BACKGROUND Long-term levodopa therapy for Parkinson's disease is complicated by the development of motor fluctuations and abnormal involuntary movements. One approach is to add a dopamine agonist at this stage of the disease to reduce the time the patient spends immobile or off and to reduce the dose of levodopa in the hope of reducing such problems in the future. OBJECTIVE To compare the efficacy and safety of adjuvant ropinirole therapy versus placebo in patients with Parkinson's disease already established on levodopa therapy and suffering from motor complications. METHODS Electronic searches of MEDLINE, EMBASE and the Cochrane Controlled Trials Register. Handsearching of the neurology literature as part of the Cochrane Movement Disorders Group's strategy. Examination of the reference lists of identified studies and other reviews. Contact with SmithKline Beecham. METHODS Randomised controlled trials of ropinirole versus placebo in patients with a clinical diagnosis of idiopathic Parkinson's disease and long-term complications of levodopa therapy. METHODS Data was abstracted independently by the authors and differences settled by discussion. The outcome measures used included Parkinson's disease rating scales, levodopa dosage, 'off' time measurements and the frequency of withdrawals and adverse events. RESULTS Three double-blind, parallel group, randomised, controlled trials have been conducted on 263 patients. The two phase II studies were relatively small, were conducted over the short term (12 weeks), and used relatively low doses of ropinirole (mean administered doses 3.3 and 3.5 mg/d) in a twice daily regime. In view of this clinical heterogeneity and some statistical heterogeneity, the results of these trials have not been included in a meta-analysis. The conclusions of this review are based on the evidence from a single phase III study which was medium term (26 weeks) and used ropinirole doses in line with the current UK licensed maximum in a thrice daily regime. In view of difficulties in assessing changes in off time in ~~ Leiberman 98~~, caused by the initial imbalance between the arms of the trial, it is unsafe to draw any firm conclusion about the effect of ropinirole on off time. However, as an adverse event, dyskinesia was significantly increased in those who received ropinirole (~~ Leiberman 98~~; odds ratio 2.90; 1.36, 6.19 95% CI; Table 8). Measurements of motor impairments and disability were poor in this study with incomplete information available. Levodopa dose could be reduced in ~~ Leiberman 98~~ with a significantly larger reduction on ropinirole than on placebo (weighted mean difference 180 mg/d; 106, 253 95% CI; Table 2). No significant differences in the frequency of adverse event reports were noted between ropinirole and placebo apart from the increase in dyskinesia with ropinirole. There was a trend towards fewer withdrawals from ropinirole in ~~ Leiberman 98~~ but this did not reach statistical significance. CONCLUSIONS Ropinirole therapy can reduce levodopa dose but at the expense of increased dyskinetic adverse events. No clear effect on off time reduction was found but this may have been due to the under-powering of the single evaluable trial. Inadequate data on motor impairments and disability was collected to assess these outcomes. These conclusions apply to short and medium term treatment, up to 26 weeks. Further longer term trials are required, with measurements of effectiveness, and also studies to compare the newer with the older dopamine agonists.