Soon after the introduction of levodopa in to the treatment of Parkinson's disease (PD), 1 marked involuntary movements or dyskinesia were identified as a common serious and potentially disabling side effect of treatment. 2 The drug developed a negative reputation for the high incidence of dyskinesia-although much of this was because of the high dosage used and the introduction of treatment in patients with advanced, but previously untreated, disease. 3 Later studies reported that early use of L-dopa resulted in a high prevalence of involuntary movements of up to 36% within 5 years and up to 88% within 10 years. 4,5 These reports and the clinical experience of early initiation of L-dopa therapy led to the widespread view that dyskinesia was an almost inevitable consequence of the use of L-dopa. Not surprisingly, the medical community and the patient population took fright and the early initiation of L-dopa therapy diminished. Avoiding L-dopa treatment until necessary became a strategy for treating the younger patient population and eventually led to the early use of dopamine agonist drugs that appeared to result in a lower incidence of dyskinesia. 3 However, more recently, L-dopa has made a return to frontline treatment using more careful treatment paradigms and using lower daily doses that seem to lessen the risk of dyskinesia induction. The PDMED study has also consolidated the role of initiating treatment of PD with L-dopa, and the current view is that there is little evidence to justify withholding the drug in early disease. 6 Nevertheless, recent reports would, at first glance, seem to suggest that dyskinesia has not gone away as a problem of clinical significance. In 2012, Manson and colleagues reviewed the incidence of dyskinesia in PD and concluded that 40% to 50% of patients develop dyskinesia within 5 years of initiation of treatment and that this increases to 50% to 75% after 10 years. 7 Also, Ahlskog and Muenter, in their 2001 review, showed that the prevalence of dyskinesia increases from a nil dyskinesia state in recently diagnosed patients to 60.7% to 95.5% in patients with a disease duration of 9 to 15 years after L-dopa treatment of variable dose schedules. 5 The data is in overall agreement with another review by Tran and colleagues, 8 which also pointed out that the prevalence of dyskinesia depended on the age of onset of PD, disease duration, and severity and duration of L-dopa therapy-as might be expected from past publications. A further pointer toward a high dyskinesia rate is also evident in the CALM-PD trial, where 222 patients were randomised to receive either pramipexole or L-dopa, and it was shown that 6 years after initiation of treatment, prevalence of dyskinesia was high, at 68%, for L-dopatreated patients and 50% for those on pramipexole. Disabling and painful dyskinesia, however, were almost nonexistent in both groups (2.8% and 3.5% reported disabling dyskinesia and 5.6% and 3.5% reported painful dyskinesia, respectively). 17 In a community-based study of 124 patients, Schrag an...