Short-term challenges with dopaminergic agents are used in patients with idiopathic Parkinson's disease (PD) to predict the therapeutic effect of sustained levodopa treatment, but false-negative results often occur. We prospectively evaluated 74 patients with clinically diagnosed IPD and compared the predictive value of a short-term levodopa test assessed by movement time (MT) with the predictive value obtained by the evaluation with the motor examination part of the Unified Parkinson's Disease Rating Scale (UPDRS-ME). The response to long-term levodopa was accurately predicted in 96% of patients by assessing the response to the short-term test with MT and in 80% of cases with UPDRS-ME. Similar predictive values were obtained by separately analyzing 21 de novo patients. The short-term test also accurately predicted the magnitude of improvement with long-term treatment. We conclude that the predictive value for long-term dopaminergic responsiveness may be further enhanced by evaluating the short-term pharmacologic challenges with MT analysis. This is particularly useful to select de novo patients for drug trials with dopaminergic agents.