BACKGROUND Mild cognitive deficits have been reported in essential tremor (ET). However, these cognitive deficits have been assessed in cross-sectional rather than longitudinal analyses. OBJECTIVE To determine whether decline in cognitive test scores occurs at a faster rate in ET cases than controls. METHODS In a population-based study of older people (≥ 65 years) in central Spain (Neurological Disorders in Central Spain, NEDICES), non-demented ET cases and controls were followed prospectively. Participants with baseline or incident Parkinson's disease or dementia were excluded as were participants who developed incident ET. At baseline (1994-1995) and at follow-up (1997-1998), a 37-item version of the mini-mental state examination (37-MMSE) was administered. RESULTS A total of 2319 participants (72.4 ± 5.8 years) included 135 prevalent ET cases and 2184 controls. At baseline, the mean 37-MMSE in cases was 28.8 ± 5.8 vs. 30.2 ± 4.8 in controls (P = 0.02). During the 3-year follow-up period, the 37-MMSE declined by 0.70 ± 3.2 points in cases vs. 0.11 ± 3.8 points in controls (P = 0.03). In analyses that adjusted for age, education, and other potential confounders, the case-control difference remained robust. DISCUSSION In this population-based, prospective study of non-demented elders, baseline cognitive test scores were lower in ET cases than controls; moreover, during the 3-year follow-up period, these scores declined at a rate that was seven-times faster in ET cases. This study provides evidence that cognitive deficits in ET are not static, and they appear to be progressing at a faster rate than in elders without this disease.