There have been numerous studies of the effects of neuroleptic discontinuation in young schizophrenic patients. Yet little information is available regarding the same in older schizophrenic patients. Longitudinal studies of the course of schizophrenia have found that over time many patients fully recover or significantly improve. At the same time, the risk of persistent tardive dyskinesia (TD) is high among older patients maintained on neuroleptics. Hence a study of the effects of neuroleptic discontinuation on relapse aid on TD in the older subject population is warranted. We will study 160 chronic schizophrenic patients over age 45 (mean age 60 to 65 years), recruited over the first 4-year period. Baseline psychiatric assessments will be done after a patient is stabilized on the present neuroleptic dose for at least 1 month. These will include: Brief Psychiatric Rating Scale (BPRS); Scales for Assessment of Positive Symptoms (SAPS); Scale for Assessment of Negative Symptoms (SANS); Positive and Negative Syndrome Scale (PANSS); Schedule for the Deficit Syndrome (SDS); Dementia Rating Scale (DRS); Abnormal Involuntary Movement Scale (AIMS); Hamilton Rating Scale for Depression (HAM-D); Cumulative Illness Rating Scale - Geriatrics (CIRS-G); and Brief Symptom Inventory (BSI). The neuroleptic medication will be reduced by 25% every other week and discontinued at the end of the sixth week. We expect 100 patients to complete neuroleptic taper. Most rating scales will be repeated biweekly during the taper and for 4 weeks afterwards, and monthly thereafter. All the ratings will be done by """"""""blind"""""""" raters. Patients will be maintained off neuroleptics unless and until they relapse. We hypothesize that: (l) Schizophrenic patients who are older, those who have later onset of illness, and those of the paranoid or residual type will be more likely to complete neuroleptic discontinuation; (2) The risk of relapse will be negatively associated with current age and age of onset of schizophrenia and positively associated with severity of positive symptoms, degree of cognitive impairment and presence of TD at baseline; (3) Among schizophrenic patients who complete neuroleptic taper, there will be a positive correlation between change in severity of TD and change in negative symptoms; and (4) Remission of TD after neuroleptic discontinuation will be associated with lower AIMS score at baseline, younger age and lower negative and affective symptom scores at baseline. We will use survival analysis and other appropriate statistical analyses for testing the hypotheses. We believe that the results of our study will have considerable clinical and theoretical implications.