A major concern regarding ventralis intermedius nucleus deep brain stimulation for essential tremor has been the loss of surgical efficacy over time in a minority of patients. Some experts have ascribed the worsening tremor to tolerance, while other evidence has suggested that disease progression may play a role. Suboptimal lead placement has also been reported to be a factor in worsening tremor following deep-brain stimulation; however, most authors consider this phenomenon to manifest within a few months of the actual surgery. We aimed to dissect the tolerance versus disease progression issue by analysing preoperative versus long-term post-surgical Fahn-Tolosa-Marin Tremor Rating Scale scores both on and off stimulation among 28 patients who underwent ventralis intermedius nucleus deep brain stimulation and 21 age-matched controls. Of the 28 patients in the treatment arm of the cohort, seven (25%) demonstrated evidence of tremor progression, and had a 34% increase in the tremor score off stimulation at the 36 month follow-up compared with a 32% increase among controls (P = 0.67). In one of the seven patients there was evidence of suboptimal lead placement given the lateral position of the lead, and the motor side effects during threshold testing. This patient demonstrated a loss of stimulation benefit between 24 and 36 months, which may have been more indicative of tolerance. The other six subjects (86%) maintained stimulation benefit throughout the follow-up period, despite worsening tremor off stimulation (at a comparable rate to that of controls), making disease progression the most likely explanation. The data suggest that deep brain stimulation tolerance may be over-reported in the literature, and that a tolerance versus disease progression work-up should include: examining the trend in off stimulation scores, accounting for image based lead locations, and during programming sessions checking for thresholds which may elicit clinical benefits and side effects.