A brief mental health outcomes scale: relationships between scale scores and diagnostic/sociodemographic variables in the long-term mentally ill The Global Assessment of Functioning (GAF) scale is a quick and simple measure of overall psychological disturbance. A recent study by our group reported the reliability and validity of the scale in a population of severely mentally ill people. The current study extends the analysis of this population to consider a number of questions, i.e. what are the associations for this group between variations in GAF scores and sociodemographic variables? Are lower scores associated with particular diagnoses for the group assessed? What are the characteristics of those patients whose scores were most variable? What were the completion rates for use of the scale? No associations were observed between GAF score and the sociodemographic variables used in this study. Lower GAF scores were associated with psychotic diagnoses. Those with most varying scores were, in the main, white European nonimmigrant males with a diagnosis of schizophrenia. Completion rates for the study were reasonable at 73.8%. This study involved the repeated assessment of a large number of clients by a changing multidisciplinary team. Inspite of the extended nature of the study, completion rates remained reasonable, suggesting that the scale was acceptable for routine clinical use. The GAF in itself, however, appears to be unsuitable for making individual treatment decisions but could be used as a first step before using more sophisticated scales.