Past exposure to neuroleptic drugs and risk of Parkinson disease in an elderly cohort.

Neuroleptics and neuroleptic-like drugs are known to induce parkinsonism, which may reveal underlying Parkinson disease (PD) in some cases. We assessed the long-term risk of developing PD after past exposure to these drugs, in a 15-year prospective population-based elderly cohort study. We used the Cox proportional hazards model to assess the relation between past exposure to neuroleptics and the risk of developing incident PD. All incident cases of parkinsonism were identified by standardized procedure and validated by a committee of experts. Of 2,991 subjects followed, 117 developed parkinsonism and 43 developed probable PD during follow-up, of whom 22.2% and 32.6%, respectively, had been exposed to neuroleptics, compared to 16.6% for subjects without parkinsonism. About a third of subjects presented transient parkinsonism during drug exposure. After adjustment for gender and past occupation, past exposure to neuroleptics was associated with incident PD (relative risk, 3.16; 95% confidence interval [CI], 1.65-6.04). The relative risk was 3.65 (95% CI, 1.41-9.45) for benzamides and 2.59 (95% CI, 1.23-5.43) for phenothiazines. The population-attributable fraction of the risk for developing PD was 8.2% for benzamides and 12.2% for phenothiazines. In a French elderly cohort, the risk of probable PD was increased by 3.2-fold after exposure to neuroleptics. This finding suggests the necessity of limiting the use of such drugs in elderly people.