Department of Psychology, Fordham University, New York, USA.
National Institutes of Health, National Institute on Drug Abuse, Baltimore, USA.
Alcohol Clin Exp Res. 2022 Aug;46(8):1539-1551. doi: 10.1111/acer.14889. Epub 2022 Aug 26.
Research conducted during the COVID-19 Pandemic has identified two co-occurring public health concerns: loneliness and substance use. Findings from research conducted prior to the pandemic are inconclusive as to the links between loneliness and substance use. This study aimed to measure associations of loneliness with three different types of substance use during COVID-19: daily number of alcoholic drinks, cannabis use, and non-cannabis drug use.
Data were obtained between October 2020 and May 2021 from 2,648 US adults (M = 38.76, 65.4% women) diverse with respect to race and ethnicity using online recruitment. Participants completed baseline surveys and daily assessments for 30 days. A daily loneliness measure was recoded into separate within- and between-person predictor variables. Daily outcome measures included the number of alcoholic drinks consumed and dichotomous cannabis and non-cannabis drug use variables. Generalized linear multilevel models (GLMLM) were used to examine within- and between-person associations between loneliness and substance use.
The unconditional means model indicated that 59.0% of the variance in the daily number of alcoholic drinks was due to within-person variability. GLMLM analyses revealed that, overall, people drank more on days when they felt a particularly high or particularly low degree of loneliness (positive quadratic effect). There was a negative and significant within-person association between daily loneliness and the likelihood of cannabis use. There was also a positive and significant within-person association between daily loneliness and the likelihood of non-cannabis drug use.
Associations between loneliness and substance use vary with substance type and whether within- or between-person differences are assessed. These findings are relevant to the persistence of substance use disorders and thus of potential clinical importance. Individuals who do not experience severe loneliness at intake but who show daily increases in loneliness above baseline levels are at heightened risk of alcohol and non-cannabis drug use. Future research could profitably examine just-in-time adaptive interventions that assess fluctuations in loneliness to prevent the development or exacerbation of substance use disorders.
在 COVID-19 大流行期间开展的研究发现了两个同时存在的公共卫生问题：孤独和物质使用。大流行前开展的研究结果对于孤独感和物质使用之间的联系尚无定论。本研究旨在衡量孤独感与 COVID-19 期间三种不同类型的物质使用之间的关联：每日饮酒量、大麻使用和非大麻药物使用。
本研究于 2020 年 10 月至 2021 年 5 月期间，使用在线招募方式从 2648 名美国成年人（M = 38.76，65.4%为女性）中获取数据，这些参与者在种族和民族方面具有多样性。参与者完成了基线调查和 30 天的每日评估。每日孤独感测量结果被重新编码为单独的个体内和个体间预测变量。每日结局测量包括饮酒量和二分法大麻和非大麻药物使用变量。使用广义线性多层模型（GLMLM）来检验孤独感与物质使用之间的个体内和个体间关联。
无条件均值模型表明，每日饮酒量的方差中有 59.0%归因于个体内变异性。GLMLM 分析显示，总体而言，当人们感到孤独感特别高或特别低时，他们的饮酒量会增加（正二次效应）。每日孤独感与大麻使用的可能性之间存在负向且显著的个体内关联。每日孤独感与非大麻药物使用的可能性之间也存在正向且显著的个体内关联。
孤独感与物质使用之间的关联因物质类型以及评估个体内差异还是个体间差异而有所不同。这些发现与物质使用障碍的持续存在有关，因此具有潜在的临床重要性。在摄入时不感到严重孤独感但在基线水平以上每日孤独感增加的个体，更有可能出现酒精和非大麻药物使用。未来的研究可以从评估孤独感波动的即时自适应干预措施中受益，以预防物质使用障碍的发展或恶化。
