The purpose of the study was to examine the predictive value of clinical and socio-vocational variables of first-admitted patients with delusional psychoses as regards clinical course and outcome, which was illustrated by the dimensions: psychotic symptoms, impairment, remission, and relapse at a 4-year re-follow-up. An excess of patients had poor course and outcome. In the multivariate logistic analysis an independent statistically significant effect was assigned to male sex, low social group, few social contacts, no work, and long duration of psychosis. The most informative clinical and social predictors yielded a maximal information of 19-50% for the different dimensions of course and outcome. One-fifth of the patients had good outcome. Living with others and at the same time having frequent social contacts was a variable with an independent statistically significant effect. The most informative predictors of good outcome yielded a maximal information of 34%. Differences between predictors of 2-year and 4-year course and outcome are discussed. It is concluded that good prognosis is exceptional in patients with delusional psychoses and from a nosological approach attached to patients with reactive psychosis. However, socio-vocational variables at index admission are predictors of higher statistical information than any clinical variable.