OBJECTIVE The effectiveness of brief psychotherapy is established, but criteria for the selection of patients remain elusive. This study examines the usefulness of some widely used criteria for the selection of patients for brief psychodynamic therapy (BPT). METHODS A checklist of 39 criteria often used in a teaching hospital to screen patients for BPT was constructed. Scores derived from a principal components analysis of this checklist provided one set of predictor criteria. A second measure was derived from the consensual agreement of the independent ratings by 3 prominent advocates of BPT. A third predictor was derived from a similar analysis of items from the checklist on which consensual agreement for their relevance was obtained from the independent ratings of 23 BPT teachers in 16 university-affiliated hospitals. Reliable chart-based ratings of improvement in symptoms, increased insight, and improvement in general functioning of a convenience sample of 43 patients treated with BPT in the outpatient clinic of a teaching hospital served as outcome measures. RESULTS No correlation between any of the derived predictors and any of the measures of outcome was significant after a Bonferroni correction. CONCLUSIONS Teachers and clinicians still do not have good criteria for the selection of patients who will benefit from BPT. Exclusion of severe behavior disorders may, however, improve outcome rates by more than 15%.