The objective of this study was to determine the putative risk factors for the development of tardive dystonia (TDt) in contrast with tardive dyskinesia (TD). Fifteen TDt patients seen in the Movement Disorders Clinic were compared with 2 groups of 15 TD controls each. The first control group was drawn from the Clinic and matched with the TDt cases for severity, using degree of dysfunction as the matching variable. The second control group comprised mild TD cases drawn from a separate study of drug-induced movement disorders in chronic schizophrenia and were matched for age and sex with the TDt cases. A number of demographic, treatment-related, diagnosis-related and historical variables suggested in the literature were examined. Most risk factors for TDt that have been suggested by previous studies were not supported. The first control group was significantly older than the TDt cases. The TDt patients had a more frequent past history of acute drug-induced dystonia and of postural tremor prior to the onset of the mental illness, although only the former reached statistical significance. The results suggested that TDt and TD do not differ in most putative risk factors, although the small sample size increases the likelihood of a type II error. It is inconclusive on the role of young age and male sex as risk factors. TDt cases may, however, be individuals vulnerable to the development of dystonia, with neuroleptics probably bringing out such a vulnerability. This finding needs to be examined in larger studies.