Roseli Boerngen-Lacerda, PhD, Universidade Federal do Paraná, Curitiba, Paraná, Brazil, presented the results of a study [Christoff AO, Boerngen-Lacerda R. Addict Behav. 2015] among college students in Brazil demonstrating that, while any intervention was better than none in reducing the prevalence of alcohol and drug use, a web-based intervention was particularly beneficial for reducing alcohol use.

Alcohol and drug use is prevalent among college students, and progression from their occasional to hazardous or harmful use has become a significant global public health issue. The World Health Organization’s Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) was developed as a clinical tool to detect psychoactive substance use. However, because of sporadic health care utilization by college students, Prof Boerngen-Lacerda and colleagues used a more tailored strategy, a web-based intervention, for this patient population that uses the Internet heavily.

The randomized controlled trial was conducted to compare the efficacy of 3 different interventions in reducing the prevalence of alcohol and drug use among college students in Brazil. A total of 815 students were invited to participate and were randomized to ASSIST intervention (n = 436) or ASSIST control (n = 373); 6 were excluded for fictitious drug use. Then, 458 patients who scored as having moderate- or high-risk use were randomized to a computer-based intervention called the ASSIST/Motivational Brief Intervention (ASSIST/MBIc; n = 144), in-person feedback plus an MBI in an interview (ASSIST/MBIi; n = 167), or feedback only (control; n = 147). Inclusion criteria included the ability and willingness to participate in 2 sessions, each lasting 5 minutes (control group) to 40 minutes (ASSIST/MBIc and ASSIST/MBIi groups), and to avoid engaging in other substance use treatments or programs before or during the study. Participants were evaluated at baseline and 90 days later.

The ASSIST scores were reduced in all 3 groups at the 90-day follow-up compared with baseline (Figure 1). This suggests that any intervention is better than none, said Prof Boerngen-Lacerda.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Effect of Intervention on Alcohol, Smoking, and Substance Involvement Screening Test Scores

Specific substance ASSIST scores at follow-up relative to baseline in college students. The data are expressed as mean ± standard error of the involvement scores in the three groups: ASSIST/MBIc, ASSIST/MBIi, and control. From the left to right the bars represent total involvement score, alcohol, tobacco, marijuana, and summation of the scores for other drugs. The symbol “*” represents a significant difference (P ≤ .05, ANOVA followed by Newman–Keuls test) compared with control.

Reprinted from Addict Behav, Vol. 45, Christoff AO et al. Reducing substance involvement in college students: A three-arm parallel-group randomized controlled trial of a computer-based intervention. Pages No. 164-171, Copyright (2015), with permission from Elsevier.

In addition, the computer-based intervention was particularly beneficial for reducing alcohol use. While the specific involvement scores for alcohol decreased to a low level of risk in all 3 groups, scores in the ASSIST/MBIc group were significantly reduced compared with the control group (P ≤ .05). The scores for each question for alcohol were reduced in the 2 intervention groups compared with baseline.

A small positive, although statistically insignificant, effect was also observed in the specific involvement scores for marijuana in the ASSIST/MBIi and control groups.

Although the specific involvement scores for tobacco decreased in all 3 groups, they continued to indicate moderate risk at 90-day follow-up, with no significant effect of either ASSIST/MBIc or ASSIST/MBIi intervention on scores compared with control.

The results of this study showed that the ASSIST/MBIc intervention may therefore be a better alternative to interview interventions for college students who are accustomed to using computer-based technologies, concluded Prof Boerngen-Lacerda. This intervention is easy to administer, and its content can be tailored to individual students and delivered in the absence of a counselor, she said.