Results of a multi-perspective examination of loneliness trajectories and its determinants in German adults during the COVID-19 pandemic.

Most research on adults' vulnerability to loneliness during the pandemic has been of limited quality. This study aimed to overcome previous limitations by examining loneliness trajectories in German adults from a population-based cohort during the pandemic using face-to-face assessment, identifying risk factors and highlighting those particularly relevant to older adults. Analyses included two measurement points before and two during the pandemic, combining data from the population-based Gutenberg Health Study and COVID-19 Study (N = 7001; baseline: Mage = 51.72, SDage = 10.04). Growth mixture models identified distinct loneliness trajectories. Factors associated with these trajectories were tested by a multinomial logistic regression model including sociodemographic, individual, and pandemic-related predictors and interactions with age. Overall, mean loneliness increased. Three distinct classes were identified: No Loneliness (59.3%), Onset (23.3%), and Temporary Increase (17.4%). In comparison to No Loneliness, Onset was associated with reduction in social contact during the pandemic and Temporary Increase with sex, high school degree, pre-pandemic depression symptoms, pandemic-related stressors and social support. No unique risk factors for older adults were found. Interventions that strengthen one's adaptability to (acute) stressors and promote social resources with special attention to women may be a promising way to prevent loneliness during the pandemic.