Some methodological issues in using aggression rating scales in intervention studies among institutionalized elderly

Abstract

Background: Aggressive behavior among geriatric psychiatry inpatients spontaneously declines during serial measurements over time with rating scales.

Method: Evidence for decline in aggressive behavior, over a 6-week period, on a continuing-care geriatric psychiatry ward was examined with the Staff Observation Aggression Scale (SOAS) and the Rating Scale for Aggressive Behavior in the Elderly (RAGE).

Results: The RAGE and all its subitems did not spontaneously decline. The total number of aggressive incidents on the SOAS, the SOAS total score, and the subitem scores, except for the SOAS results subitem, spontaneously declined after Week 4 and became established by Week 5.

Conclusion: A "run in" period of at least 4 weeks may be needed in intervention studies using the SOAS to reduce contamination by spontaneous decline.