Objectives To explore the impact of depression on heart rate (HR) and heart rate variability (HRV) as a marker of autonomic nervous system (ANS) impairment in depressed and non-depressed patients with advanced type 2 diabetes mellitus (T2DM) and to explore possible effects of an acceptance- and mindfulness-based group intervention (MBSR) on HR and HRV.Methods Alongside a prospective clinical trial, we collected demographic, psychosocial and clinical data from 113 chronic T2DM patients in a standardized setting. At baseline and after one year, depressive mood was assessed with the Patient Health Questionnaire (PHQ-9), and autonomic function was determined by measuring HR and HRV markers. A subsample was randomly assigned to take part in eight MBSR sessions. Results Of the 113 T2DM patients (77.9% men; mean age=58.8±7.0 years; diabetes duration 11.5±7.0 years), 33 showed clinically relevant depressive symptoms at baseline. In cross-sectional analysis, we found no association between depression and HR/HRV (all comparisons p>0.05). In prospective regression analysis depression did not predict follow-up scores of HRV. The patients who participated in the MBSR intervention showed a tendency toward improved parasympathetic control (RMSSD, CV, E-I-Ratio) with small-to-moderate effect sizes (d≤0.38). Conclusions Depression was not directly associated with cardiac autonomic control in this sample, but MBSR training may have positively influenced HR and HRV. In advanced diabetes, somatic and behavioral parameters seem to be more predictive than depression for the course of autonomic functioning, but the pathways remain unclear.