Threat-related adverse childhood experiences and loneliness among middle-aged and older adults in China: indirect associations involving social activity participation breadth and functional limitations

BackgroundAdverse childhood experiences (ACEs) are heterogeneous, and threat-related adversities involving actual or threatened harm may be particularly relevant to later social and emotional functioning. However, evidence regarding the association between threat-related ACEs and loneliness in later life, and the potential roles of social activity participation breadth and functional limitations, remains limited among middle-aged and older Chinese adults. This study examined these associations using data from the China Health and Retirement Longitudinal Study.MethodsData were drawn from the 2014 and 2020 China Health and Retirement Longitudinal Study (CHARLS). Threat-related ACEs—physical abuse, household substance abuse, domestic violence, unsafe neighborhood, and bullying—were retrospectively reported in 2014 and summed into a score ranging from 0 to 5. Social activity participation breadth, basic and instrumental activities of daily living (BADL/IADL) limitations, and loneliness were assessed concurrently in 2020. Although data from two CHARLS time points were linked, the specified indirect pathway was not longitudinal. Regression-based serial indirect-association models were estimated in Stata using product-of-coefficients calculations. Standard errors and 95% confidence intervals for the specific and total indirect associations were obtained from 10,000 nonparametric bootstrap replications. All estimates were interpreted as associational statistical quantities rather than causal effects.ResultsHigher threat-related ACE scores were associated with higher loneliness scores. In both the IADL- and BADL-based serial indirect-association models, the largest indirect association was observed through functional limitations. The indirect associations through social activity participation breadth alone and through the specified sequential pathway were statistically distinguishable from zero but small in magnitude. The negative coefficients reflect the algebraic decomposition of fitted associations and do not indicate that greater social activity participation breadth reduces loneliness or that threat-related ACEs have a protective effect.ConclusionThreat-related ACE scores were associated with loneliness among middle-aged and older Chinese adults. The statistical decompositions identified small indirect associations involving social activity participation breadth and functional limitations, with the largest specific indirect estimate involving functional limitations. These findings apply specifically to the five-item threat-related ACE construct used in this study and should not be generalized to deprivation-related experiences or the full range of ACEs. They do not demonstrate mediation as a temporal or causal process because the proposed intermediate variables and loneliness were measured concurrently.