Improving the assessment of Intellectual Disability (ID) within the UK Prison Service (who define ID using an IQ below 80)

Tools

Smith, L, 2016. Improving the assessment of Intellectual Disability (ID) within the UK Prison Service (who define ID using an IQ below 80). PhD, Nottingham Trent University.

Preview

Text
LORRAINE.SMITH-2016.pdf - Published version
Download (3MB) | Preview

Abstract

The main topic of this thesis is the assessment of Intellectual Disabilities (ID) within the UK prison service. ID is characterised by deficits in intellectual ability, such as reasoning, problem solving and understanding new or complex information (impaired intelligence), deficits in adaptive functioning (AF), which reduce the individual’s ability to function independently within their social environment, and these deficits begin before adulthood (Diagnostic and Statistical Manual (DSM)-5, APA, 2013). It is important to note that the assessments at the centre of this thesis are based on the English prison system definition of ID; IQ below 80. This is in contrast to the internationally recognised and accepted definition of ID defined as an IQ less than 70 (although this criteria has been removed from the most recent DSM, the DSM-5 (APA, 2013)). The prison service definition of ID using an IQ of below 80 includes those with IQ in the borderline range for determining ID, for which the Becoming New Me (BNM) treatment programme is suitable for, in addition to those with an IQ less than 70. This thesis comprises three empirical studies that focussed on improving the current assessment of ID (as defined by the prison service as having an IQ less than 80) within the UK prison service. All of the studies employed quantitative methodologies and participants were recruited from a UK prison for sex offenders. The first study aimed to assess the psychometric properties of a new IQ screening measure, the OASys Screening Tool (OASys ST), which was developed by NOMS to replace the Wechsler Abbreviated Scale of Intelligence (WASI) as an IQ screening measure used to identify individuals with an Intelligence Quotient (IQ) indicative of ID as defined by the UK prison service (below 80). The sample comprised 80 adult male prisoners, whose OASys data was accessed and used to complete the OASys ST. The analysis includes a probabilistic model of the data which was developed to assess the effectiveness of the OASys ST, using IQ data (WASI and WAIS scores) and Treatment Programme (TP) data. A logistic regression was also conducted and, in order to inform item redundancy, pairwise correlations were calculated. The OASys ST was found to be an accurate predictor of whether an individual’s IQ is above or below the threshold of 80; using this cut-off it was possible to classify all the individuals who scored two or less on the OASys ST as above the IQ of 80 threshold and the probability of making a mistake with these classifications was at most 3%, this rose to 7% if the cut-off was three and 15% if the cut-off was four. These individuals could be placed straight onto the CORE sex offending treatment programme without any further IQ testing. As described within this thesis, historically IQ has been the sole criterion relied on for determining treatment suitability within the prison service (Sparrow et al., 2005). However, as shown in the DSM-5 (APA, 2013), an ID diagnosis requires an assessment of both IQ and AF. There is evidence of poor identification of offenders with ID, including sex offenders, by the Criminal Justice System (CJS) (Banes, 2002; HMIP, 2015) because there is no commonly used process for this identification (Beebee, 2009; HMIP, 2015), since current measures employed to measure AF in the community are inappropriate for use on incarcerated populations (Young, Boccaccini, Conroy, & Lawson, 2007). Previously, Sex Offenders with ID (SOIDs) have been at a disadvantage regarding treatment programmes and supports available, but they have recently been the focus of research and policies, resulting in the creation of the Becoming New Me (BNM) treatment programme which was designed specifically to meet the needs of SOIDs (defined by the prison service as having an IQ less than 80). The existing literature indicates that having a reliably sound AF measure suitable for use within prisons is important in ensuring prisoners are placed onto the most appropriate treatment programme and that adequate supports are implemented in line with the Disability Discrimination Act (DDA) (2005). As such, the aim of study two was to develop an adaptive functioning screening measure which will be used alongside measures of IQ to assess ID. The stages employed to develop the new measure are summarised briefly as follows: A conceptual framework of AF was developed via consulting the diagnostic criteria set out in the DSM-5 (APA, 2013) and by reviewing the current community measures of AF and the ID literature. A sample of 11 prisoner and 11 staff participants took part in interviews about daily life inside prison, the results of which were used to produce the items. Originally 115 items were developed and pre-tested by a sample of experts. Item response theory was utilised to reduce the item pool. Forty-six items were retained in the scale which produced 95% of the maximum certainty of the original 115 item scale. The scale produced was named the Adaptive Functioning Assessment Tool (AFAT). The AFAT is the first AF assessment tool that is appropriate to use within a prison environment that has been created following a systematic process of scale development, the stages of which are explained in more detail within this thesis. The final study aimed to assess the psychometric properties of the new AF measure, the AFAT, to see whether it is a valid and reliable measure of AF that could potentially be rolled out throughout the prison service. The results indicated that the AFAT is a reliable measure of AF; the Cronbach’s alphas for each of the subscales were all above the .7 level recommended by Nunnally (1978) and all four sub-scales correlated positively with one another, as well as with the full scale AFAT score. Although the reliability levels vary from item to item, an average inter-item correlation of .91 was obtained, with all the individual correlations exceeding the recommended limit of .3 (Cronbach & Meehl, 1955). The AFAT was also found to have a good level of validity; by using scores on the AFAT, an accurate prediction could be made on which treatment programme participants’ had been referred for (Chi Squared test residual deviance = 32.45, p < 0.001). As expected, there was a significant negative correlation between the AFAT and LD diagnosis (r = -.67, p < .01), and there were significant positive correlations between the AFAT and WASI scores (r = .62, p < .01) and ratings of overall AF (r = .65, p < .01). There was also a significant negative correlation between scores on the AFAT and OASys ST full scale scores (r = -.68, p < .01). Also, as predicted, there was no correlation between participants’ age and their score on the AFAT (r = .11, p > .05). The AFAT showed a high level of content validity; 40 out of the 46 items were rated as content valid by all experts resulting in an S-CVI of .87 and the S-CVI-Average was .96. In an attempt to interpret the test scores on the AFAT, two Latent Class Analyses (LCA) were conducted; one treating the responses as categorical and the other treating the responses as continuous variables. Both analyses revealed three distinct classes of individuals, as expected, reflecting high, medium and low AF groups. The second LCA analysis also revealed two minor classes. The interpretation of each class is described, including how these varying AF levels present themselves among the different classes, including how the different levels of AF are manifested across the four sub-scales constituting the AFAT. The thesis offers an insight into the effectiveness of the OASys ST and also highlights the value in having an AF measure which is important in assessing support needs (HMIP, 2015), determining the most effective treatment programme and informing treatment delivery in line with the Risk Need Responsivity principles (Andrews & Bonta, 2010). The studies offer an original contribution to the knowledge regarding the assessment of prisoners defined by the prison service as having ID (IQ below 80), since the OASys ST is a new tool, that before this research had yet to receive an evaluation of the psychometric properties other than during the development of the tool itself. The AFAT is the first AF measure that is suitable to be used within a prison setting, which has been developed systematically and been subjected to reliability and validity testing. Both tools have direct implications for the prison service, if adopted by the prison service they are quicker than the current available tools and can be used by non-psychology and non-psychometric trained staff and have both been shown to produce results that can be relied upon. However, because the sample consisted of sex offenders only, and was conducted in a single UK prison, further research and testing is recommended.

Item Type:	Thesis
Description:	This research programme was carried out in collaboration with HMP Whatton
Creators:	Smith, L.
Date:	March 2016
Rights:	This work is the intellectual property of the author. You may copy up to 5% of this work for private study, or personal, non-commercial research. Any re-use of the information contained within this document should be fully referenced, quoting the author, title, university, degree level and pagination. Queries or requests for any other use, or if a more substantial copy is required, should be directed to the owner of the intellectual property rights.
Divisions:	Schools > School of Social Sciences
Record created by:	Linda Sullivan
Date Added:	28 Jun 2016 15:29
Last Modified:	28 Jun 2016 15:29
URI:	https://irep.ntu.ac.uk/id/eprint/28031