The correspondence between the Staff Observation Aggression Scale-Revised and two other indicators for aggressive incidents

Previous research has shown good psychometric properties of the Staff Observation Aggression Scale-Revised (SOAS-R). However, it has never been investigated what proportion of aggressive incidents occurring in facilities is documented with the SOAS-R. Furthermore, if incidents are underreported, the consequences for the categorization of clients into aggressive and nonaggressive subgroups based on the SOAS-R are unknown. To examine this, in four inpatient psychiatric facilities for adults with mild intellectual disabilities, aggressive incidents were documented with the SOAS-R and two other indicators of aggressive incidents: the daily staff reports on clients' behavior and reports on of the use of restraints. Less than half of the incidents documented with the staff and restraint reports were also documented with the SOAS-R. On the other way around, however, it was also found that a substantial proportion of incidents reported on SOAS-R forms were not documented in the daily staff reports, which points to a more general problem of underreporting aggressive behavior. Apart from that, categorization of clients into an aggressive and a nonaggressive subgroup with SOAS-R data collected during 1 month or longer corresponded largely with the categorization based on both other indicators. This study showed that underreporting of aggressive incidents is likely to occur with the SOAS-R, making the instrument less suitable to assess absolute aggression incidence in facilities. Still, the SOAS-R seems a good instrument to categorize clients into aggressive and nonaggressive subgroups. Ways to improve the "compliance" of the ward team to document all aggressive incidents are addressed in the Discussion section of this article. © 2009 Elsevier Inc. All rights reserved.