Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jan 1.
Published in final edited form as: Drug Alcohol Depend. 2022 Nov 19;242:109706. doi: 10.1016/j.drugalcdep.2022.109706

Reduced alcohol use increases drink-refusal self-efficacy: Evidence from a contingency management study for DWI arrestees

Erin E Wood 1, Yuanyuan Liang 2, Tae-Joon Moon 1, Alexander M Wasserman 3, R J Lamb 1, John D Roache 1, Nathalie Hill-Kapturczak 1, Donald M Dougherty 4
PMCID: PMC9784176  NIHMSID: NIHMS1855945  PMID: 36493503

Abstract

Background.

Several therapies and interventions to reduce drinking first target drink-refusal self-efficacy (DRSE) to influence drinking behavior. While higher self-efficacy scores are correlated with better outcomes, it is unclear that increased self-efficacy is the causative step leading to improved outcomes. Instead, this correlation may result from reduced drinking that increased self-efficacy. The current study sought to understand how changes in drinking behavior can influence DRSE.

Methods.

Data were from 211 driving while intoxicated (DWI) arrestees participating in an 8-week contingency management (CM) study to reduce drinking. Some of participants were mandated by the courts to wear transdermal alcohol monitoring devices (Mandated group) and some were not mandated (Non Mandated group). All wore a transdermal alcohol monitor during the 8-week study and were randomized to CM or a Control condition stratified by the mandate group. Participants completed weekly assessments of DRSE. Group-based trajectory-modeling identified three drinking behavior trajectory groups.

Results.

While there were no differences in baseline DRSE between the three trajectory groups, participants in the low- and moderate-frequency drinking behavior groups significantly increased DRSE across the study.

Conclusion.

The present study indicates that being able to maintain abstinence or reduce heavy drinking may increase DRSE.

Keywords: Alcohol, Contingency Management, Drink-Refusal Self-Efficacy, SCRAM monitors, Transdermal Alcohol Concentration, Behavioral Intervention

1. Introduction

Heavy alcohol use increases the risk of negative medical, legal, and social consequences (i.e., unemployment, chronic illness, and death; Rehm et al., 2003; White & Hingson, 2013). The National Institutes of Alcohol Abuse (NIAAA; 2022) defines heavy alcohol use as more than 14 drinks per week or 4 drinks per day for men, and more than 7 drinks per week or 3 drinks per occasion for women. In the United States, heavy alcohol use is responsible for approximately 95,000 deaths and upwards of $249 billion in public and private damages each year (NIAAA, 2022; Sacks et al., 2015). Therefore, understanding how individuals come to reduce their heavy drinking is important to aid in the development of more effective interventions.

From early clinical and behavioral research, there has been evidence that changes in thought processes can lead to changes in behavior and that, conversely, changes in behavior can lead to changes in thought processes (Beck, 1970, Ellis, 1962). Relatedly, self-efficacy (i.e., the belief in one’s ability to change behavior) has been considered an important and malleable example of a thought process that can lead to behavior change. Bandura (1989) suggested that self-efficacy plays a crucial and causal role in the likelihood of actual behavior change. According to this hypothesis, an individual’s belief in their ability to reduce heavy drinking precedes any actual changes in this behavior (Sheeran et al., 2016). Additionally, many cognitive-behavioral therapies hypothesize that while changes in thought processes can lead to changes in behavior, changes in behavior can conversely lead to changes in thought processes. Interventions using cognitive-behavioral theory to promote self-efficacy for change commonly include web- or mobile-based brief motivational interventions based on motivational interviewing principles (Field et al., 2020; Fowler et al., 2016; Gajecki et al., 2014; Satre et al., 2016; Voogt et al., 2014). These treatments often utilize goal-oriented motivation to promote change by understanding an individual’s underlying motivations for change (Miller & Rollnick, 2012). Participants may be encouraged to understand their desire to change their behavior in these settings by receiving information regarding normative alcohol use and harms of excessive alcohol use (Pedersen et al., 2017; Terlecki et al., 2014), learning how to enhance their coping skills and self-efficacy (Moos, 2007; Morgenstern & Longabaugh, 2000), or by setting goals to help motivate and sustain abstinence or moderation (Bujarski et al., 2013; Oddo et al., 2021). In an integrative data analysis, Kruger and colleagues (2021) found that treatments including various cognitive-behavioral therapies, medication management, and motivation enhancement treatments were all seen to increase drink-refusal self-efficacy (DRSE: i.e., their confidence in their ability to refuse alcohol) both during and up to a year after the treatment period, but there was no indication of how this influenced drinking behaviors or in turn how self-efficacy was influenced by changes in individual’s drinking. However, Kelly and Greene (2014) also found that for individuals with low DRSE, treatments to increase motivation led to increases in both DRSE and alcohol abstinence at the same time - suggesting that changes in DRSE and alcohol use might be simultaneous. Thus, alternate explanations for the relationship between DRSE and drinking may be warranted.

Such an alternate explanation of the self-efficacy-behavior relationship has emerged in the smoking cessation literature. That literature suggests that behavior change might precede change in self-efficacy. For instance, individuals who abstain from or maintain reduced cigarette use show significantly greater increases in self-efficacy (Clyde et al., 2019; Romanowich et al., 2009; Tolli & Schmidt, 2008). Furthermore, those individuals showing the most behavior change were more likely to maintain their abstinence or reduced use over the long-term (Gwaltney et al., 2005, 2009; Holloway & Watson, 2002; Yzer & Van Den Putte, 2006). While many treatments for alcohol use disorder seek first to increase DRSE in order to decrease alcohol use, the smoking cessation literature argues decreased smoking leads to increased self-efficacy (Gwaltney et al., 2009; McKellar et al., 2008; Romanowich et al., 2009). Although changes in DRSE resulting from reduced drinking have yet to be fully examined, findings within the smoking cessation literature suggest that the longer an individual abstains or maintains reduced drinking, the more they will be: i) confident in their ability to avoid alcohol use, which ii) in turn, increases the likelihood of future success. However, the corollary of this hypothesis is that individuals who were unable to abstain or maintain reduced alcohol use, will also be less likely to believe they will be successful in the future – which in turn increases the likelihood that heavy drinking patterns will be maintained (Holloway & Watson, 2002; Stimmel, 2009).

1.1. Current Study

The current study was a secondary analysis of data collected in a study of DWI arrestees (some mandated to wear a transdermal alcohol monitor and some not) who participated in a contingency management (CM) study to reduce their heavy drinking (Dougherty et al., 2022). In the initial study, participants were randomized into either a CM group or a control group. This resulted in four groups: Non-Mandated Control, Non-Mandated CM, Mandated Control, and Mandated CM. Participants in each CM group were encouraged to reduce alcohol use by receiving incentive payments for each week they maintained abstinence. The purpose of the present study was to examine how DRSE changed as a function of drinking behavior across the 8-week intervention. The analyses of these data considered two overarching hypotheses based on: (a) drinking behavior across the 8-week intervention, and (b) within each of the four study groups. Our first hypothesis was that participants who refrained from heavy drinking across the 8-week study period would have greater increases in DRSE when compared to participants who continued heavy drinking patterns. Our second hypothesis was that participants in the contingency management groups would have greater increases in DRSE than participants in the control group.

2. Methods

2.1. Participants and Criteria

The present analyses used data from a previous study (Dougherty et al., 2022) that was designed to assess the efficacy of CM on drinking behavior in individuals who had been arrested for driving while intoxicated (DWI). The current study used a subsample of 211 participants from the initial study who had data for all weekly assessments.

Participants were recruited from pretrial supervision after being arrested for a DWI and the local provider of transdermal alcohol monitoring (Recovery Monitoring Solutions Corporation). Inclusion criteria included having a DWI, and an Alcohol Use Disorders Identification Test (AUDIT; Babor & Higgins-Biddle, 2001) score >7 for women and >8 for men. Participants were excluded if they had either significant alcohol withdrawal symptoms as indicated by a Clinical Institute Withdrawal Assessment Alcohol Scale Revised (CIWA-Ar; Sullivan et al., 1989) score >10, had a medical (e.g., pregnancy, scheduled surgery) or psychiatric (e.g., inability to comprehend informed consent; presence of a DSM-5 psychiatric disorder with symptoms of psychosis) contra-indicating participation, or were incarcerated. Of the 211 participants, 70 participants were mandated (Mandated) by judicial courts to wear the Secure Continuous Remote Alcohol Monitor (SCRAM, Alcohol Monitoring Systems [AMS], 2021) and 141 participants were not court-mandated (Non-Mandated) but volunteered to wear the SCRAM in order to participate in the study. See Dougherty et al. (2022) for detailed information regarding the court-mandating process for DWI arrestees in this study. The local institutional review board approved this study and all participants provided written informed consent prior to enrollment (ClinicalTrial.gov Identifier: NCT03638596).

2.2. Study Procedure

Participants were randomized within each mandate status group (Mandated vs. Non-Mandated) to either a contingency management (CM) or a control (Control) group using stratified, block randomization to balance groups based on sex and drinking behavior during the 28 days before study entry. This resulted in four groups: Non-Mandated Control (n = 68), Non-Mandated CM (n = 73), Mandated Control (n = 36), and Mandated CM (n = 34). Once randomized, all participants were required to wear a SCRAM to measure transdermal alcohol concentrations (TAC) during the 8-week study. While participants in the Mandated group were already wearing the monitor prior to randomization, participants in the Non-Mandated group were fitted with a transdermal alcohol monitor by study staff. All participants came in once weekly for 8 weeks to have their transdermal alcohol monitoring data read and uploaded, and complete questionnaires regarding their drinking behaviors and attitudes, including self-efficacy, and to receive study-related financial compensation. For a study flow chart that includes an overview of study procedures, please see Figure 1.

Figure 1.

Figure 1.

Open in a new tab

Study flow chart.

NOTE: The Brief Situational Control Questionnaire was used to assess Drink-Refusal Self-Efficacy

All contingency payments for the CM groups were based on the individual participant’s TAC data only. At each weekly visit, participants in the CM groups received $50 if they never exceeded the TAC criteria cut-off on any day during the past week. TAC data were automatically collected every 30 minutes throughout the study and digital data were processed to identify drinking events according to rules described previously (Roache et al., 2015). The CM contingency cut-off was designed to incentivize the absence of any heavy drinking. As previously defined (Karns-Wright et al., 2016; Roache et al., 2019), a heavy drinking event was defined as any positive TAC event that ever had two or more TAC readings > 0.02 g/dL. The 0.02 g/dL cutoff was initially suggested by the manufacturer of the transdermal alcohol monitor (SCRAM, Alcohol Monitoring Systems [AMS], 2021) Research using the SCRAM device in conjunction with a breath alcohol concentration device has found that a TAC reading of 0.02 g/dL corresponds to a breath alcohol concentration of 0.06–0.08 g/dL (Hill-Kapturczak et al., 2014; Karns-Wright, 2016). Participants in the CM group who exceeded the TAC criterion cut-off level received no incentive payment. Weekly contingency payments for the Control groups were yoked to the CM group’s payments (Higgins et al., 2000). For details of the yoking procedure for the current sample, please refer to (Dougherty et al., 2022).

2.3. Measures and Materials

2.3.1. Baseline Variable.

2.3.1.1. Drinking History.

To assess participant drinking prior to study entry, participants completed the Timeline Follow-back (TLFB; Sobell & Sobell, 1992). For the TLFB, participants were asked to describe which days out of the past 28 days they consumed alcohol and the quantity consumed at each drinking event. Based on TLFB results individuals were placed into one of three drinking frequency groups: “no” previous drinking (i.e., reporting alcohol use on 0% of the days included in the TLFB; n = 71), “moderate” previous drinking (i.e., reported alcohol use on 1–30% of the days included in the TLFB; n=85), or “frequent” previous drinking (i.e., reported alcohol use on more than 30% of the days included in the TLFB; n=55).

2.3.2. Outcome Variable.

2.3.2.1. Drink-refusal self-efficacy.

To measure DRSE, participants completed the 8-item Brief Situational Control Questionnaire (Breslin et al., 2000). Participants were asked to rate their perceived self-efficacy in controlling their drinking in eight alcohol-tempting situations (e.g., experiencing unpleasant emotions, social pressure) on a scale from 0–100 (“0” = “not at all confident”, “100” = “totally confident”). For the present study, participants reported their drink-refusal self-efficacy at baseline and each of the weekly visits. For each week, scores were averaged across the eight items. Higher scores indicated greater drink-refusal self-efficacy. The scale is reliable with Cronbach’s α ranging from 0.91 – 0.95 in the present study.

2.3.3. Primary Variables

2.3.3.1. Transdermal Alcohol Monitoring.

SCRAM was used to characterize actual drinking patterns as in previous studies (Dougherty et al., 2012; Hill-Kapturczak et al., 2014) and determine if contingencies were met (in the case of the CM group). The SCRAM was worn on the ankle and used a sensor to detect ethanol excreted through the skin every 30 minutes, 24 hours/day. Deception and tampering were minimized by a locking mechanism and sensors that detected proximity to the skin and body temperature. Transdermal alcohol concentration data was stored on web-based AMS servers which were accessed directly from AMS in a manner approved by our institutional review board. All study-related readings and interpretation of TAC data were conducted independently, and no feedback was provided to pretrial supervision services. The Non-Mandated participants were paid $10/day for wearing the SCRAM while we covered the costs of the SCRAM for Mandated participants (an offset of their costs of $10/day).

2.3.3.2. Heavy Drinking Behavior Trajectory.

To understand how drinking behavior across the 8-week study may have influenced DRSE, group-based trajectory modeling was used to identify three distinct frequency patterns of heavy drinking over time. Using binary (heavy vs. not heavy) drinking data collected from the daily TAC readings, participants were placed into one of three patterns of drinking behavior (Figure 2) using the posterior probabilities and the maximum probability group assignment rule. The Group Based Trajectory Modeling (GBTM) grouping pattern optimized simplicity of interpretation and model fit indices including Bayesian Information Criterion, Akaike Information Criterion, and entropy (Nagin & Odgers, 2010). The three drinking behavior trajectory groups shown in Fig.2 included: i) a low-frequency heavy drinking behavior trajectory group (n=123) who consistently had no or low frequency of heavy drinking over time (i.e., the median number of heavy drinking days = 1 day); ii) a moderate-frequency trajectory group (n=56) who exhibited moderate frequencies of heavy drinking (i.e., a median number of 10 heavy drinking days); and iii) a high-frequency trajectory group (n=32) who commonly drank heavily with a nonlinear trend over time and a median number of 29.5 heavy drinking days.

Figure 2.

Figure 2.

Open in a new tab

Three distinct patterns of drinking behavior identified by group-based trajectory modelling (GBTM) of daily TAC heavy drinking data

2.4. Analytic procedure

Participant baseline characteristics were summarized using descriptive statistics and compared among the three drinking behavior trajectory groups using Kruskal-Wallis H test for continuous variables and Chi-square test or Fisher’s exact test for categorical variables as appropriate. For the primary outcome (i.e., weekly measured DRSE), a linear mixed-effects model (LMM) was used to model the self-efficacy trajectories over time. In the LMM, the time variable was the weekly visit (visits 1 to 8) treated as a continuous variable and different self-efficacy trajectories/slopes were modeled by including the following interaction terms: study group (4 levels) × time, sex × time, baseline age × time, baseline drinking history (3 levels) × time, baseline self-efficacy × time, and drinking behavior trajectory group (3 levels) × time; and the corresponding main effects. Model-based marginal means of self-efficacy were estimated for each subgroup of interest (e.g., study group and drinking behavior trajectory group) at each weekly visit with corresponding 95% confidence intervals. The slopes (i.e., the rates of change) of self-efficacy trajectories and corresponding 95% CIs were calculated and compared between groups based on the LMM. Across the 8-week study period, 18 (8.5%) of participants were lost due to attrition, and missing data was at random. In the LMM, all available data were utilized, no data imputation or model/variable selection was implemented. All analyses were performed using Stata/SE (version 17).

3. Results

Table 1 provides descriptive information for the study sample and by the drinking behavior trajectory groups separately. Of the 211 subjects, the majority were Non-Mandated (66.8%) and male (76.8%) with an average age of 38.2 years. The largest proportion of participants identified as white (48.4%) of Hispanic descent (73.9%). At baseline, participants who had been Mandated to court-monitored abstinence had higher AUDIT scores than participants who were Non-Mandated (M = 18.9 vs. 21.9; t(209) = −2.48, p = 0.01). There were no significant differences in AUDIT scores across the CM and Control groups. When assessed across the three drinking behavior trajectory groups there were no significant differences in age, sex, or race. However, because the three drinking history groups were defined by their frequency of heavy drinking, there were significant differences in the drinking history in the 28 days before study entry (risk of heavy previous drinking: 21.95% vs. 25% vs. 43.75%, p=0.003). Notably, there were no Mandated participants in the high-frequency drinking behavior trajectory group. At baseline, the average reported DRSE across the three drinking trajectory groups was 69.2 on a 100-point scale.

Table 1.

Baseline participants’ characteristics by drinking behavior trajectory group

Variable Low-Frequency Heavy Drinking Group (N = 123) Moderate-Frequency Heavy Drinking Group (N = 56) High-Frequency Heavy Drinking Group (N = 32) Total (N = 211) P-value
Entries are count and column percent [N(%)] unless specified otherwise.
Mandate Status < 0.01 1 ***
Mandated 57 (46.34) 13 (23.21) 0 (0) 70 (33.18)
Non-Mandated 66 (53.66) 43 (76.79) 32 (100) 141 (66.82)
Treatment 0.082
CM 70 (56.91) 25 (44.64) 12 (37.5) 107 (50.71)
Control 53 (43.09) 31 (55.36) 20 (62.5) 104 (49.29)
Study group < 0.01 1 ***
Mandated CM 27 (21.95) 7 (12.5) 0 (0) 34 (16.11)
Non-Mandated CM 43 (34.96) 18 (32.14) 12 (37.5) 73 (34.6)
Mandated Control 30 (24.39) 6 (10.71) 0 (0) 36 (17.06)
Non-Mandated Control 23 (18.7) 25 (44.64) 20 (62.5) 68 (32.23)
Sex 0.482
Female 26 (21.14) 13 (23.21) 10 (31.25) 49 (23.22)
Male 97 (78.86) 43 (76.79) 22 (68.75) 162 (76.78)
Age 0.073
Mean±SD 39.81±11.6 35.95±9.96 35.81±9.39 38.18±11
Median [Q1, Q3] 38 [30, 48.5] 36 [27.75, 43] 36 [28.75, 41] 37 [29, 46]
Min, Max 21, 71 22, 64 20, 62 20, 71
Ethnicity 0.072
Non-Hispanic/Latinx 28 (22.76) 21 (37.5) 6 (18.75) 55 (26.07)
Hispanic/Latinx 95 (77.24) 35 (62.5) 26 (81.25) 156 (73.93)
Race 0.312
White 60 (48.78) 27 (48.21) 15 (46.88) 102 (48.34)
Black 11 (8.94) 10 (17.86) 2 (6.25) 23 (10.9)
American Indian/Alaskan Native 4 (3.25) 1 (1.79) 0 (0) 5 (2.37)
Multi-racial 21 (17.07) 12 (21.43) 5 (15.62) 38 (18.01)
Unknown 27 (21.95) 6 (10.71) 10 (31.25) 43 (20.38)
Drinking History in the 28 days before study entry
No Previous Drinking <0.01 1 ***
Moderate Previous Drinking 52 (42.28) 16 (28.57) 3 (9.38) 71 (33.65)
Frequent Previous Drinking 44 (35.77) 26 (46.43) 15 (46.88) 85 (40.28)
27 (21.95) 14 (25) 14 (43.75) 55 (26.07)
DSRE_baseline
Mean ± SD 71.6±23.34 66.75±24.4 64.31±21.27 69.21±23.4 0.23
Median [Q1, Q3] 72.62 [53.88, 93.81] 66.44 [50, 87.88] 63.69 [49.78, 78.53] 68.75 [50.56, 91.75]
Min, Max 0, 100 5.12, 100 19, 100 0, 100
Open in a new tab

Note: DRSE = Drink Refusal Self Efficacy

***

P-value < 0.001

1.

Fisher’s exact test.

2.

Chi-square test.

3.

Kruskal–Wallis H test

Table 2 provides the estimated LMM coefficients for variables predicting DRSE over time. After adjusting for confounding variables (sex and age), three factors were significant predictors of the rate of change in self-efficacy -namely study group (p=0.002), drinking trajectory group (p=0.003), and baseline self-efficacy (p<0.001). On average, a higher self-efficacy score at baseline was associated with a higher self-efficacy score at each visit but a significantly slower increase in self-efficacy score over time (coefficient for the interaction between baseline self-efficacy and time=−0.04, p<0.001) due possibly to the ceiling effects (Figure 3).

Table 2.

Coefficient estimates of change in DSRE based on linear mixed-effects model

Coefficient 95% confidence Interval p
Study group
 Non-Mandated Control ref -- --
 Mandated Control 4.79 [−3.06, 12.63] .23
 Non-Mandated CM 3.49 [−2.31, 9.29] .24
 Mandated CM 4.48 [−3.12, 12.08] .24
Time, week 5.24 [3.86, 6.63] <.01 ***
Study group × Time
 Non-Mandated Control ref -- --
 Mandated Control −1.20 [−1.98, −.42] <.01 **
 Non-Mandated CM .05 [−.53, .64] .86
 Mandated CM .25 [−.52, 1.01] .53
Sex (Male vs. Female) .92 [−4.59, 6.44] .74
Sex × Time −.41 [−.96, .15] .15
Age, year .04 [−.18, .26] .74
Age × Time −.02 [−.04, .00] .12
Drinking History Group
 No Previous Drinking Group ref -- --
 Moderate Previous Drinking Group 1.94 [−3.87, 7.75] .51
 Frequent Previous Drinking Group 1.39 [−5.37, 8.13] .69
Drinking History Group × Time
 No Previous Drinking Group ref -- --
 Moderate Previous Drinking Group .36 [−.22, .94] .23
 Frequent Previous Drinking Group .55 [−.13, 1.22] .11
DRSE Baseline .78 [−7.18, 4.15] <.01 ***
DRSE Baseline × Time −.04 [−4.38, 10.34] <.01 ***
Drinking behavior trajectory group
Low-Frequency Heavy Drinking Group ref -- --
Medium-Frequency Heavy Drinking Group −1.52 [−7.19, 4.15] .60
High- Frequency Heavy Drinking Group 2.98 [−4.38, 10.34] .43
Drinking behavior trajectory group × Time
 Low- Frequency Heavy Drinking Group ref -- --
 Moderate-Frequency Heavy Drinking Group −.21 [−.77, .35] .46
 High- Frequency Heavy Drinking Group −1.33 [−2.10, −.57] <.01 **
Open in a new tab

NOTE: Drinking History is calculated using drinking reports for 28 days prior to study entry in which “No Previous Drinking” indicated 0% of drinking days for the past 28 days; Moderate: > 0% to <= 30%; Heavy: >30%., Drinking Group is calculated by using TAC data to establish the level of drinking throughout the study; DRSE = Drink Refusal Self Efficacy;

**

P-value < 0.01,

***

P-value < 0.001.

Figure 3.

Figure 3.

Open in a new tab

Patterns of changes in participant-rated self-efficacy across the 8-week study period based on baseline levels of DRSE

3.1. Changes in drink-refusal self-efficacy based on drinking behavior trajectory groups

Baseline DRSE scores did not differ between the three drinking trajectory groups (P>0.10). Figure 4 depicts the model-based marginal mean self-efficacy across time based on the LMM described in Table 2 for the three groups. Participants in the low-frequency group increased their self-efficacy score by 1.33 points per week [95% CI: 1.02 to 1.64], and the moderate-frequency group increased their self-efficacy score by 1.12 points per week [95% CI: 0.67 to 1.56]. In contrast, the DRSE changes (0.01 per week) in the high-frequency group were not significant. Participants in the low- and moderate-frequency groups had significantly greater DRSE change when compared to participants in the high-frequency group (1.33 vs. −0.01, p = .001; 1.12 vs. −0.01, p = 0.005). The average self-reported DRSE at 8 weeks was 81.28, 78.08, and 73.60 for the low-, moderate-, and high-frequency drinking behavior trajectory groups, respectively (81.28 vs. 73.60, p=0.04; 78.08 vs. 73.60, p=0.24).

Figure 4.

Figure 4.

Open in a new tab

Patterns of changes in participant-rated self-efficacy across the 8-week study period based on drinking behavior trajectory group

NOTE: Results are based on the linear mixed-effects model described in Table 2

3.2. Changes in drink-refusal self-efficacy based on study groups

Figure 5 depicts model-based marginal mean self-efficacy at each visit based on the LMM described in Table 2 for each study group. Participants in both the Non-Mandated CM and Non-Mandated Control groups saw increases in DRSE across the study period of 1.48 points per week [95% CI: 0.88 to 2.08] and 1.23 points per week [95% CI: 0.80 to 1.66], respectively. Participants in the Mandated CM group also saw 1.29 points per week [95% CI: 0.88 to 1.69] increase in DRSE across the study. Participants in the Mandated Control group had no significant increase in DRSE (0.03 points per week [95% CI: −0.57 to 0.63]). Non-Mandated CM/Control and Mandated CM participants were found to have significantly greater increases in self-efficacy than participants in the Mandated Control group (all p<0.03).

Figure 5.

Figure 5.

Open in a new tab

Patterns of changes in participant-rated self-efficacy across the 8-week study period based on study group

NOTE: Results are based on the linear mixed-effects model described in Table 2

4. Discussion

The present study examined changes in DRSE measured self-efficacy during a CM study conducted on DWI arrestees. The results supported our first hypothesis that during treatment, individuals who are able to refrain from heavy drinking have greater increases in DRSE than individuals who maintain heavy drinking patterns. We also found partial support for our second hypothesis that participants in the CM group saw greater increases in DRSE than Control. However, participants in the Non-Mandated CM group saw greater change in DRSE than participants in the Mandated CM group which was not expected based on our prior hypotheses.

The findings of this study suggest an interesting relationship between self-efficacy and alcohol use. Grounded in the social learning approach, most of the previous studies have focused on the role of self-efficacy as either as a predictor (e.g., Oei et al., 2007; Sitharthan & Kavanagh, 1990) or a moderator (e.g., Kenney et al., 2014; Miller et al., 2019) of alcohol use and related consequences. However, the results from this study show that the association can be reversed: individuals who abstained or refrained from heavy drinking showed greater increases in DRSE than individuals who maintained heavy drinking patterns. In a similar vein, there has been anecdotal evidence showing that reduced alcohol use precedes increases in self-efficacy (McKellar et al., 2008). For instance, de Visser and Nicholls (2020) found that temporary alcohol abstinence for a month predicted increases in DRSE. Jenzer et al. (2020) also found that the DRSE can be enhanced by the experience of successful control of drinking behavior, not just functioning as a predictor of alcohol use. Additionally, the findings of our research are consistent with observations from studies using CM to reduce smoking in which cross-lagged analysis showed that reductions in smoking predicted increases in self-efficacy, but changes in self-efficacy did not predict changes in smoking (Romanowich et al., 2009). Our findings resonate with these recent studies and suggest an alternate approach to understanding the relationship between self-efficacy and alcohol use.

Furthermore, the present study could indicate that an individual’s previous successes or failures shape one’s beliefs about their future probability of success. In the second leg of our analyses, we found that individuals who were in the Mandated CM group or were Non-Mandated (both CM and Control) saw greater increases in DRSE than participants who were in the Mandated Control group. These results are of interest, specifically when comparing the Mandated participants, because while all Mandated participants were identified as belonging to the low- and moderate-frequency groups (i.e., the groups seen to show the greatest increases in DRSE) only the Mandated CM participants saw increases in DRSE when assessing change in DRSE as a function of the study group. Although assessing the underlying processes which lead to changes in DRSE is outside the scope of the present study, we theorize that differences between the Mandated CM and Control group might have been due to their attributions towards the factors that were associated with their drinking. For example, Mandated participants in the CM group may have had greater increases in DRSE because they attributed their success in reducing heavy drinking to their own ability as they were moderating their drinking to earn incentives. Mandated CM participants might not have seen significant changes in DRSE because they may have attributed their success to the coercive judicial consequences of drinking heavily and not to their own skill at reducing their drinking. However, future research is needed to fully understand how internal and external attributes influence changes in DRSE and drinking behavior.

4.1. Limitations

While the present study provides evidence that changes in drinking behavior may precede changes in self-efficacy (DRSE), it had some limitations which might impact the interpretation of these findings. In the present study, whether a participant was Non-Mandated or Mandated was determined by judges’ personal preferences. Thus, we cannot attribute any differences between the Mandated and Non-Mandated groups to whether individuals were mandated, as the expected a priori outcomes of these individuals are not necessarily the same. Also, due to the nature of the sample, being DWI arrestees who had been instructed to remain abstinent to improve their trial outcomes, regardless of whether they were Mandated or Non-Mandated, probably explains why 33% of participants reported no alcohol use in the 28-days prior to participation and why there were no Mandated participants identified within the continued heavy drinking group. This suggests a bias towards lower-level drinking than would normally be seen in persons who drink heavily, and also suggests a “floor effect” that could explain why CM did not reduce drinking within the Mandated group. Additionally, concerns about legal consequences may also have enhanced the effects of CM within the Non-Mandated group and enhanced self-efficacy within the mandated groups. Furthermore, the static nature of the linear trends seen across the 8-week study period makes the data unamenable to the cross-lagged design needed to assess any bi-directional relationship between changes in drinking behavior and DRSE. Lastly, in the present study, while we found evidence that participants who were able to maintain abstinence or reduced drinking saw increases in DRSE, we were unable to assess whether these changes persisted after the intervention/mandate period.

4.2. Future Directions & Implications

Individuals who have a history of alcohol abuse or have received an alcohol use disorder diagnosis may be hesitant or have barriers to accepting treatment (Mullen et al., 2015). However, the findings of the present study show that interventions such as CM may be a useful catalyst to reduce alcohol use in DWI arrestees. The findings of the current study indicate that incentives to maintain abstinence (i.e., financial compensation for participants in the CM groups) may assist individuals with maintaining abstinence. Ultimately, these incentives may lead an individual to experience more frequent days of abstinence which may lead to increases in DSRE which could promote long-term changes in drinking behaviors. The results of the present study suggest that, for groups who may be more resistant to treatment, external motivations (e.g., financial incentives) may be beneficial in starting the pathway to abstinence even if an individual does not yet have internal motivation (i.e., self-efficacy, desire to change). However, more research is needed to fully understand how behavioral interventions and abstinence can lead to long-term behavior change, particularly in populations resistant to treatment.

4.3. Conclusions

In conclusion, the present study examined the influence that drinking behaviors may have on DRSE in a DWI population within a CM framework. For the individuals in this study, motivation to reduce heavy drinking was likely high given the legal contingencies put upon them in addition to the payments for meeting contingency. The data indicate that reducing alcohol use through contingencies can lead to increases in DRSE which might have the potential to lead to long-term changes in heavy drinking behavior. Future research is needed to better understand how to better serve individuals in the DWI population to reduce heavy drinking behavior. In conclusion, the present findings indicate that to reduce heavy drinking, treatments should consider first targeting the reduction of heavy drinking to increase DRSE—not expect that increased DRSE will reduce heavy drinking.

Highlights.

  • Reduced heavy drinking increases drink-refusal self-efficacy (DRSE).

  • Contingency management (CM) can assist increase DRSE.

  • CM might benefit individuals required to wear a SCRAM after DWI arrest.

Funding:

Research reported in this paper was supported by National Institutes of Health grants R01-AA014988, and T32-DA031115

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Competing interest statement

The authors do not have any competing interests to declare.

The authors have no disclosures.

References

  1. Babor TF, & Higgins-Biddle JC (2001). Brief Intervention for Hazardous and Harmful Drinking: A Manual for Use in Primary Care. World Health Organization, 53. [Google Scholar]
  2. Bandura A (1989). Human agency in social cognitive theory. In American Psychologist (Vol. 44, Issue 9, pp. 1175–1184). American Psychological Association. 10.1037/0003-066X.44.9.1175 [DOI] [PubMed] [Google Scholar]
  3. Beck AT (1970). Cognitive therapy: Nature and relation to behavior therapy. Behav Ther, 1(2), 184–200. 10.1016/S0005-7894(70)80030-2 [DOI] [Google Scholar]
  4. Best JR, Miller PH, & Jones LL (2009). Executive functions after age 5: Changes and correlates. In Developmental Review. 10.1016/j.dr.2009.05.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Breslin FC, Sobell LC, Sobell MB, & Agrawal S (2000). A comparison of a brief and long version of the Situational Confidence Questionnaire. Behaviour Research and Therapy, 38(12), 1211–1220. 10.1016/S0005-7967(99)00152-7 [DOI] [PubMed] [Google Scholar]
  6. Clyde M, Pipe A, Reid R, Els C, & Tulloch H (2019). A bidirectional path analysis model of smoking cessation self-efficacy and concurrent smoking status: impact on abstinence outcomes. Addict Biol, 24(5), 1034–1043. 10.1111/adb.12647 [DOI] [PubMed] [Google Scholar]
  7. de Visser RO, & Nicholls J (2020). Temporary abstinence during Dry January: predictors of success; impact on well-being and self-efficacy. Psychology & Health, 35(11), 1293–1305. 10.1080/08870446.2020.1743840 [DOI] [PubMed] [Google Scholar]
  8. Dougherty DM, Moon T-J, Liang Y, Roache JD, Lamb R, Mathias CW, Wood E, Wasserman A, & Hill-Kapturczak N (2022). Effectiveness of contingency management using transdermal alcohol monitoring to reduce heavy drinking among driving while intoxicated (DWI) arrestees. Alcohol Clin Exp Res, 46(S1), 70–289. 10.1111/acer.14833 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dougherty DM, Charles NE, Acheson A, John S, Furr RM, & Hill-Kapturczak N (2012). Comparing the detection of transdermal and breath alcohol concentrations during periods of alcohol consumption ranging from moderate drinking to binge drinking. Exp Clin Psychopharmacol, 20(5), 373–381. 10.1037/a0029021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ellis A (1962) Reason and emotion in psychotherapy. Secaucus, NJ: Lyle Stuart and Citadel Press. [Google Scholar]
  11. Field CA, Richards DK, Castro Y, Alonso Cabriales J, Wagler A, & von Sternberg K (2020). The Effects of a Brief Motivational Intervention for Alcohol Use through Stages of Change among Nontreatment Seeking Injured Patients. Alcohol Clin Exp Res, 44(11), 2361–2372. 10.1111/acer.14466 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fowler LA, Holt SL, & Joshi D (2016). Mobile technology-based interventions for adult users of alcohol: A systematic review of the literature. Addict Behav, 62, 25–34. 10.1016/J.ADDBEH.2016.06.008 [DOI] [PubMed] [Google Scholar]
  13. Gajecki M, Berman AH, Sinadinovic K, Rosendahl I, & Andersson C (2014). Mobile phone brief intervention applications for risky alcohol use among university students: a randomized controlled study. Addict Sci Clin Pract, 9(1), 11. 10.1186/1940-0640-9-11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gwaltney CJ, Metrik J, Kahler CW, & Shiffman S (2009). Self-efficacy and smoking cessation: a meta-analysis. Psychol Addict Behav, 23(1), 56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gwaltney CJ, Shiffman S, Balabanis MH, & Paty JA (2005). Dynamic self-efficacy and outcome expectancies: prediction of smoking lapse and relapse. J Abnorm Psychol, 114(4), 661. [DOI] [PubMed] [Google Scholar]
  16. Higgins ST, Wong CJ, Badger GJ, Haug Ogden DE, & Dantona RL (2000). Contingent reinforcement increases cocaine abstinence during outpatient treatment and 1 year of follow-up. J Consult Clin Psychol, 68(1), 64–72. 10.1037/0022-006X.68.1.64 [DOI] [PubMed] [Google Scholar]
  17. Hill-Kapturczak N, Lake SL, Roache JD, Cates SE, Liang Y, & Dougherty DM (2014). Do variable rates of alcohol drinking alter the ability to use transdermal alcohol monitors to estimate peak breath alcohol and total number of drinks? Alcohol Clin Exp Res, 38(10), 2517–2522. 10.1111/acer.12528 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Holloway A, & Watson HE (2002). Role of self-efficacy and behaviour change. Int J Nurs Pract, 8(2), 106–115. 10.1046/j.1440-172x.2002.00352.x [DOI] [PubMed] [Google Scholar]
  19. Iguchi MY, Stitzer ML, Bigelow GE, & Liebson IA (1988). Contingency management in methadone maintenance: effects of reinforcing and aversive consequences on illicit polydrug use. Drug Alcohol Depend, 22(1–2), 1–7. 10.1016/0376-8716(88)90030-0 [DOI] [PubMed] [Google Scholar]
  20. Jenzer T, Egerton GA, & Read JP (2021). Learning from drinking experiences in college: A test of reciprocal determinism with drinking refusal self-efficacy. Psychol Addict Behav, 35(1), 85–92. 10.1037/adb0000675 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Karns-Wright TE, Roache JD, Hill-Kapturczak N, Liang Y, Mullen J, & Dougherty DM (2016). Time Delays in Transdermal Alcohol Concentrations Relative to Breath Alcohol Concentrations. Alcohol Alcohol, 52(1), 35–41. 10.1093/alcalc/agw058 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kelly JF, & Greene MC (2014). Where there’sa will there’sa way: a longitudinal investigation of the interplay between recovery motivation and self-efficacy in predicting treatment outcome. Psychol Addict Behav, 28(3), 928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kenney SR, Napper LE, & LaBrie JW (2014). Social anxiety and drinking refusal self-efficacy moderate the relationship between drinking game participation and alcohol-related consequences. Am J Drug Alcohol Abuse, 40(5), 388–394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kruger ES, Serier KN, Pfund RA, McKay JR, & Witkiewitz K (2021). Integrative data analysis of self-efficacy in 4 clinical trials for alcohol use disorder. Alcohol Clin Exp Res, 45(11), 2347–2356. 10.1111/acer.14713 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. McKellar J, Ilgen M, Moos BS, & Moos R (2008). Predictors of changes in alcohol-related self-efficacy over 16 years. J Subst Abuse Treat, 35(2), 148–155. 10.1016/j.jsat.2007.09.003 [DOI] [PubMed] [Google Scholar]
  26. Miller CM, Whitley RB, Scully KA, Madson MB, & Zeigler-Hill V (2019). Protective behavioral strategies and alcohol-related outcomes: The moderating effects of drinking refusal self-efficacy and sex. Addictive Behaviors, 99, 106110. 10.1016/j.addbeh.2019.106110 [DOI] [PubMed] [Google Scholar]
  27. Miller WR, & Rollnick S (2012). Motivational interviewing: Helping people change. Guilford press. [Google Scholar]
  28. Moos RH (2007). Theory-based active ingredients of effective treatments for substance use disorders. Drug Alcohol Depend, 88(2–3), 109–121. 10.1016/J.DRUGALCDEP.2006.10.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Morgenstern J, & Longabaugh R (2000). Cognitive–behavioral treatment for alcohol dependence: A review of evidence for its hypothesized mechanisms of action. Addiction, 95(10), 1475–1490. [DOI] [PubMed] [Google Scholar]
  30. Mullen J, Ryan SR, Mathias CW, Dougherty DM (2015) Treatment needs of driving while intoxicated offenders: The need for a multimodal approach to treatment. Traffic Injury Prev. 16:637–644. 10.1080/15389588.2015.1013189 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nagin DS, & Odgers CL (2010). Group-based trajectory modeling in clinical research. Annu Rev Clin Psychol, 6(1), 109–138. [DOI] [PubMed] [Google Scholar]
  32. National Institute of Alcohol Abuse and Alcoholism. (Updated 2022a). Alcohol use in the United States. Retrieved from https://www.niaaa.nih.gov/publications/brochures-and-fact-sheets/alcohol-facts-and-statistics
  33. National Institute of Alcohol Abuse and Alcoholism. (Updated 2022b). The basics: defining how much alcohol is too much Retrieved from https://www.niaaa.nih.gov/health-professionals-communities/core-resource-on-alcohol/basics-defining-how-much-alcohol-too-much
  34. Oei TP, Hasking P, & Phillips L (2007). A comparison of general self-efficacy and drinking refusal self-efficacy in predicting drinking behavior. Am J Drug Alcohol Abuse, 33(6), 833–841. [DOI] [PubMed] [Google Scholar]
  35. Pedersen ER, Parast L, Marshall GN, Schell TL, & Neighbors C (2017). A randomized controlled trial of a web-based, personalized normative feedback alcohol intervention for young-adult veterans. J Consult Clin Psychol, 85(5), 459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Rehm J, Room R, Graham K, Monteiro M, Gmel G, & Sempos CT (2003). The relationship of average volume of alcohol consumption and patterns of drinking to burden of disease: an overview. Addiction, 98(9), 1209–1228. [DOI] [PubMed] [Google Scholar]
  37. Roache JD, Karns-Wright TE, Goros M, Hill-Kapturczak N, Mathias CW, & Dougherty DM (2019). Processing transdermal alcohol concentration (TAC) data to detect low-level drinking. Alcohol, 81, 101–110. 10.1016/J.ALCOHOL.2018.08.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Roache JD, Karns TE, Hill-Kapturczak N, Mullen J, Liang Y, Lamb RJ, & Dougherty DM (2015). Using Transdermal Alcohol Monitoring to Detect Low-Level Drinking. Alcohol Clin Exp Res, 39(7), 1120–1127. 10.1111/acer.12750 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Romanowich P, Mintz J, & Lamb RJ (2009). The relationship between self-efficacy and reductions in smoking in a contingency management procedure. Exp Clin Psychopharmacol, 17(3), 139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sacks JJ, Gonzales KR, Bouchery EE, Tomedi LE, & Brewer RD (2015). 2010 National and State Costs of Excessive Alcohol Consumption. Am J Prev Med, 49(5), e73–e79. 10.1016/J.AMEPRE.2015.05.031 [DOI] [PubMed] [Google Scholar]
  41. Satre DD, Leibowitz A, Sterling SA, Lu Y, Travis A, & Weisner C (2016). A randomized clinical trial of Motivational Interviewing to reduce alcohol and drug use among patients with depression. J Consult Clin Psychol, 84(7), 571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Sheeran P, Maki A, Montanaro E, Avishai-Yitshak A, Bryan A, Klein WMP, Miles E, & Rothman AJ (2016). The impact of changing attitudes, norms, and self-efficacy on health-related intentions and behavior: A meta-analysis. Health Psychol, 35(11), 1178–1188. 10.1037/hea0000387 [DOI] [PubMed] [Google Scholar]
  43. Sitharthan T, & Kavanagh DJ (1990). Role of self-efficacy in predicting outcomes from a programme for controlled drinking. Drug Alcohol Depend, 27, 87–94. [DOI] [PubMed] [Google Scholar]
  44. Sobell LC, & Sobell MB (1992). Timeline Follow-Back BT - Measuring Alcohol Consumption: Psychosocial and Biochemical Methods (Litten RZ & Allen JP (eds.); pp. 41–72). Humana Press. 10.1007/978-1-4612-0357-5_3 [DOI] [Google Scholar]
  45. Stimmel B (2009). From addiction to abstinence: Maximizing the chances of success. Fam Court Rev, 47(2), 265–273. [Google Scholar]
  46. Sullivan JT, Sykora K, Schneiderman J, Naranjo CA, & Sellers EM (1989). Assessment of alcohol withdrawal: the revised clinical institute withdrawal assessment for alcohol scale (CIWA-Ar). Br J Addict, 84(11), 1353–1357. [DOI] [PubMed] [Google Scholar]
  47. Terlecki MA, Ecker AH, & Buckner JD (2014). College Drinking Problems and Social Anxiety: The Importance of Drinking Context HHS Public Access. Psychol Addict Behav, 28(2), 545–552. 10.1037/a0035770 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tolli AP, & Schmidt AM (2008). The Role of Feedback, Causal Attributions, and Self-Efficacy in Goal Revision. J Appl Psychol, 93(3), 692–701. 10.1037/0021-9010.93.3.692 [DOI] [PubMed] [Google Scholar]
  49. Voogt CV, Kuntsche E, Kleinjan M, & Engels RCME (2014). The effect of the ‘What Do You Drink’ web-based brief alcohol intervention on self-efficacy to better understand changes in alcohol use over time: Randomized controlled trial using ecological momentary assessment. Drug Alcohol Depend, 138(1), 89–97. 10.1016/J.DRUGALCDEP.2014.02.009 [DOI] [PubMed] [Google Scholar]
  50. White A, & Hingson R (2013). The burden of alcohol use: excessive alcohol consumption and related consequences among college students. Alcohol Res. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Yzer MC, & Van Den Putte B (2006). Understanding smoking cessation: The role of smokers’ quit history. Psychol Addict Behav, 20(3), 356. [DOI] [PubMed] [Google Scholar]

RESOURCES