Personalized Boosters after a Computerized Intervention Targeting College Student Drinking: Perceived Descriptive Norms As a Mechanism of Change

Introduction: Computerized interventions are widely used because of their logistic advantages over in-person interventions. They are more cost-effective and can quickly deliver tailored individual feedback to large groups of students. However, in-person interventions are generally more efficacious. A preliminary study demonstrated the ability of emailed boosters to improve the efficacy of a popular online intervention using personalized feedback delivered remotely in the short term. The current study is an examination of longer-term effects and mechanisms of change.

Methods: For the current study, data were collected from N = 561 college student volunteers ages 18 to 24, mostly female (67%), and mostly Caucasian/White (49%) or African-American/Black (37%). Participants were randomized into one of three conditions: 1) general health information session only, 2) alcohol intervention only, or 3) alcohol intervention plus booster. Participants engaged in 60 minutes with the online intervention, and were assessed on alcohol consumption and related problems at baseline (pre-intervention), as well as 2, 4, and 6 weeks post, and 3, 6, and 9 months post. Emailed boosters sent after the two-week assessment contained personalized feedback regarding tailored descriptive norms (i.e., information on students at the same institution by gender), and protective behavioral strategies (PBS).

Results: Dual process latent growth models revealed that booster emails with personalized feedback yielded significant reductions in perceived descriptive norms, b = -1.36 (β = -0.293), which were in turn associated with significant reductions in drinking, b = 0.09 (β = 0.263). This significant indirect effect, b = -0.12 (β = -0.077), supports perceived descriptive norms as a mechanism of change. Pairing norms (motivating information) with PBS (means of harm reduction) appears to facilitate drinking reductions through correcting normative misperceptions. These effects are maintained through the 9-month follow-up, with drinking displaying only minor fluctuations across time (b = .003, p = .843).

Discussion: The temporal relationship between changes in normative perceptions and alcohol consumption, use of experimental manipulations, and strong associations among variables support perceived descriptive norms as a true mechanism of change. Moreover, these findings are consistent with the literature regarding descriptive norms as a mechanism of change across interventions for college drinking. The observed efficacy of the personalized booster delivered via email has positive clinical implications. Its low cost and easy dissemination make it a very efficient way to potentially reduce high risk drinking among students long-term.