Abstract
Background Seclusion is a restrictive intervention that results in some form of containment and social isolation of a person from others. Little is known about the relationships between individual and organisation factors and the use of seclusion in disability services.
Method The reported use of seclusion in disability services in Victoria, Australia, was examined over a 3-year period, with a focus on the characteristics of those who were secluded (n = 146) and the characteristics of organisations that reported seclusion compared to others who were reported to be restrained but not secluded (n = 2,482).
Results Results from a logistic regression showed that the individual factors of age, the presence of autism and/or a psychiatric disorder put people at risk of being secluded. In terms of organisational factors, receiving accommodation services in institutions or in the community and the location of the organisation were risk factors.
Conclusions The findings are consistent with previous research but add to this literature by showing that certain organisational characteristics are also risk factors for seclusion. Understanding these factors is important in order to help disability support staff find other more ethical and appropriate alternatives to seclusion.
Author note
Some of the data in this paper was presented at the 47th Annual Conference for Australasian Society for Intellectual Disability: Research to Practice, Wellington, New Zealand, in November, 2012.
Acknowledgements
The authors would like to thank Anthony La Sala and Robin Dale for their assistance with the data collation and coding.
Notes
† This manuscript was accepted under the Editorship of Susan Balandin.
1. The majority of this restraint is in the form of “routine chemical restraint,” defined as medication administered on a regular basis (e.g., daily) to control behaviour.
2. Additional logistic regression models were run to explore interactions between diagnosis (autism, psychiatric disorder) and service setting in addition to diagnosis and region (rural versus metropolitan); however, none of these interactions were significant and the models are not reported here.
3. Unfortunately, we could not include quality of support plan in the main analysis. Not all support plans are rated for quality; instead, this data exist only for a small subset of the total individuals included in the sample, precluding statistical consideration of that variable.