To the Editor: In the era of typical antipsychotic medications, tardive dyskinesia was a common and often undiagnosed disorder (1). The newer atypical antipsychotic medications hold the promise of a lower prevalence of movement disorders. We conducted a study whose goal was to determine if this promise has translated from the systematic research to a naturalistic sample.

Evaluations were completed for 162 patients from Central State Hospital in Petersburg, Va. The sample consisted of chronically ill, treatment-resistant adult patients, most of whom had psychotic disorders (70%) or bipolar disorder (9%) based on DSM-IV criteria. The majority (94%) of patients were taking antipsychotic medication. Of this subset, 25% were taking a typical antipsychotic but not an atypical antipsychotic, 52% were taking an atypical antipsychotic but not a typical antipsychotic, and 23% were taking both. Each patient was rated with the Involuntary Movement Scale (2). The primary rater (D.E.R.) was trained by the developers of this scale and achieved a high level of reliability with respect to gold standard ratings (intraclass correlation coefficient [ICC]=0.96 for parkinsonism global rating, p<0.001; ICC=0.79 for dyskinesia global rating, p<0.001). A diagnosis of parkinsonism was defined as having a rating of 2 (mild) or higher on the global rating scale. Tardive dyskinesia was diagnosed by the criteria of Schooler and Kane (3).

Per hospital policy, almost all of these patients had a previous examination for dyskinesia by their attending physician with the Abnormal Involuntary Movement Scale (4). These data were used to compare the prevalence of diagnoses of tardive dyskinesia between the attending physicians and the authors of the current study.

Tardive dyskinesia was present in 40% of the patients (40% of the patients taking only typical antipsychotics, 39% of the patients taking only atypical antipsychotics, and 47% of the patients taking typical and atypical antipsychotics). Of the patients with tardive dyskinesia, 57% had not been diagnosed previously with tardive dyskinesia. Of the patients newly diagnosed with tardive dyskinesia, 48% were mildly ill, 48% were moderately ill, and 4% were severely ill.

For patients taking any antipsychotic medication, 49% had parkinsonism (67% of the patients taking only typical antipsychotics, 34% of the patients taking only atypical antipsychotics, and 68% of the patients taking typical and atypical antipsychotics). Parkinsonism was uncommonly mentioned as an adverse effect, and no rating scale was used by the attending physicians to assess parkinsonism.

A substantial number of patients had tardive dyskinesia or parkinsonism. The high prevalence of newly diagnosed cases of tardive dyskinesia suggests a continuing need for better identification of these disorders and the opportunity for further preventing or treating these disorders.