Factors that influence the variation in occurrence of antipsychotic-induced parkinsonism (AIP) in the elderly have not been well elucidated. The aim of this study was to investigate the association between parkinsonism in elderly users of haloperidol and prescribed dose, plasma concentration, and duration of use of haloperidol in a cross-sectional design. This study included 150 inpatients aged 65 years and older who were treated with haloperidol. Parkinsonism assessed by the Simpson Angus Scale was present in 46% of the included patients. Prescribed haloperidol dose varied from 0.3 to 5 mg/d. Plasma concentration ranged from 0.13 to 4.11 μg/L, with one outlying measurement (21.43 μg/L). Dose is moderate but significantly associated with haloperidol plasma concentration (weighted R = 0.32; P < 0.001). Variability in the total score on the Simpson Angus Scale could not be explained by the variability in dose, concentration (respectively R = 0.003 and 0.001) nor duration of use of haloperidol. Smoking showed to be not significantly protective in the development of AIP (crude odds ratio, 0.39; 95% confidence interval, 0.15-0.997; and adjusted odds ratio, 0.44; 95% confidence interval, 0.17-1.17). In a clinical practice-setting dose, neither plasma concentration nor duration of use of haloperidol is associated with an increased occurrence of AIP. This study does not support the hypothesis of the peripheral pharmacokinetic explanation for the high prevalence of AIP and differences in AIP sensitivity in the elderly during treatment with haloperidol.