BACKGROUND: Tardive dyskinesia (TD) is a disabling movement disorder associated with the prolonged use of neuroleptic medication. Several strategies have been examined in the treatment of TD. Currently, however, there is no clear evidence of the effectiveness of these drugs in TD and they have been associated with many side effects. One particular strategy would be to use pharmaceutical agents which are known to influence the catecholaminergic system at various junctures. OBJECTIVES: To determine whether catecholaminergic drugs for people with schizophrenia or other chronic mental illnesses are associated with a reduction in neuroleptic-induced tardive dyskinesia. SEARCH STRATEGY: We searched the Cochrane Schizophrenia Group's Register (January 1996), Biological Abstracts (1982-1995), EMBASE (1980-1995), LILACS (1982-1996), MEDLINE (1966-1995) and PsycLIT (1974-1995). We searched the Cochrane Schizophrenia Group's Register again in December 2002 and September 2005. We also searched references of all relevant studies for further trial citations and contacted principal authors of trials. SELECTION CRITERIA: We selected studies if they were randomised controlled trials focusing on people with schizophrenia or other chronic mental illnesses who also suffered from neuroleptic-induced tardive dyskinesia. We compared the use of catecholaminergic interventions versus placebo or no intervention. DATA COLLECTION AND ANALYSIS: We independently extracted data. For homogenous dichotomous data, we calculated the random effects, relative risk (RR), and 95% confidence interval (CI) and, where appropriate, the numbers needed to treat (NNT) on an intention-to-treat basis. For continuous data, we calculated weighted mean differences (WMD). MAIN RESULTS: We excluded 20 studies, mainly due to an inability to extract data from the first arm of the study crossover. One included study has shown that patients on placebo were no more likely to leave the study early than those on tiapride (n=24). The other included study (n=35) also reported equivocal data (RR 5.28 CI 0.3 to 102.6) for leaving the study early when participants were randomised to either celiprolol or placebo. However, in both studies, sample size was limited. AUTHORS' CONCLUSIONS: Although there has been a large amount of research in this area, most studies were excluded due to inherent problems in the nature of their crossover designs. Usually data are not reported before the crossover and the nature of TD and its likely response to treatments makes it imprudent to use this data. The review provides little usable information for service users or providers and more well designed and reported studies are indicated.