Methods
Consecutively admitted geropsychiatric inpatients were examined with the Abnormal Involuntary Movement Scale within 72 hours of admission; comprehensive demographic, diagnostic, and psychometric data were also obtained.

Results
Seventy-four (19.2%) of 386 patients received diagnoses of dyskinesia. Lifetime duration of neuroleptic use was strongly correlated with dyskinesia rates. After accounting for the effect of lifetime duration of neuroleptic use in a stepwise logistic regression, only associations with Global Assessment Scale score and presence of dental problems remained statistically significant. In comparison with the duration of neuroleptic use, however, the contribution of these factors was minor. Sixteen percent of patients with less than 3 months of neuroleptic use, 29% with 3 to 12 months of neuroleptic use, 30% with 1 to 10 years of neuroleptic use, and 41% with more than 10 years of neuroleptic use had dyskinesia. Compared with patients with no history of neuroleptic treatment, the relative risks for these durations of neuroleptic use were 1.62 (95% confidence limits [CL],0.81, 3.24), 2.89 (95% CL, 1.50, 5.55), 3.08 (95% CL, 1.66, 5.70), and 4.11 (95% CL, 2.12, 7.96), respectively.

Conclusions
Within elderly populations, duration of exposure to neuroleptics is the strongest predictor of risk for tardive dyskinesia, and this risk increases rapidly within the first year of total lifetime neuroleptic use.