[Skip to Content]
[Skip to Content Landing]
March 1991

National Institute of Mental Health Longitudinal Study of Chronic SchizophreniaPrognosis and Predictors of Outcome

Author Affiliations

From the Maryland Psychiatric Research Center, Department of Psychiatry, University of Maryland School of Medicine, Baltimore (Dr Breier); the Section on Clinical Studies, Clinical Neuroscience Branch, National Institute of Mental Health, Bethesda, Md (Ms Schreiber and Dr Pickar); and the Department of Psychiatry, University of Pittsburgh (Pa) (Ms Dyer).

Arch Gen Psychiatry. 1991;48(3):239-246. doi:10.1001/archpsyc.1991.01810270051007

• We performed a longitudinal study of chronic schizophrenic patients who were hospitalized for research purposes at the National Institute of Mental Health (NIMH) Intramural Program in the 1970s and early 1980s. We assessed present course, outcome and predictor data from the initial cohort of 58 young chronic schizophrenic patients who were followed up for 2 to 12 years following their NIMH index hospitalization. At follow-up, the sample showed substantial functional impairment and levels of symptoms with only about 20% of the sample demonstrating a good outcome. In addition, strong intercorrelation was noted among the symptom and functioning indexes at follow-up. Moreover, neuropsychologic tests of frontal cortical functioning were significantly correlated with outcome levels of negative symptoms and social functioning but not with levels of positive symptoms. During the period from the index hospitalization to the follow-up assessment, 78% of the sample suffered a relapse, 38% attempted suicide and 24% had episodes of major affective illness. Furthermore, levels of positive and negative symptoms ascertained when patients received optimal neuroleptic treatment during the index hospitalization significantly predicted outcome levels of symptoms and functioning and time spent hospitalized during the follow-up period. In contrast, levels of index positive and negative symptoms ascertained during the drugfree state did not predict outcome symptoms or functioning. These data suggest that treatment response is a critical predictor variable. We examined the implication of these data for the course of illness in schizophrenics.