Peter Boris Centre for Addiction Research, St Joseph's Healthcare Hamilton.
Department of Psychological Science, Vassar College.
Psychol Addict Behav. 2024 Sep;38(6):656-667. doi: 10.1037/adb0001006. Epub 2024 Apr 18.
Social network analysis (SNA) characterizes the structure and composition of a person's social relationships. Network features have been associated with alcohol consumption in observational studies, primarily of university undergraduates. No studies have investigated whether indicators from a person's social network can accurately identify the presence of alcohol use disorder (AUD), offering an indirect strategy for identifying AUD.
Two cross-sectional case-control designs examined the clinical utility of social network indicators for identifying individuals with AUD (cases) versus demographically matched drinkers without AUD (controls). Study 1 ( = 174) used high-resolution egocentric SNA assessment, whereas Study 2 ( = 189) used a brief assessment.
In Study 1, significant differences between AUD+ participants and controls were present for network alcohol severity (i.e., heavy drinking days; d = 1.23) and frequency ( = 0.35), but not network structural features. Network alcohol severity exhibited very good classification of AUD+ individuals versus controls (area under the curve [AUC] = 0.80), whereas network frequency did not (AUC = 0.61). In Study 2, significant differences were present for network alcohol severity ( = 1.02), quantity ( = 0.74), and frequency ( = 0.43), and severity exhibited good differentiation (AUC = 0.76).
Social network indicators of alcohol involvement robustly differentiated AUD+ individuals from matched controls, and the brief assessment performed almost as well as the high-resolution assessment. These findings provide proof-of-concept for severity-related SNA indicators as promising novel clinical assessments for AUD. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
社会网络分析（SNA）描述了一个人的社会关系结构和组成。网络特征已在观察性研究中与饮酒行为相关联，这些研究主要针对的是大学生。目前还没有研究调查一个人的社交网络指标是否可以准确识别出酒精使用障碍（AUD）的存在，这提供了一种间接识别 AUD 的策略。
两项横断面病例对照设计研究了社交网络指标在识别患有 AUD（病例）的个体与无 AUD 的同年龄、性别匹配的饮酒者（对照）方面的临床效用。研究 1（n = 174）使用高分辨率的自我中心 SNA 评估，而研究 2（n = 189）使用简短的评估。
在研究 1 中，AUD+参与者与对照组之间在网络酒精严重程度（即大量饮酒天数；d = 1.23）和频率（ = 0.35）上存在显著差异，但网络结构特征没有差异。网络酒精严重程度对 AUD+个体与对照组的分类非常好（曲线下面积 [AUC] = 0.80），而网络频率则不然（AUC = 0.61）。在研究 2 中，网络酒精严重程度（ = 1.02）、数量（ = 0.74）和频率（ = 0.43）均存在显著差异，且严重程度表现出良好的区分度（AUC = 0.76）。
酒精卷入的社交网络指标可以可靠地区分 AUD+个体与匹配的对照组，并且简短评估的表现几乎与高分辨率评估一样好。这些发现为基于严重程度的 SNA 指标作为 AUD 的有前途的新型临床评估提供了概念验证。
