Guidelines for rating Global Assessment of Functioning (GAF)

Abstract

BACKGROUND: Global Assessment of Functioning (GAF) is a scoring system for the severity of illness in psychiatry. It is used clinically in many countries, as well as in research, but studies have shown several problems with GAF, for example concerning its validity and reliability. Guidelines for rating are important. The present study aimed to identify the current status of guidelines for rating GAF, and relevant factors and gaps in knowledge for the development of improved guidelines. METHODS: A thorough literature search was conducted. RESULTS: Few studies of existing guidelines have been conducted; existing guidelines are short; and rating has a subjective element. Seven main categories were identified as being important in relation to further development of guidelines: (1) general points about guidelines for rating GAF; (2) introduction to guidelines, with ground rules; (3) starting scoring at the top, middle or bottom level of the scale; (4) scoring for different time periods and of different values (highest, lowest or average); (5) the finer grading of the scale; (6) different guidelines for different conditions; and (7) different languages and cultures. Little information is available about how rules for rating are understood by different raters: the final score may be affected by whether the rater starts at the top, middle or bottom of the scale; there is little data on which value/combination of GAF values to record; guidelines for scoring within 10-point intervals are limited; there is little empirical information concerning the suitability of existing guidelines for different conditions and patient characteristics; and little is known about the effects of translation into different languages or of different cultural understanding. CONCLUSIONS: Few studies have dealt specifically with guidelines for rating GAF. Current guidelines for rating GAF are not comprehensive, and relevant points for new guidelines are presented. Theoretical and empirical studies, and international expert panels would be valuable, as well as production of a manual with more information about scoring. Computerised assessment may well be the future.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 346 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.