increase reliability (Shrout & Lane, 2012; Stevenson et al., 2019). As a result, while our DTC-depression and -anxiety items were similar to items from the Modified DMQ-R scale (Grant et al., 2007), our items likely do not capture the full range of DTC-depression and DTC-anxiety. Further, the wording of our DTC-depression item (“to feel less guilty or depressed”) collapsed two affect descriptors into one item. Although this broadened the scope of this single item, future examination should use multiple items to separately assess drinking to reduce depression and guilt. In addition, the wording of our DTC-anxiety item (“to feel more relaxed or calm”) could be considered positive reinforcement more so than negative reinforcement. Although our wording was consistent with an item from the Modified DMQ-R (“to relax”), future work should increase the number of momentary items to more fully sample the Modified DMQ-R scales and examine the reliability and validity of momentary versions of these scales.

Second, the EMA protocol prompted participants to report on their momentary drinking motives whenever they reported that they just had a drink or drank since the last prompt. Thus, drinking itself could influence what the reported motive was. For example, if someone felt that the drink they just consumed made them feel less anxious, they might be more likely to report that they drank because they wanted to feel more relaxed. EMA designs that prompt participants to make reports prior to any drinking that they expect to engage in might address this (e.g., as in Stevenson et al., 2019). Similarly, a third limitation is that motives and subjective relief were reported at the same prompt. Again, this could influence reporting such that if an individual endorsed drinking to feel less depressed, they might be motivated to also report that the drink relieved an unpleasant feeling. Conversely, if an individual felt that drinking relieved an unpleasant feeling, they might infer that they were drinking for that purpose and be more likely to endorse DTC. The current study addressed this by using DTC motives at one moment to predict subjective relief at the next moment, thereby assessing whether DTC motives have a prospective effect on subjective relief, which would likely be less influenced by any concurrent-moment consistency of reporting. In addition, because momentary- and episode-level predictors are included simultaneously in each model and their effects disaggregated, the presence of episode-level effects of DTC on subjective relief suggests an association between DTC and relief above and beyond what would be assumed based on concurrent-moment consistency of reporting.

A fourth limitation is that our method is unable to distinguish between expectancy versus pharmacological effects of alcohol, and we do not interpret our results as necessarily reflecting a pharmacological effect of alcohol. Although an expectancy effect of subjective relief may be enough to serve as a negative reinforcer of alcohol use and, as such, is still a valuable target for assessment and intervention, it is worth noting that expectancy and pharmacological effects have different implications for intervention. For instance, if the positive association between DTC motives and subjective relief is driven by an expectation that DTC will provide relief from unpleasant feelings, then an expectancy challenge intervention might be indicated, which would provide individuals with the opportunity to examine what role expectations play in their drinking experiences (Dunn, Fried-Somerstein, Flori, Hall, & Dvorak, 2020; Scott-Sheldon, Terry, Carey, Garey, & Carey, 2012). In contrast, if the positive association between DTC motives and subjective relief is driven by pharmacology, then intervention might focus on whether any longer-term negative outcomes are present that might benefit from replacing DTC with other coping strategies that might provide not only short-term relief, but longer-term positive outcomes as well. Therefore, future work should seek to disaggregate the relative effects of expectancy versus pharmacology on momentary outcomes of DTC.

A fifth limitation is that our sample was partially clinical in nature and had a range of psychopathology including BPD and other diagnoses (Table 1). Although we viewed this as a strength to ensure adequate base rates for processes like coping-motivated drinking that may be difficult to capture during EMA timeframes with samples that do not experience as much NA or problematic drinking patterns, it raises the question of how generalizable our findings are to other populations. Sixth, because we excluded individuals who were in or seeking treatment for substance use, who had past-year attempts to cut down or stop drinking, or who experienced significant past-year physiological withdrawal symptoms, our results may not generalize to individuals with more severe AUD. Future studies should examine whether our results replicate in a sample of individuals with a more representative range of AUD symptoms.

Finally, a seventh limitation is that our sample was primarily (78.2%) female. The overrepresentation of women was a result of recruiting participants with BPD through treatment centers, in which women with BPD tend to be over-represented (e.g., Lenzenweger, Lane, Loranger, & Kessler, 2007). At the same time, this prohibited examination of potential gender differences and limits the generalizability of our findings. Based on emerging evidence that within-person associations among DTC motives and problems related to alcohol use may be stronger in men than in women (e.g., Dvorak, Pearson, & Day, 2014), it is possible that we would have found stronger associations among DTC motives, subjective relief, and change in negative affect had our sample included more men. Greater inclusion of men also would have provided sufficient power to test gender as a possible moderator of the association between DTC motives and alcohol use. This will be important to assess in future work.

Conclusions

The current study adds to an emerging literature seeking to measure drinking motives and their correlates in daily life. Although daily-life DTC was largely associated with endorsement of drinking-contingent relief, depression and anxiety did not improve in the moment. In fact, momentary depression increased after drinking when DTC-depression was endorsed. Our findings support the measurement of drinking motives in daily life and shed light on an important discrepancy between subjective relief attributed to alcohol use and momentary changes in NA, indicating that, although individuals generally report drinking-contingent relief, DTC may not reduce NA. This discrepancy highlights subjective relief as a potential reinforcer of DTC, whereas a lack of improvement in affect during DTC suggests that alcohol use is not an effective coping strategy. Findings from this study importantly move forward motivational models of alcohol use in the context of daily life.

Supplementary Material

Supplemental Material

General Scientific Summary:

Individuals who report drinking alcohol to cope with negative affect are at risk for experiencing alcohol-related problems. Findings from this study indicate that when individuals endorse drinking alcohol to cope with negative affect in their daily lives, they report drinking-contingent relief from unpleasant feelings or symptoms, but their moment-to-moment changes in negative affect do not show an improvement. This suggests that subjective drinking-contingent relief may reinforce drinking to cope, whereas awareness of a lack of improvement in negative affect may provide useful counterevidence for whether drinking is an effective coping strategy.

Acknowledgments

This work was supported by the National Institutes of Health P60 AA011998 (Trull/Heath), T32 AA013526 (Sher), F31 AA027958 (Wycoff), and T32 AA007459 (Monti).

Footnotes

Declarations of interest: none.

1

Day-level aggregates can be thought of as representing the level of the drinking episode in our study because they aggregate all momentary motives from drinking reports on a given day and exclude non-drinking observations. For days with multiple drinking episodes, which occurred on 5.4% of the drinking days that we analyzed, we only included the first drinking episode of the day in our analyses. Consistent with Carpenter et al. (2017), we defined multiple drinking episodes as occurring when three or more hours had passed between the report of any two drinks and estimated blood alcohol concentration had fallen to .000 g% during that same timeframe.

2

Several articles have been published using data from this study. The article most similar to the current study used only the BPD group and focused on baseline coping and enhancement motives as person-level moderators of negative and positive affect throughout the day predicting subsequent initiation and continuation of a drinking episode (Wycoff et al., 2020). The current study is distinct in that it focuses on real-time reports of DTC and examines subjective relief and NA as outcomes of DTC.

3

At all prompts, participants were asked whether they had consumed alcohol since their previous report as a safety net in case participants did not report their initial drink immediately after finishing.

4

Participants were also asked to rate whether they drank for four additional reasons: “because it would make it easier to be around people,” “because it would make me more courageous or daring,” “because it would make it easier to think or perform a task,” and “because it would make me more aggressive or tough.” We did not include these items in the current analyses because they do not clearly map onto the other DMQ-R motives for drinking.

5

Participants were also asked to rate how much the drink was “pleasurable” and “made me feel worse.” We did not include these subjective effects in the currently analyses because they did not directly address negative reinforcement of drinking.

6

We also tested whether momentary DTC was associated with concurrent subjective relief as a sensitivity analysis to examine whether associations between DTC and subjective relief differ over different time courses. See Supplemental Material (Model S1) for analytic method, results, and discussion of this model.

References

  1. American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders: DSM-5. Washington, DC: American Psychiatric Association. [Google Scholar]
  2. Arbeau KJ, Kuiken K, & Wild CT (2011). Drinking to enhance and cope: A daily process study of motive specificity. Addictive Behaviors, 36, 1174–1183. [DOI] [PubMed] [Google Scholar]
  3. Armeli S, O’Hara RE, Covault J, Scott DM, & Tennen H (2016). Episode-specific drinking-to-cope motivation and next-day stress-reactivity. Anxiety, Stress & Coping, 29, 673–684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Armeli S, O’Hara RE, Ehrenberg E, Sullivan T, & Tennen H (2014). Episode-specific drinking-to-cope motivation, daily mood, and fatigue-related symptoms among college students. Journal of Studies on Alcohol and Drugs, 75, 766–774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baker TB, Piper ME, McCarthy DE, Majeskie MR, & Fiore MC (2004). Addiction motivation reformulated: An affective processing model of negative reinforcement. Psychological Review, 111, 33–51. [DOI] [PubMed] [Google Scholar]
  6. Carpenter RW, Trela CJ, Lane SP, Wood PK, Piasecki TM, & Trull TJ (2017). Elevated rate of alcohol consumption in borderline personality disorder patients in daily life. Psychopharmacology, 234, 3395–3406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cho SB, Su J, Kuo SI, Bucholz KK, Chan G, Edenberg HJ, … Dick DM (2019). Positive and negative reinforcement are differentially associated with alcohol consumption as a function of alcohol dependence. Psychology of Addictive Behaviors, 33, 58–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cooper ML (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6, 117–128. [Google Scholar]
  9. Cooper ML, Frone MR, Russell M, & Mudar P (1995). Drinking to regulate positive and negative emotions: A motivational model of alcohol use. Journal of Personality and Social Psychology, 69, 990–1005. [DOI] [PubMed] [Google Scholar]
  10. Cooper ML, Kuntsche E, Levitt A, Barber LL, & Wolf S (2016). Motivational models of substance use: A review of theory and research on motives for using alcohol, marijuana, and tobacco. The Oxford Handbook of Substance Use and Substance Use Disorders, 1, 375–421. [Google Scholar]
  11. Cox WM, & Klinger E (1988). A motivational model of alcohol use. Journal of Abnormal Psychology, 97, 168–180. [DOI] [PubMed] [Google Scholar]
  12. Curran PJ, & Bauer DJ (2011). The disaggregation of within-person and between-person effects in longitudinal models of change. Annual Review of Psychology, 62, 583–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dunn ME, Fried-Somerstein A, Flori JN, Hall TV, & Dvorak RD (2020). Reducing alcohol use in mandated college students: A comparison of a Brief Motivational Intervention (BMI) and the Expectancy Challenge Alcohol Literacy Curriculum (ECALC). Experimental and Clinical Psychopharmacology, 28, 87–98. [DOI] [PubMed] [Google Scholar]
  14. Dvorak RD, Pearson MR, & Day AM (2014). Ecological momentary assessment of acute alcohol use disorder symptoms: Associations with mood, motives, and use on planned drinking days. Experimental and Clinical Psychopharmacology, 22, 285–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. First MG, Spitzer RL, Gibbon M, & Williams JB (1995). Structured clinical interview for DSM-IV Axis I disorder–patient ed. (SCID-I/P, Version 2). New York: New York State Psychiatric Institute, Biometrics Research Department. [Google Scholar]
  16. Gorka SM, Hedeker D, Piasecki TM, & Mermelstein R (2017). Impact of alcohol use motives and internalizing symptoms on mood changes in response to drinking: An ecological momentary assessment investigation. Drug and Alcohol Dependence, 173, 31–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Grant VV, Stewart SH, O’Connor RM, Blackwell E, & Conrod PJ (2007). Psychometric evaluation of the five-factor Modifed Drinking Motives Questionnaire-Revised in undergraduates. Addictive Behaviors, 32, 2611–2632. [DOI] [PubMed] [Google Scholar]
  18. Grant VV, Stewart SH, & Mohr CD (2009). Coping-anxiety and coping-depression motives predict different daily mood-drinking relationships. Psychology of Addictive Behaviors, 23, 226–237. [DOI] [PubMed] [Google Scholar]
  19. Koob GF, & Le Moal M (2008). Addiction and the brain antireward system. Annual Review of Psychology, 59, 29–53. [DOI] [PubMed] [Google Scholar]
  20. Koob GF, & Volkow ND, (2010). Neurocircuitry of addiction. Neuropsychopharmacology, 35, 217–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kuntsche E, Knibbe R, Gmel G, & Engels R (2005). Why do young people drink? A review of drinking motives. Clinical Psychology Review, 25, 841–861. [DOI] [PubMed] [Google Scholar]
  22. Lane SP, Carpenter RW, Sher KJ, & Trull TJ (2016). Alcohol craving and consumption in borderline personality disorder: When, where, and with whom. Clinical Psychological Science, 4, 775–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lane SP & Hennes EP (2019). Conducting sensitivity analyses to identify and buffer power vulnerabilities in studies examining substance use over time. Addictive Behaviors, 94, 117–123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lenzenweger Lane, Loranger, & Kessler (2007). DSM-IV personality disorders in the National Comorbidity Survey Replication. Biological Psychiatry, 62, 553–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Linehan MM (1993). Cognitive-behavioral treatment of borderline personality disorder. Guilford Press, New York. [Google Scholar]
  26. Merrill JE, Wardell JD, & Read JP (2014). Drinking motives in the prospective prediction of unique alcohol-related consequences in college students. Journal of Studies on Alcohol and Drugs, 75, 93–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Miltenberger RG (2011). Behavior modification: Principles and procedures. Cengage Learning. [Google Scholar]
  28. O’Hara RE, Armeli S, & Tennen H (2014a). Drinking-to-cope motivation and negative mood-drinking contingencies in a daily diary study of college students. Journal of Studies on Alcohol and Drugs, 75, 606–614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. O’Hara RE, Armeli S, & Tennen H (2015). College students’ drinking motives and social-contextual factors: Comparing associations across levels of analysis. Psychology of Addictive Behaviors, 29, 420–429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. O’Hara RE, Boynton MH, Scott DM, Armeli S, Tennen H, Williams C, & Covault J (2014b). Drinking to cope among African American college students: An assessment of episode-specific motives. Psychology of Addictive Behaviors, 28, 671–681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Pfohl B, Blum N, & Zimmerman M (1994). Structured interview for DSM-IV personality disorders. Iowa City: University of Iowa Hospitals and Clinics. [Google Scholar]
  32. Piasecki TM, Cooper ML, Wood PK, Sher KJ, Shiffman S, & Heath AC (2014). Dispositional drinking motives: Associations with appraised alcohol effects and alcohol consumption in an ecological momentary assessment investigation. Psychological Assessment, 26, 363–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Scott-Sheldon LAJ, Terry DL, Carey KB, Garey L, & Carey MP (2012). Efficacy of expectancy challenge interventions to reduce college student drinking: A meta-analytic review. Psychology of Addictive Behaviors, 26, 393–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Sher KJ, & Grekin ER (2007). Alcohol and affect regulation. In Gross JJ (Ed.), Handbook of Emotion Regulation (pp. 560–580). New York, NY, US: The Guilford Press. [Google Scholar]
  35. Sher KJ, & Levenson RW (1982). Risk for alcoholism and individual differences in the stress-response-dampening effect of alcohol. Journal of Abnormal Psychology, 91, 350–367. [DOI] [PubMed] [Google Scholar]
  36. Shrout PE, & Lane SP (2012). Psychometrics. In Mehl MR & Conner TS (Eds.), Handbook of Research Methods for Studying Daily Life (pp. 302–320). New York, NY, US: The Guilford Press. [Google Scholar]
  37. Stevenson BL, Dvorak RD, Kramer MP, Peterson RS, Dunn ME, Leary AV, & Pinto D (2019). Within- and between-person associations from mood to alcohol consequences: The mediating role of enhancement and coping drinking motives. Journal of Abnormal Psychology, 128, 813–822. [DOI] [PubMed] [Google Scholar]
  38. Stone AA, & Shiffman S (1994). Ecological momentary assessment (EMA) in behavioral medicine. Annals of Behavioral Medicine, 16, 199–202. [Google Scholar]
  39. Treloar H, Piasecki TM, McCarthy DM, Sher KJ, & Heath AC (2015). Ecological evidence that affect and perceptions of drink effects depend on alcohol expectancies. Addiction, 110, 1432–1442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Trull TJ, Freeman LK, Vebares TJ, Choate AM, Helle AC, & Wycoff AM (2018). Borderline personality disorder and substance use disorders: An updated review. Borderline Personality Disorder and Emotion Dysregulation, 5, 15–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Trull TJ, Sher KJ, Minks-Brown C, Durbin J, & Burr R (2000). Borderline personality disorder and substance use disorders: A review and integration. Clinical Psychology Review, 20, 235–253. [DOI] [PubMed] [Google Scholar]
  42. Wycoff AM, Carpenter RW, Hepp J, Lane SP, & Trull TJ (2020). Drinking motives moderate daily-life associations between affect and alcohol use in individuals with borderline personality disorder. Psychology of Addictive Behaviors. Advance online publication. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
NIHMS1716291-supplement-Supplemental_Material.pdf (283.3KB, pdf)

RESOURCES