Reducing substance involvement in college students: A three-arm parallel-group randomized controlled trial of a computer-based intervention

Highlights

• Computer ASSIST plus brief intervention reduced substance involvement score.

• Computer-based intervention reduces ASSIST scores for alcohol.

• Computer-based intervention for substances is effective in college students.

• Computer-based ASSIST plus brief intervention is comparable to interview.

Abstract

The prevalence of alcohol and other drug use is high among college students. Reducing their consumption will likely be beneficial for society as a whole. Computer and web-based interventions are promising for providing behaviorally based information. The present study compared the efficacy of three interventions (computerized screening and motivational intervention [ASSIST/MBIc], non-computerized screening and motivational intervention [ASSIST/MBIi], and screening only [control]) in college students in Curitiba, Brazil. A convenience sample of 458 students scored moderate and high risk on the ASSIST. They were then randomized into the three arms of the randomized controlled trial (ASSIST/MBIc, ASSIST/MBIi [interview], and assessment-only [control]) and assessed at baseline and 3 months later. The ASSIST involvement scores decreased at follow-up compared with baseline in the three groups, suggesting that any intervention is better than no intervention. For alcohol, the specific involvement scores decreased to a low level of risk in the three groups and the MBIc group showed a positive outcome compared with control, and the scores for each question were reduced in the two intervention groups compared to baseline. For tobacco, involvement scores decreased in the three groups, but they maintained moderate risk. For marijuana, a small positive effect was observed in the ASSIST/MBIi and control groups. The ASSIST/MBIc may be a good alternative to interview interventions because it is easy to administer, students frequently use such computer-based technologies, and individually tailored content can be delivered in the absence of a counselor.

Introduction

The progression from occasional to harmful/hazardous drug use is a worldwide health concern (United Nations Office on Drugs and Crime, 2011). Notably, prevalence of alcohol and other drug use is high among college students (Andrade, Duarte & Oliveira, 2010; Scott-Sheldon, Carey, Elliott, Garey, & Carey, 2014).

The early detection of substance involvement allows different levels of intervention and improves the prognosis of substance-related disorders (Ali et al., 2002, Johnston et al., 2012). The World Health Organization (WHO) developed the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) for all psychotropic substances. Individuals are scored for each type of drug used, with a ranking of their level of risk that is given as feedback. The ASSIST was validated as a structured interview, to be used by healthcare professionals in many countries, including Brazil (Henrique et al., 2004, Humeniuk et al., 2010). Barreto, Christoff, and Boerngen-Lacerda (2014) showed that adaptation of the ASSIST to a self-report version was as acceptable as the interview, and the two formats' scores were comparable in college students. This adaptation is a necessary step in the development of a computer-based format. Considering that college students seldom visit healthcare services, giving them more tailored approaches is recommended, such as computer- and web-based interventions.

Web-based programs are promising for college students because they have frequent access to the Internet (Gross, 2004). In Brazil, Internet access reached 43% of the total population in 2013. Seventy-seven percent of Brazilians aged 16–24 years are connected to the Internet. Although many computer- and web-based interventions are available for young people in other countries, adapting a simple, rapid, and valid intervention for Brazilian students is necessary when considering their lifestyle and cultural differences.

Recent reviews reported that different theoretically based interventions, delivered electronically, effectively reduced the frequency of alcohol (Carey et al., 2012, Fachini et al., 2012, Scott-Sheldon et al., 2014), tobacco (Hutton et al., 2011), and cannabis (Tait, Spijkerman, & Riper, 2013) use, mainly in college students. For other drugs, two reviews (Copeland and Martin, 2004, Moore et al., 2011) and two RCTs also confirmed its effectiveness (Campbell et al., 2014, Carroll et al., 2014) in general population.

The present RCT evaluated the efficacy of a computer-based intervention program, called ASSIST/Motivational Brief Intervention (ASSIST/MBIc) on substance involvement compared with those receiving only feedback about their ASSIST scores (control group) and others receiving feedback plus MBI in an interview (ASSIST/MBIi). The intervention was based on the traditional motivational interview (Prochaska, DiClemente, & Norcross, 1992), using some of the main elements: Feedback, Responsibility, Advice, Menu of Options, Empathy, and Self-Efficacy (FRAMES). The intervention sought to raise awareness of the risks of substance use, committing the students to use self-management skills to change their behavior (Miller & Rollnick, 2012). Both interventions were designed to be short and easily linked to the ASSIST results and the meaning of the scores, give advice about identifying potential problems, encourage behavioral change, allow the participants to report their substance-related problems, list the advantages and disadvantages of using the substance(s), list skills to cope with risky behaviors related to the substance(s), and provide goals to change behaviors in the short, medium, and long terms. The ASSIST plus brief intervention has already proven to be effective when individuals are interviewed face-to-face by health professionals in a very similar intervention model (Humeniuk et al., 2012).

We hypothesized that the ASSIST/MBIc and ASSIST/MBIi groups at follow-up would show less involvement with substances compared with the control group.