BACKGROUND: Evidence from clinical trials with monoamine oxidase type B inhibitors (TEMPO, ADAGIO and DATATOP) and levodopa (ELLDOPA) suggests that Parkinson's disease patients may benefit from treatment being commenced immediately on diagnosis rather than waiting for functional disability to develop, as is traditional clinical practice. METHODS: We performed a narrative literature review and meta-analysis of delayed-start design trials in Parkinson's disease. RESULTS: There was inconsistency in the results of the two rasagiline delayed-start design trials, with early treatment with a 2 mg dose significantly superior in the TEMPO trial, but the 1 mg dose significantly better in the ADAGIO trial, making interpretation difficult. Further, the benefits of immediate treatment were small in terms of total unified Parkinson's disease rating scale scores, with a mean difference of 0.91 units (95% confidence interval 0.01, 1.80; P=0.05) in a meta-analysis of the TEMPO and ADAGIO delayed-start design trials. Such small differences are unlikely to be of clinical relevance. There is also little information on whether immediate treatment has a beneficial effect on patient quality of life with an acceptable adverse reaction profile, and we have no data on whether imediate treatment is cost-effective. DISCUSSION: Based on the evidence available, changing clinical practice to immediate therapy on diagnosis is not warranted and further trials are needed. © 2011 Movement Disorder Society.