Skip to main content
NCBI home page
Journal of Studies on Alcohol and Drugs logoLink to Journal of Studies on Alcohol and Drugs
. 2022 Sep 21;83(5):640–645. doi: 10.15288/jsad.21-00207

Who Opts In to Alcohol Feedback and How Does That Impact Behavior? A Pilot Trial

Cassandra l Boness a,,*, Ashley C Helle b, Mary Beth Miller b,,c, Melissa Gordon Wolf d, Kenneth J Sher b
PMCID: PMC9523752  PMID: 36136433

Abstract

Objective:

Personalized feedback interventions are effective in reducing alcohol consumption and related problems. However, little is known about the role of choice in outcomes. The current study sought to (a) characterize individuals who opt in for brief alcohol-related feedback, (b) assess participants’ consistency in that choice over two time points, and (c) evaluate changes in peak alcohol consumption among those who did and did not receive feedback.

Method:

Participants reporting past-12-month alcohol consumption were recruited through Prolific. At the outset of the survey, participants were asked if they would like to receive feedback on their drinking at the end of the survey (“opt in”). Participants at Time 1 (T1; N = 732) were 41% female, 91% White, and 8% Hispanic (mean age = 36, SD = 12.25, range: 18–80). A subset was invited back for a 30-day retest (Time 2 [T2]; n = 234).

Results:

Those reporting higher maximum drinks and more drug use were more likely to opt in to feedback than those with lower use. Further, 85% of participants were consistent in their choice of whether to receive feedback across T1 and T2 (κ = .65). Among heavy drinking participants with T1 and T2 data (n = 163), there was an effect of feedback on intensity of consumption at T2.

Conclusions:

Individuals who engage in heavy alcohol use are more likely to opt in to personalized alcohol feedback, and most do so consistently. Among heavy drinkers, feedback at T1 reduced intensity of consumption at T2, but the effect was small and requires future replication in more diverse samples.

Personalized feedback interventions (PFIs) prompt consideration of future drinking by highlighting discrepancies between personal drinking and others’ drinking, reframing use in terms of other costs (e.g., calories), and/or evaluating risk in terms of some standard (Miller et al., 2013). Given the range of ways to prompt such consideration, PFIs vary considerably in content (Miller et al., 2013; Ray et al., 2014). Still, brief interventions incorporating personalized feedback are among the most effective alcohol prevention and treatment strategies for heavy-drinking young adults (Tanner-Smith & Lipsey, 2015).

The efficacy of brief alcohol interventions involving personalized feedback has been studied extensively among young adults, with most studies demonstrating small but significant reductions in alcohol use, alcohol-related consequences, or both (Black et al., 2016; Carey et al., 2012; Cole et al., 2018; Huh et al., 2015; Tanner-Smith & Lipsey, 2015). PFIs also demonstrate consistency across modalities, delivery sites, formats, and length (Tanner-Smith & Lipsey, 2015), although there is some evidence that PFI effects may be moderated by sex (e.g., Doumas et al., 2020). However, few studies have examined the efficacy of a single-component PFI in reducing subsequent drinking behaviors (Ray et al., 2014).

It is uncommon in treatment trials to continue collecting data from PFI participants who “opt out” of feedback. That is, participants opt out of a study by declining or withdrawing participation, but then relevant outcomes, such as alcohol use, are not assessed. Someone who saw an advertisement for drinking feedback on a digital marketing campaign would have to click a link or provide personal data before receiving feedback. Similarly, in a doctor's office, a person would have to offer drinking information before being flagged for potential intervention. These contingencies provide people multiple opportunities to opt out of intervention.

Because the opt-out option in research is typically withdrawal of all future participation, current trials are unable to characterize those who do not opt in for alcohol feedback in the first place.

This study extends previous research by examining factors associated with choosing to receive drinking feedback. First, we aimed to better understand how those who do and do not choose to receive feedback differ in terms of sociodemographics, substance use patterns, and mental health symptoms. Second, we hypothesized that people who opted in to feedback at baseline (Time [T] 1) would also opt in 30 days later (T2). Finally, we hypothesized that heavy drinkers who opted in to feedback at T1 would report greater reductions in drinking intensity at T2 than those who did not opt in.

Method

Participants were recruited through Prolific to participate in a larger study on diagnostic measures of alcohol use disorder (AUD). Respondents on Prolific (https://www.prolific.co), an online research subject pool, completed a pre-screening survey that is used to determine their eligibility for subsequent studies. Participants who reported that they were 18 years or older, U.S. residents, and consumed an average of 10 or more units of alcohol1 per week at pre-screening were invited to participate. Because the date of pre-screening was unknown, participants were also required to report alcohol use in the past 12 months at T1.

After providing informed consent, participants were asked: “As part of participating in this study, you have the option to receive personalized feedback on your alcohol use behaviors. Would you like to receive feedback on your alcohol use? If you choose, ‘Yes, I would like to receive feedback,’ you will be provided this information at the end of the survey.” Participants subsequently completed T1 measures, and then feedback was presented to those who opted in (see Appendix A for sample materials). (A supplemental appendix appears as an online-only addendum to this article on the journal's website.) Participants were debriefed on the goals of the study at the end of T1.

Participants who completed the T1 assessment within the first 48 hours of data collection (May 2020) were eligible to participate in a 30-day retest at T2 (June 2020). Because of funding constraints, only the first 300 individuals to sign up were able to participate at T2. Participants were compensated $6.50 for each time point. All procedures were approved by the institutional review board.

A total of 833 individuals completed the T1 survey. However, 101 participants were excluded because they denied past-12-month alcohol use (n = 5), provided incomplete data (n = 78), failed more than 80% of attention checks (n = 10), demonstrated implausible response time (n = 1), or provided illogical responses (n = 7). Of the 732 participants with valid responses for T1, 264 completed T2. However, 30 T2 participants were excluded because they denied past-year alcohol use (n = 2), provided incomplete data (n = 12), failed attention checks (n = 7), demonstrated implausible response time (n = 2), or provided illogical responses (n = 7), resulting in a T2 sample of 234. Participant sociodemographics are provided in Table 1.

Table 1.

Participant characteristics by time point and feedback group

graphic file with name jsad.21-00207tbl1.jpg

Variable Full sample No feedback at T1 Feedback at T1 Differences
T1 (N =732) M (SD) or n (%) T2 (n =234) M (SD) or n (%) T1 (n =232) M (SD) or n (%) T2 (n =75) M (SD) or n (%) T1 (n =500) M (SD) or n (%) T2 (n =159) M (SD) or n (%) No feedback vs. feedback at T1 t or χ2 (p)
Age 36.13 (12.25) 38.25 (12.12) 38.19 (12.89) 39.43 (12.00) 35.17 (11.84) 37.69 (12.19) 3.01 (.003)
Sex
 Female 302 (41%) 94 (40%) 95 (41%) 30 (40%) 207 (41%) 64 (40%) 0.01 (.94)
 Male 428 (58%) 140 (60%) 137 (59%) 45 (60%) 291 (58%) 95 (60%) 0.01 (.94)
Racial/ethnic identity
 White Black/African 666 (91%) 208 (89%) 209 (91%) 66 (88%) 457 (91%) 142 (89%) 0.19 (.66)
 American Hispanic or Latino/ 40 (5%) 14 (6%) 15 (6%) 5 (7%) 25 (5%) 9 (6%) 0.41 (.52)
 Latina 57 (8%) 14 (6%) 10 (8%) 3 (4%) 47 (9%) 11 (7%) 5.02 (.02)
Substance use
 Age at first full drink 12-mo. drinking 16.20 (2.76) 16.30 (3.02) 16.35 (2.69) 16.57 (2.57) 16.13 (2.79) 16.18 (3.21) 1.00 (.32)
 frequency 12-mo. drinking 3.99 (2.01) 3.90 (2.06) 3.92 (2.10) 3.93 (2.12) 4.02 (1.97) 3.88 (2.03) -0.64 (.52)
 quantity 12-mo. drunk 5.37 (3.87) 5.18 (3.42) 5.03 (3.28) 4.63 (2.92) 5.53 (4.11) 5.44 (3.61) -1.78 (.08)
frequency 12-mo. binge 1.63 (1.73) 1.45 (1.70) 1.57 (1.76) 1.53 (1.83) 1.67 (1.71) 1.41 (1.63) -0.73 (.47)
 frequency 12-mo. frequency 1.49 (1.90) 1.61 (2.00) 1.27 (1.90) 1.50 (2.05) 1.60 (1.89) 1.66 (1.98) -2.17 (.03)
  ≥12 drinks 0.72 (1.49) 0.61 (1.32) 0.66 (1.56) 0.48 (1.23) 0.75 (1.46) 0.67 (1.36) -0.70 (.49)
  30-day max. drinksa 7.95 (5.69) 7.90 (5.16) 6.90 (4.57) 6.81 (4.34) 8.44 (6.09) 8.40 (5.43) −3.79 (<.001)
  12-mo. max. drinksa 11.73 (6.92) 11.56 (6.72) 10.41 (6.25) 9.96 (5.93) 12.35 (7.13) 12.31 (6.95) −3.73 (<.001)
  Lifetime max. drinksa 16.10 (8.00) 15.82 (8.19) 15.13 (7.66) 14.47 (7.56) 16.56 (8.12) 16.47 (8.42) -2.26 (.02)
Family history AUD 322 (44%) 89 (38%) 98 (42%) 27 (36%) 224 (45%) 62 (39%) 0.38 (.54)
AUDIT 12.26 (7.38) 11.58 (7.13) 11.17 (7.27) 10.76 (7.36) 12.77 (7.39) 11.97 (7.01) -2.74 (.006)
DUDIT 4.92 (7.93) 3.81 (6.85) 3.70 (6.20) 3.20 (6.05) 5.49 (8.57) 4.10 (7.20) −3.14 (.002)
Psychiatric symptoms
  Anxiety (GAD-7) 7.80 (5.91) 6.70 (5.70) 7.26 (6.03) 5.40 (5.26) 8.05 (5.84) 7.31 (5.81) −1.67 (.10)
  Depression (PHQ-9) 7.66 (6.56) 6.87 (6.35) 6.68 (5.93) 6.04 (5.86) 8.11 (6.79) 7.26 (6.54) −2.89 (.004)
Open in a new tab

Notes: The “No feedback” columns characterize the differences between those who did not opt in for feedback at Time 1 across the two time points. The “Feedback” columns characterize the differences between those individuals who did opt in for feedback at Time 1 across the two time points. Bonferroni-corrected p value = .003 (.05/20). Significant tests are displayed in bold. T = Time; mo. = month; max. = maximum; AUD = alcohol use disorder; AUDIT = Alcohol Use Disorder Identification Test; DUDIT = Drug Use Disorder Identification Test; GAD-7 = Generalized Anxiety Disorder 7-item; PHQ-9 = Patient Health Questionnaire 9-item. Ns may not add up to overall N if participants chose not to answer the question.

a

Values over 40 were Winsorized to 40.

Self-report measures

Alcohol consumption. Participants reported on their alcohol use in the past 12 months. Items assessed quantity/ frequency of drinking, frequency of binge drinking (5+/4+ drinks for males/females in a 2-hour period on one day), frequency of feeling drunk, and frequency of 12 or more drinks in a single sitting. Participants also reported the maximum number of drinks they consumed within a 24-hour period for the past 12 months, past 30 days, and lifetime. Quantity, frequency, and binge frequency items were adapted from the National Institute on Alcohol Abuse and Alcoholism's (NIAAA) recommended alcohol consumption questions (NIAAA, 2003) and the Intensive, Multivariate, Prospective Alcohol College-Transition Study survey (Sher & Rutledge, 2007). Maximum drinks variables were Winsorized to 40 drinks (Cooper & Weekes, 1983). Because maximum drinking quantity was the only variable assessed over a 30-day timeframe, this was the only variable for which we examined changes over time.

Participants completed the 10-item Alcohol Use Disorders Identification Test (AUDIT; Saunders et al., 1993), which assesses past-12-month alcohol consumption and risky drinking (range: 1–40). Higher scores are associated with a greater likelihood of AUD, and scores greater than 7 are indicative of hazardous alcohol use (Babor et al., 2001; Saunders et al., 1993).

Drug use. Participants completed the 11-item Drug Use Disorders Identification Test (DUDIT; Bergman et al., 2003), which assesses drug consumption and harmful use. Higher scores indicate more drug-related problems (range: 0–44).

Mental health symptoms. Participants completed the Generalized Anxiety Disorder-7 (GAD-7; Spitzer et al., 2006) and Patient Health Questionnaire-9 (PHQ-9; Kroenke et al., 2001). Both are screening tools on which higher scores indicate more severe anxiety (range: 0–21) or depression (range: 0–27), respectively, in the previous 2 weeks.

Alcohol feedback

Individuals who chose to receive feedback at T1 were presented with their AUDIT score and a description of the risk level for their score: 0–7 = low risk of alcohol-related harm; 8–15 = moderate risk; 16–19 = high risk (drinking that will eventually result in harm, if not already doing so); and ≥20 = high risk (definite harm, also likely to have AUD; Babor et al., 2001; Saunders et al., 1993).

Analyses

All relevant code and data can be found at https://osf.io/7szw9. Differences between those choosing and not choosing to receive feedback were examined using a series of two-sample t tests and chi-square tests in R version 4.0.2 (R Core Team, 2020). Bonferroni correction was applied to correct for multiple tests (p = .05 / 20 = .003; see variables in Table 1).

To determine whether those who chose to receive feedback at T1 would also be likely to choose to receive feedback at T2, we estimated Cohen's kappa using the R psych package (Revelle, 2021).

Finally, to determine changes in maximum drinks from T1 to T2, a mixed-effects model was used to test the fixed effects of sex, feedback, time, and the Feedback × Time interaction in the subsample of heavy-drinking participants with data at both time points (n = 163). Participant was treated as a random effect to account for the repeated assessment of past-30-day maximum drinks. Sex was included as a covariate, given known sex differences in consumption, rates of absorption, and response to PFIs. To estimate this model, we used the lmer() function in the “lme4” R package (R Core Team, 2020).

Results

Comparisons between those choosing and not choosing to receive feedback

Those with higher maximum drinks in the past 30 days, t(582.45) = -3.79, p < .001, and 12 months, t(508.40) = -3.73, p < .001, were more likely to opt in to feedback than those with lower maximum drinks. Similarly, individuals with higher DUDIT scores (M = 5.49, SD = 8.57) were more likely to opt in to feedback than those with lower scores (M = 3.70, SD = 6.20; t(579.56) = -3.14, p = .002). Results are presented in Table 1.

Consistency across time points

Among participants with data for both time points (n = 234), 84.62% were consistent in their choice of whether to receive feedback. Specifically, 59.40% chose feedback at both time points, and 25.21% did not choose to receive feedback at both time points. Of those who were inconsistent in their responses, 8.55% (n = 20) wanted T1 but not T2 feedback, and 6.84% (n = 16) wanted T2 but not T1 feedback.

Cohen's κ was .65, indicating high agreement between T1 and T2 choices.2

Change in drinking intensity

Change in maximum drinks from T1 to T2 was assessed among heavy drinkers who provided data at both time points (n = 163). Relative to those who did not receive feedback, participants who received feedback reported small but significant reductions in past-30-day maximum drinks from T1 to T2, b = -1.93, SE = 0.89, t(158.00) = -2.15, p = .03; η2 = .03 (Figure 1). Men also reported a greater number of maximum drinks than women, b = 2.64, SE = 0.76, t(158.00) = 3.46, p < .001, and individuals who opted in for feedback at T1 reported higher maximum drinks than those who did not, b = 5.59, SE = 1.56, t(260.44) = 3.46, p < .001. Descriptive statistics for groups across time points are provided in Supplemental Table 1.

Figure 1.

Figure 1

Open in a new tab

Raw means by Time 1 feedback group for past-30-day maximum drinks across Time 1 and Time 2 for all participants with Time 1 and Time 2 data (n = 163). This figure displays means for past-30-day maximum drinks across the two time points for those who received feedback and those who did not. Means correspond to those presented in Supplemental Table 1. Error bars represent standard errors. Past-30-day maximum drinks at both time points have been Winsorized to 40.

Discussion

Research has largely failed to consider who opts in for alcohol feedback, given the choice. This study demonstrates that people who do and do not opt in for feedback vary on key characteristics. The finding that people with greater alcohol and drug use risk scores were more likely to opt in for feedback is unsurprising, given literature suggesting that individuals with heavier consumption and more alcohol-related problems may be more likely to volunteer for treatment research (Strohmetz et al., 1990). This is important because it suggests that individuals at high risk for harmful substance use are receptive to receiving drinking feedback. It also challenges the notion that people who use substances are defensive or unwilling to receive feedback about their use. Moreover, it suggests that brief personalized feedback delivered online is a realistic way to improve the reach of brief alcohol interventions for those at risk of alcohol-related harm.

The finding that those who opted in for feedback at T1 tended to opt in for feedback again at T2 may have both positive and negative consequences for prevention and intervention efforts. That is, people who are open to feedback may remain open across time, whereas people who are unreceptive to feedback may remain unreceptive. In this case, intervention efforts may focus on increasing one's openness to feedback, perhaps using methods such as motivational interviewing (Miller & Rollnick, 2013). Of note, a small percentage (~7%) of participants who did not opt in for feedback at T1 did opt in at T2, raising the question of what might have changed and/or influenced their decision to opt in the second time. We did not identify any significant predictors of this “inconsistency” in responding. This is an important area for future longitudinal and event-level research, as it may point to an ideal “time” to provide feedback.

Although participants who opted in for feedback reduced the intensity of their drinking at 30-day follow-up, the size of the effect was small. As such, it is worth considering the type and “dose” of feedback provided. Feedback in this study included participants’AUDIT score and corresponding risk level (e.g., low risk), which is an exceptionally minimal amount of feedback. In their integrative analysis of more than 6,000 participants and 31 brief intervention conditions, Ray and colleagues (2014) found an interaction between the number of feedback components and the extent to which feedback was personalized. Specifically, more content was preferable for highly personalized feedback, but less content was preferable for less personalized feedback. In this study, the single intervention component seems to have been sufficient to reduce drinking intensity among high-risk drinkers who elected to receive feedback. However, the single intervention component may also have contributed to the small effect size observed.

Limitations and future directions

This study improves our understanding of who might be open to drinking feedback and provides preliminary evidence that even single-component feedback may have an impact on drinking behavior. However, this study was not without limitations. Because maximum drinks was the only outcome we measured using a 30-day timeframe, this is the only outcome for which we could assess intervention effects. We encourage future researchers to use consensus outcomes that facilitate synthesis of findings across studies (e.g., typical quantity/frequency, alcohol-related consequences; Shorter et al., 2021). There was a significant between-group difference in maximum drinks at T1, in which case changes in both groups may represent regression to the mean.

Regarding the intervention itself, we also asked twice about participant interest in receiving feedback—once at T1 and again at T2. Although participants were largely consistent in their choices, we did not directly assess factors that influenced decisions to receive feedback. For example, if a participant received feedback at T1, they might not want identical feedback at T2. Thus, “consistency” estimates may be limited.

Generalizability may also be limited. Specifically, although we aimed to characterize who did and did not choose personalized feedback, individuals who participate in research in general may be different from those who do not (Henrich et al., 2010; Nielsen et al., 2017). Second, brief feedback in a research study may be perceived as relatively “low stakes” compared with being asked by a provider about one's alcohol use. Thus, it is unclear if results would generalize to traditional clinical settings. Finally, the sample was relatively homogeneous with respect to racial and ethnic identity, and we did not assess the geographical location of participants. Future research examining minimal alcohol feedback across more diverse samples may elucidate what type and how much feedback is associated with decreases in alcohol consumption across populations.

Conclusion

Findings from this study suggest that individuals in greatest need of intervention (those reporting hazardous drinking and other drug use) are more likely than lower-risk adults to opt in for alcohol feedback. Those who opt in for feedback are also likely to opt in again later, suggesting that it may be viable to recruit high-risk drinkers online for a brief web-based intervention approach. Although data are preliminary, heavy drinkers who opted in for feedback reported greater short-term decreases in drinking intensity than those who did not receive feedback. Thus, additional research examining the efficacy and effectiveness of minimal feedback interventions in diverse settings is warranted.

Footnotes

1

Prolific is a United Kingdom–based company that measures alcohol in units, rather than standard drinks (1 unit of alcohol = 1 small glass of wine, half pint of beer, pub measure of spirits).

2

We also tested a post hoc model with T1 feedback choice, T1 consumption (AUDIT), T2 consumption (AUDIT), age, sex, race, ethnicity, family history of AUD, DUDIT, GAD-7, and PHQ-9 as predictors of T2 feedback choice to predict the “inconsistency” in the choice of feedback across time points. None of these predictors were significant.

This research was partially funded by a Student Research Fellowship from the American Psychological Association's Division 12, Section IX (Assessment) to Cassandra L. Boness. Investigator effort was supported in part by National Institute on Alcohol Abuse and Alcoholism grants F31AA026177 (principal investigator: Cassandra L. Boness) and K08AA028543 (principal investigator: Ashley C. Helle). Funding sources had no involvement in the conduct of the research or preparation of the manuscript.

This manuscript has been posted to PsyArvix: https://osf.io/g9r46. Supplemental materials including data and R code can be found here: https://osf.io/7szw9.

References

  1. Babor T. F., de la Fuente J. R., Saunders J., Grant M. Geneva, Switzerland: World Health Organization; 2001. The Alcohol Use Disorders Identification Test: Guidelines for use in primary care. [Google Scholar]
  2. Bergman A. H., Bergman H., Palmstierna T., Schlyter F. Stockholm, Sweden: Karolinska Institute; 2003. DUDIT: The Drug Use Disorders Identification Test: MANUAL. [Google Scholar]
  3. Black N., Mullan B., Sharpe L. Computer-delivered interventions for reducing alcohol consumption: Meta-analysis and meta-regression using behaviour change techniques and theory. Health Psychology Review. 2016;10:341–357. doi: 10.1080/17437199.2016.1168268. doi:10.1080/17437199.2016.1168268. [DOI] [PubMed] [Google Scholar]
  4. Carey K. B., Scott-Sheldon L. A. J., Elliott J. C., Garey L., Carey M. P. Face-to-face versus computer-delivered alcohol interventions for college drinkers: A meta-analytic review, 1998 to 2010. Clinical Psychology Review. 2012;32:690–703. doi: 10.1016/j.cpr.2012.08.001. doi:10.1016/j.cpr.2012.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cole H. A., Prassel H. B., Carlson C. R. A meta-analysis of computer-delivered drinking interventions for college students: A comprehensive review of studies from 2010 to 2016. Journal of Studies on Alcohol and Drugs. 2018;79:686–696. doi:10.15288/jsad.2018.79.686. [PubMed] [Google Scholar]
  6. Cooper R. A., Weekes A. J. Totowa, NJ: Barnes and Noble Books; 1983. Data, models and statistical analysis. [Google Scholar]
  7. Doumas D. M., Esp S., Turrisi R., Bond L., Flay B. Efficacy of the eCHECKUP TO GO for high school seniors: Sex differences in risk factors, protective behavioral strategies, and alcohol use. Journal of Studies on Alcohol and Drugs. 2020;81:135–143. doi: 10.15288/jsad.2020.81.135. doi:10.15288/jsad.2020.81.135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Henrich J., Heine S. J., Norenzayan A. The weirdest people in the world? Behavioral and Brain Sciences. 2010;33:61–83. doi: 10.1017/S0140525X0999152X. discussion 83–135. doi:10.1017/S0140525X0999152X. [DOI] [PubMed] [Google Scholar]
  9. Huh D., Mun E.-Y., Larimer M. E., White H. R., Ray A. E., Rhew I. C., Atkins D. C. Brief motivational interventions for college student drinking may not be as powerful as we think: An individual participant-level data meta-analysis. Alcoholism: Clinical and Experimental Research. 2015;39:919–931. doi: 10.1111/acer.12714. doi:10.1111/acer.12714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kroenke K., Spitzer R. L., Williams J. B. The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine. 2001;16:606–613. doi: 10.1046/j.1525-1497.2001.016009606.x. doi:10.1046/j.1525-1497.2001.016009606.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Miller M. B., Leffingwell T., Claborn K., Meier E., Walters S., Neighbors C. Personalized feedback interventions for college alcohol misuse: An update of Walters & Neighbors. Psychology of Addictive Behaviors. 2013;2005;27:909–920. doi: 10.1037/a0031174. doi:10.1037/a0031174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Miller W. R., Rollnick S. (3rd Ed.). New York, NY: Guilford Press; 2013. Motivational interviewing: Helping people change. [Google Scholar]
  13. National Institute on Alcohol Abuse and Alcoholism. Recommended alcohol questions. 2003. Retrieved from https://web.archive.org/web/20210709043804/https://www.niaaa.nih.gov/research/guidelines-and-resources/recommended-alcohol-questions.
  14. Nielsen M., Haun D., Kärtner J., Legare C. H. The persistent sampling bias in developmental psychology: A call to action. Journal of Experimental Child Psychology. 2017;162:31–38. doi: 10.1016/j.jecp.2017.04.017. doi:10.1016/j.jecp.2017.04.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. R Core Team. Vienna, Austria: R Foundation for Statistical Computing; 2020. R: A language and environment for statistical computing. Retrieved from https://www.R-project.org/ [Google Scholar]
  16. Ray A. E., Kim S. Y., White H. R., Larimer M. E., Mun E.-Y., Clarke N., Huh D. the Project INTEGRATE Team. When less is more and more is less in brief motivational interventions: Characteristics of intervention content and their associations with drinking outcomes. Psychology of Addictive Behaviors. 2014;28:1026–1040. doi: 10.1037/a0036593. doi:10.1037/a0036593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Revelle W. Evanston, IL: Northwestern University; 2021. psych: Procedures for Psychological, Psychometric, and Personality Research. R package version 2.1.3. Retrieved from https://CRAN.R-project.org/package=psych. [Google Scholar]
  18. Saunders J. B., Aasland O. G., Babor T. F., de la Fuente J. R., Grant M. Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption II. Addiction. 1993;88:791–804. doi: 10.1111/j.1360-0443.1993.tb02093.x. doi:10.1111/j.1360-0443.1993.tb02093.x. [DOI] [PubMed] [Google Scholar]
  19. Sher K. J., Rutledge P. C. Heavy drinking across the transition to college: Predicting first-semester heavy drinking from pre-college variables. Addictive Behaviors. 2007;32:819–835. doi: 10.1016/j.addbeh.2006.06.024. doi:10.1016/j.addbeh.2006.06.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Shorter G. W., Bray J. W., Heather N., Berman A. H., Giles E. L., Clarke M., Newbury-Birch D. The “Outcome Reporting in Brief Intervention Trials: Alcohol” (ORBITAL) core outcome set: International consensus on outcomes to measure in efficacy and effectiveness trials of alcohol brief interventions. Journal of Studies on Alcohol and Drugs. 2021;82:638–646. doi: 10.15288/jsad.2021.82.638. doi:10.15288/jsad.2021.82.638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Spitzer R. L., Kroenke K., Williams J. B., Löwe B. A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine. 2006;166:1092–1097. doi: 10.1001/archinte.166.10.1092. doi:10.1001/archinte.166.10.1092. [DOI] [PubMed] [Google Scholar]
  22. Strohmetz D. B., Alterman A. I., Walter D. Subject selection bias in alcoholics volunteering for a treatment study. Alcoholism: Clinical and Experimental Research. 1990;14:736–738. doi: 10.1111/j.1530-0277.1990.tb01236.x. doi:10.1111/j.1530-0277.1990.tb01236.x. [DOI] [PubMed] [Google Scholar]
  23. Tanner-Smith E. E., Lipsey M. W. Brief alcohol interventions for adolescents and young adults: A systematic review and meta-analysis. Journal of Substance Abuse Treatment. 2015;51:1–18. doi: 10.1016/j.jsat.2014.09.001. doi:10.1016/j.jsat.2014.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Studies on Alcohol and Drugs are provided here courtesy of Rutgers University. Center of Alcohol Studies

RESOURCES