Predictors of loneliness among middle childhood and adolescence during the COVID-19 pandemic

• Published in: PloS one 2024-08, Vol.19 (8), p.e0308091
• Main Authors: Benhayoun, Ashley, Olsavsky, Anna, Akard, Terrah Foster, Gerhardt, Cynthia, Skeens, Micah A
• Format: Article
• Language: English
• Subjects: Adolescence; Adolescent; Adolescents; Child; Childhood; Children; Comparative analysis; Context; COVID-19; COVID-19 - epidemiology; COVID-19 - psychology; Developmental stages; Epidemics; Family Relations - psychology; Female; Humans; Impact analysis; Interaction models; Loneliness; Loneliness - psychology; Longitudinal Studies; Male; Mediation; Medicine and Health Sciences; Pandemics; Parents; People and Places; Regression analysis; Regression models; Research and Analysis Methods; SARS-CoV-2; Social Sciences; Surveys; Surveys and Questionnaires; Teenagers; Time measurement
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0308091

Social contexts (e.g., family, friends) are important in predicting and preventing loneliness in middle childhood (MC) and adolescence; however, these social contexts were disrupted during the COVID-19 pandemic. Comparison of social context factors that may differentially contribute to loneliness at each developmental stage (MC vs. adolescence) during the COVID-19 pandemic have been overlooked. This study examined longitudinal predictors of loneliness, including social contexts and COVID-19 impact, within MC (8-12y) and adolescence (13-17y). Parents reported on demographic information, and their children completed surveys on COVID-19 impact, loneliness, and family functioning using the COVID-19 Exposure and Family Impact Survey (CEFIS), the NIH Toolbox Loneliness (Ages 8-17) measure, and the PROMIS Family Relationships Short Form 4a measure, respectively. Regression models examined time one (T1; May-June 2020) predictors of time two (T2; November 2020-January 2021) MC child (n=92, Mage=10.03) and adolescent (n=56, Mage=14.66) loneliness. For the MC child model, significant predictors of higher loneliness included worse family functioning as well as higher COVID-19 impact and lower family income. On the other hand, higher adolescent loneliness was significantly predicted by not having married/partnered parents and was marginally significantly predicted by higher COVID-19 impact. The regression model with the full sample and interaction terms revealed no significant interactions, but that lower family functioning and higher COVID-19 impact were significant predictors of higher loneliness. Lower family income and lower in-person communication were marginally significant predictors of higher loneliness in the combined interaction model. Lastly, further exploratory mediation analyses displayed that family functioning significantly mediated the relationship between COVID-19 impact and T2 loneliness only for MC children and the full sample. Results support future interventions focused on optimizing family functioning to help mitigate MC loneliness in the context of adversity, such as a global pandemic.