Predictors of mortality in a large cohort of elders with essential tremor.

Abstract

Background: Prospective data show an increased risk of mortality in essential tremor (ET). Understanding the contributors to this increased mortality is of value.

Methods: 347 cases enrolled in a prospective study of cognition in elders with ET (M baseline age = 79.6 years, M follow-up length = 2.7 years). At each visit, cases completed clinical assessments, a cognitive test battery and were assigned a Clinical Dementia Rating (CDR). Cox proportional hazards regression equations identified predictors of cases' relative risk of mortality.

Results: 41 (11.8%) of 347 cases died during follow-up. Deceased cases were older at baseline, had fewer years of education, used more medication, had more severe tremor, had more gait impairment, and reported more falls, less physical activity, and more depressive symptoms than did living cases. In univariate analyses, a CDR score of 1 or 2 (dementia) was associated with a six-fold increased the risk of mortality (Hazards ratio, HR = 6.33). Moderate rather than milder levels of dementia (CDR 2 rather than 1) were driving this effect, with multivariate models demonstrating HRs in excess of 7.5 (p <0.001). Gait impairment and falls were also associated with significant increased risk of mortality in multivariate models.

Conclusion: We identified several factors that may be used to risk-stratify ET patients with respect to mortality. The information from these analyses can be of value in identifying subgroups of ET patients who demonstrate risk for early death; such subgroups can then be targeted for interventions as well as preventive measures.