A sample of 100 schizophrenic patients admitted to London area psychiatric hospitals were examined as part of the International Pilot Study of Schizophrenia. Clinical and social outcome were variable. At 5-year follow-up, 49% had good symptomatic outcome and 42%, good social outcome. Poor social functioning at inclusion was predictive of poor symptomatic outcome. Illness history (IH), occupational functioning (OF), social relationship functioning (SRF), negative and non-specific symptoms (NNS) at initial evaluation, and their interaction with sex and race accounted for 32% of the 5-year symptomatic outcome variance (n = 84; F=3.48; P<0.001). OF, SRF, housing status, NNS and their interaction with sex, race and age accounted for 47% of the social outcome variance (n = 62; F=2.62; P<0.007).