Current knowledge about the advantages of using split version of GAF scale – having separate Symptom (GAF-S) and Functioning (GAF-F) subscales is insufficient and has number of gaps. Present study analyzed the manner in which young adult remitted psychotic patients with different functional levels vary in their symptom severity and how GAF-F reflects presence/absence of well-known psychosis risk factors.

Sample comprised 37 remitted psychotic patients (56.8% male, 29.4 ± 6.1 years), categorized based on GAF-F scores into groups: ‘Moderate remission’ (GAF-F 41–60, n = 19) and “Good remission” (GAF-F > 61, n = 18). Participants were assessed using the GAF-S, WAIS-R, Nottingham Onset Schedule, Premorbid Adjustment and Brief Core Schema scales, childhood trauma and bullying questionnaires. Mann–Whitney and Student's t-tests were used to assess between-group differences.

Groups did not differ in age, sex and education. When compared to “Good remission” group, “Moderate remission” group exhibited earlier illness’ onset (P = 0.01), greater symptom severity (P = 0.00), negative self-evaluation (P = 0.02), more childhood physical abuse (P = 0.01) and bullying (P = 0.01). Moreover, trend-like significances (P = 0.08) were observed for poorer adolescence adjustment and negative evaluation of others. There were no between-group differences regarding IQ and duration of untreated/treated illness.

GAF scale is in the continual use since early 1990, however, evidence of further development of the instrument itself is sparse. Present research is demonstrating that GAF-F scores reflect diverse factors related to psychosis risk, the illness’ course and quality of remission. Moreover, the results contribute to surmount some of the gaps in knowledge about the split version of GAF scale.

The authors have not supplied their declaration of competing interest.