Understanding the Transition from Stress to Depression: A Longitudinal Mediational Analysis of Anxiety in Adults from the Metropolitan Region of Chile

Marcos E. Gomez^*Gustavo MarcosWenceslao Unanue

• Business School, Universidad Adolfo Ibáñez, Santiago, Chile

Drawing on recent advances in Beck's cognitive model — which traditionally conceptualizes anxiety and depression as correlated but does not explicitly address their temporal ordering — this study tests whether anxiety operates as a sequential mediator linking sustained stress to depressive symptoms in a non-clinical adult population. Prior longitudinal and mediation studies have examined associations among stress, anxiety, and depression, but differences in design, population, and analytical focus limit their applicability to non-clinical adult contexts. We extend this literature with data obtained from 805 adults in the Metropolitan Region of Santiago, Chile, followed across three waves at two-month intervals during the COVID-19 pandemic, although no pre-pandemic baseline was available. Accordingly, findings should be interpreted as evidence of the stress–anxiety–depression sequence within pandemic conditions. A cross-lagged panel model (CLPM) was used to estimate the temporal relations among stress, anxiety, and depression. This statistical method accounts for the stability of each construct across repeated measurements while estimating directional relationships over time. Results confirmed a significant partial mediation: perceived stress at T1 predicted higher anxiety at T2, which in turn predicted increased depressive symptoms at T3 (standardized indirect effect β = 0.049, 95% CI [0.016, 0.091], p = .009). To the best of our knowledge, this is the first longitudinal study conducted with a non-clinical adult sample from the Metropolitan Region of Chile that validates this mediation sequence. The findings advance Beck’s model by demonstrating anxiety’s role as a sequential mediator, contribute methodologically through the use of a three-wave CLPM to test temporal precedence, and support preventive interventions targeting early detection of subclinical anxiety to disrupt trajectories toward depression. Together, the results update the cognitive model and inform both clinical practice and public health strategies in emotionally demanding contexts.