Reid Allecia E, Carey Kate B, Merrill Jennifer E, Carey Michael P
Center for Alcohol and Addiction Studies, Brown University.
Department of Behavioral and Social Sciences, Program in Public Health, Brown University.
J Consult Clin Psychol. 2015 Feb;83(1):36-44. doi: 10.1037/a0037634. Epub 2014 Aug 11.
To determine whether (a) social networks influence the extent to which college students initiate and/or maintain reductions in drinking following an alcohol intervention and (b) students with riskier networks respond better to a counselor-delivered, vs. a computer-delivered, intervention.
Mandated students (N = 316; 63% male) provided their perceptions of peer network members' drinking statuses (e.g., heavy drinker) and how accepting each friend would be if the participant reduced his or her drinking. Next, they were randomized to receive a brief motivational intervention (BMI) or Alcohol Edu for Sanctions (EDU). In latent growth models controlling for baseline levels on outcomes, influences of social networks on 2 phases of intervention response were examined: initiation of reductions in drinks per heaviest week, peak blood alcohol content (BAC), and consequences at 1 month (model intercepts) and maintenance of reductions between 1 and 12 months (model slopes).
Peer drinking status predicted initiation of reductions in drinks per heaviest week and peak BAC; peer acceptability predicted initial reductions in consequences. Peer Acceptability × Condition interactions were significant or marginal for all outcomes in the maintenance phase. In networks with higher perceived acceptability of decreasing use, BMI and EDU exhibited similar growth rates. In less accepting networks, growth rates were significantly steeper among EDU than BMI participants. For consumption outcomes, lower perceived peer acceptability predicted steeper rates of growth in drinking among EDU but not BMI participants.
Understanding how social networks influence behavior change and how interventions mitigate their influence is important for optimizing efficacy of alcohol interventions.
确定（a）社交网络是否会影响大学生在酒精干预后开始和/或维持饮酒量减少的程度，以及（b）社交网络风险较高的学生对咨询师提供的干预与计算机提供的干预，哪种反应更好。
被要求参与的学生（N = 316；63%为男性）提供了他们对同伴网络成员饮酒状况（例如，酗酒者）的看法，以及如果参与者减少饮酒，每个朋友会有多接受。接下来，他们被随机分配接受简短的动机干预（BMI）或制裁酒精教育（EDU）。在控制结果基线水平的潜在增长模型中，研究了社交网络对干预反应两个阶段的影响：在饮酒量最高的一周开始减少饮酒量、最高血液酒精含量（BAC）以及1个月时的后果（模型截距），以及在1至12个月之间维持饮酒量减少（模型斜率）。
同伴饮酒状况预测了饮酒量最高的一周开始减少饮酒量以及最高BAC；同伴接受度预测了后果的初步减少。在维持阶段，同伴接受度×条件交互作用对所有结果均显著或接近显著。在对减少饮酒的接受度较高的社交网络中，BMI和EDU表现出相似的增长率。在接受度较低的社交网络中，EDU参与者的增长率明显高于BMI参与者。对于饮酒量结果，较低的同伴接受度预测了EDU参与者但不是BMI参与者饮酒量增长更快。
了解社交网络如何影响行为改变以及干预如何减轻其影响对于优化酒精干预的效果很重要。
