Assessing interactions between residents and staff in long term care
Research type
Research Study
Full title
A comparison of conditional probability analysis to experimental functional assessment for individuals diagnosed with dementia who demonstrate disruptive behaviour in long term care facilities
IRAS ID
226287
Contact name
Rebecca Sharp
Contact email
Sponsor organisation
Bangor University
Duration of Study in the UK
2 years, 4 months, 29 days
Research summary
Behaviour that challenges care providers whether familial or professional can have a serious negative effect on the quality of communication that occurs between the individual with dementia and their circle of support and on the quality of life of both the individual and their carers(Allen-Burge, 1999). Therefore finding a successful treatment intervention is crucial. It has been shown that identifying the cause of inappropriate behaviour is essential to providing the most effective form of treatment (Mace, 1991). Behaviour analysts refer to cause as function and there are various forms of functional behaviour assessment. The most reliable method is experimental functional analysis (Iwata, 1982) as it is the only method which demonstrates function with experimental control.
An issue with functional analysis is that it requires a level of discimination ability between different conditions in the analysis. This is known as stimulus control and there is emerging evidence that this may be impaired in people with dementia (Gallagher & Keenan, 2009). A possible alternative to experimental functional analysis is to use descriptive recording and then conduct a conditional probability analysis. This examines relationships between events in the environment and the behaviour of interest to determine if there is a correlation between specific event types and behaviour that challenges.This research aims to investigate the use of conditional probability as a tool for determining the function of behaviour that challenges and compare it to experimental functional analysis to find if there are more reliable results with one method over the other. If conditional probability is found to be more reliable in this population, there is scope for this to be used as a method for determining function that can inform effective intervention design and does not require manipulation of the environment, and thus is more naturalistic.
REC name
Wales REC 5
REC reference
17/WA/0246
Date of REC Opinion
19 Sep 2017
REC opinion
Further Information Favourable Opinion