The robust reciprocal relationship between loneliness and depressive symptoms among the general population: Evidence from a quantitative analysis of 37 studies

Abstract

Background: Loneliness has long been associated with depressive symptoms, but there is no evidence on the question of “which comes first”. To help us to answer this question, this study conducted a meta-analysis using a cross-lagged formula to examine the reciprocal relations between loneliness and depressive symptoms, as well as potential moderators. Methods: The meta-analytic dataset consisted of 37 studies with sample sizes ranging from 52 to 5991. Effect sizes (including autoregressive effects and lagged coefficients) were included. We computed weighted mean effect sizes for the correlation effect and cross-lagged effect between loneliness and depressive symptoms. We also performed a meta-regression to investigate potential moderators. Results: The results showed that loneliness and depressive symptoms reciprocally predicted each other over time with similar effect sizes. There was no moderation of our overall effect size in either direction by time interval between measurements, the proportion of female participants, mean age of the sample, or type of measurement. Limitations: Caution should be taken in drawing final conclusions about the relative strength of reciprocal effects between loneliness and depressive symptoms due to the potential influence of varied research focus in the included studies, which could affect the effect size. Conclusions: These findings advance the study of the relationship between loneliness and depressive symptoms by providing evidence that the link between them is symmetrically reciprocal and robust, which may help clinicians to develop effective intervention methods.