Aims/hypothesis The aim of the study was to determine whether diabetic peripheral neuropathy (DPN) is a risk factor for depressive symptoms and examine the potential mechanisms for this relationship. Methods This longitudinal study (9 and 18 month followup) of 338 DPN patients (mean age 61 years; 71% male; 73% type 2 diabetes) examined the temporal relationships between DPN severity (mean±SD; neuropathy disability score [NDS], 7.4±2.2; mean vibration perception threshold, 41.5±9.5 V), DPN somatic experiences (symptoms and foot ulceration), DPN psychosocial consequences (restrictions in activities of daily living [ADL] and social selfperception) and the Hospital Anxiety and Depression subscale measuring depressive symptoms (HADS-D; mean 4.9±3.7).Results Controlling for baseline HADS-D and demographic/disease variables, NDS at baseline significantly predicted increased HADS-D over 18 months. This association was mediated by baseline unsteadiness, which was significantly associated with increased HADS-D. Baseline ADL restrictions significantly predicted increased HADS-D and partly mediated the association between baseline unsteadiness and change in HADS-D. Increased pain, unsteadiness and ADL restrictions from baseline to 9 months each significantly predicted increased HADS-D over 18 months. Change in social self-perception from baseline to 9 months significantly predicted increased HADS-D and partly mediated the relationships of change in unsteadiness and ADL restrictions with change in HADS-D. Conclusions/interpretation These results confirm that neuropathy is a risk factor for depressive symptoms because it