bs_bs_banner
Journal of Intellectual Disability Research 1004
doi: 10.1111/jir.12109
volume 58 part 11 pp 1004–1014 november 2014
Intellectual disability in young people in custody in New South Wales, Australia – prevalence and markers L. Haysom,1 D. Indig,2 E. Moore3 & C. Gaskin1 1 Adolescent Health, Justice Health & Forensic Mental Health Network, Sydney, NSW, Australia 2 University of New South Wales, Sydney, NSW, Australia 3 Centre for Research into Criminal Justice, Justice Health & Forensic Mental Health Network, Sydney, NSW, Australia
Abstract Background Intellectual disability (ID) is known to be more common in incarcerated groups, especially incarcerated youth. Aboriginal young people have higher rates of ID, and make up half of all youth in juvenile custody in New South Wales (NSW), Australia. We aimed to describe the prevalence of possible ID and borderline intellectual functioning (BIF) in young people in NSW custody, and to describe the association between possible ID and Aboriginality after adjusting for the inequalities in social disadvantage. Methods Baseline study of all youth in NSW Custodial Centres between August and October 2009, with 18-month follow-up. Using Wechsler Intelligence Scale for Children – Fourth Edition (WISCIV) and Wechsler Adult Intelligence Scale – Fourth Edition (WAIS-IV) cognitive assessments, possible ID was defined as Extremely Low Intellectual Quotient range (Full Scale Intellectual Quotient, FSIQ < 70), and possible BIF was defined as Correspondence: Dr Leigh Haysom, Adolescent Health, Justice Health & Forensic Mental Health Network, Suite 305, 152 Bunnerong Road, Pagewood, Sydney, NSW 2035, Australia (e-mail: [email protected]).
Borderline IQ range (FSIQ < 80). Risk factors for possible ID and BIF included age, gender, Aboriginality, socio-economic disadvantage, offending history and psychological disorders. Results N = 295 (65%) of all young people in NSW custody completed cognitive and psychological assessments (87% male, 50% Aboriginal, average age 17 years). Almost one half (45.8%) of young people had borderline or lower intellectual functioning (by IQ assessment), and 14% had an IQ in the extremely low range (FSIQ < 70), indicating a possible ID. Aboriginal participants were three times more likely than non-Aboriginal participants to have a possible ID, but after accounting for the excess disadvantage in the Aboriginal group, Aboriginality was no longer a marker of ID. Incarceration from a young age and psychosis were significantly associated with possible ID in Aboriginal participants, compared with Aboriginal participants first incarcerated at a later age, and Aboriginal participants without psychosis. Conclusion The inequalities in criminal justice between Aboriginal and non-Aboriginal youth may exacerbate or contribute to the intellectual impairment of those incarcerated from a young age. Aboriginal young people with psychosis are also at high
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research
volume 58 part 11 november 2014
1005 L. Haysom et al. • ID in young people in custody
risk of cognitive impairments that might indicate a possible co-morbid ID, and these patients should be diverted at court into community assessment services, rather than incarcerated. These results highlight a need for better and earlier identification of young people (particularly Aboriginal youth) at risk of ID and other co-morbidities in the juvenile justice system. Keywords ADHD, intellectual disability, mental health, psychiatric disorders
Introduction Intellectual disability (ID) in people involved with the criminal justice system has been described in a number of settings, with varying rates of frequency. A large meta-analysis from 2008 described rates of ID in prisoner groups that were similar to the community of 0.5% to 1.5% (Fazel et al. 2008). An Australian study also described rates of ID in adult prisoners that were similar to the community prevalence of 1.3% (Holland & Persson 2011), but in another study, over 10% of Australians presenting to local magistrates courts had possible ID, with a further 20% in the borderline intellectual functioning (BIF) range (Vanny et al. 2009). A review of UK studies showed a prevalence of ID or BIF of 7% to over 50% (Mason & Murphy 2002; Herrington 2009). This wide disparity in results reflects the different age and socio-economic (SE) background of offenders (Hayes 2006), their co-morbidities (Raina & Lunsky 2010; Lindsay 2011), offence types (Sondenaa et al. 2008), the range of custodial settings (court and community versus custodial, and high versus low security environments) (Hogue et al. 2006), and the application of screening tools (versus diagnostic instruments) and non-standardised diagnostic systems. Young Australian offenders on community orders had rates of possible ID of 15% (Frize et al. 2008), which is comparable to rates of 10% to 17% among young people entering UK and Australian juvenile detention (Allerton et al. 2003; Herrington 2009). The reasons for this over-representation in younger offenders have not been properly explored. Incarcerated young people have very high rates of mental illness (Allerton et al. 2003), a known association with ID as the ‘Dual Diagnosis’ (Einfeld et al. 2011;
Matson & Shoemaker 2011), and previously described in adults in contact with the criminal justice system (Riches et al. 2006; Vanny et al. 2009). Almost one half (48%) of all young people in custody in New South Wales (NSW) are Aboriginal, a 17 times over-representation (Juvenile Justice NSW 2012). According to census figures, 7% of Indigenous Australians aged 15 years and over have an ID (ABS Australian Social Trends 2006), which is three times the rates in the Australian population. Aboriginal children have lower rates of literacy and numeracy than non-Aboriginal children (Ministerial Council for Education, Early Childhood Development and Youth Affairs 2008; D’Aprano et al. 2011), and risk for developmental delay is higher because of factors such as lower birth weights (Sayer et al. 2003), a higher prevalence of foetal alcohol syndrome (Burns et al. 2009), social disadvantage and limited access to specialised education (Penman 2009). Intellectual disability and its co-morbidities affect the outcomes for young people at every stage of the criminal justice process, including offence-type classification, length of sentence, presentation in the courtroom, behaviour in custody, and meeting conditions of parole (Smith et al. 1990; Riches et al. 2006; Vanny et al. 2009; Baldry et al. 2012). Accurately identifying young people with ID presenting to court or entering custody is therefore crucial in establishing opportunities for diversion and referral (Hayes 2006; Lindsay et al. 2010; Mulder et al. 2010), but has been fraught with difficulties in the Australian setting because of limited access to young people in these environments, interagency issues, and the cultural appropriateness of using screening tools and diagnostic instruments that have only been validated in nonIndigenous populations (Wechsler 2003, 2008; McKenzie et al. 2012). This study is the first to describe the prevalence of possible ID and possible BIF in Aboriginal and non-Aboriginal young people in custody in NSW, Australia, and the association of possible ID with demographic, offending and psychosocial risk factors. This knowledge will assist in the development of age and culturally appropriate screening tools within the NSW juvenile justice system to better identify those young people at risk of an ID.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1006 L. Haysom et al. • ID in young people in custody
Methods
Risk factors
Study design
At baseline, participants were asked to complete a questionnaire (administered face-to-face) which included information on demographic characteristics. Indigenous status was self-reported, and determined by asking ‘are you of Aboriginal and/or Torres Strait Islander origin?’. Postcode of residence was used as a proxy measure of SE status based on the Australian Bureau of Statistics’ Index of Relative Disadvantage (Australian Bureau of Statistics SEIFA 2011), which includes indices on income, educational attainment, unemployment and proportion of people in unskilled occupations. This SE Disadvantage Score is an ordinal measure (based on a mean score of 1000 with a standard deviation of 100), with lower scores indicating lower SE status and higher disadvantage. Lifetime psychological disorders, including major depressive episode, anxiety disorders, psychotic disorders, attention deficit hyperactivity disorder (ADHD) and conduct disorder were assessed at baseline using the Kiddie Schedule for Affective Disorders and Schizophrenia – Present and Lifetime Version (KSADS-PL) 2009 Working Draft (Axelson et al. 2009). The 2009 version was adapted from the KSADS-PL, and this has demonstrated good reliability and validity (Kaufman et al. 1997). The Kessler Psychological Distress Scale (K10) was used to screen for psychological distress, with strong associations between high or very high scores (K10 > 22), and a current diagnosis of an affective or anxiety disorder (Kessler et al. 2002). Participants self-reported self-harm and suicide attempts, experience of out-of-home care placements, age of the first placement, exclusions and expulsions from school and any parental incarcerations. Attendance at a special school or class for specific educational purposes was also self-reported. Information on incarcerations and offence types at baseline, and re-incarcerations at 18-month follow-up were determined through data linkage to the NSW Juvenile Justice and Correctives Services’ databases.
Data were collected at baseline and at 18-month follow-up as part of the NSW Young People in Custody Health Survey (Indig et al. 2011), conducted between August and October 2009 in eight Juvenile Justice Centres, and one high security Juvenile Correctional Centre. Informed consent by participants or their carers was a requirement for participation. Ethics approvals were obtained from the Justice Health Human Research and Ethics Committee, the Juvenile Justice Research Committee, the Corrective Services NSW Ethics Committee and the Aboriginal Health and Medical Research Council Ethics Committee.
Participants All young people who were in custody between August and October 2009 were approached for participation. Exclusion criteria included inability to speak sufficient English or being unavailable at the time of the survey because of work or court commitments.
Cognitive assessments Cognitive functioning was measured at baseline using the Wechsler Adult Intelligence Scale – Fourth Edition (WAIS-IV) Australian and New Zealand Language Adaptation for young people aged 17 years and over (Wechsler 2008). All young people under the age of 16 years received the Wechsler Intelligence Scale for Children – Fourth Edition (WISC-IV) Australian Standardised Edition (Wechsler 2003). Participants who were 16 years old were able to complete either test, because of the overlap allowed in the age brackets for the tests, and was determined by the psychologists. Based upon Australian normative data, WISC-IV and WAIS-IV intelligence scores were categorised as ‘Extremely Low’ (Full Scale Intellectual Quotient, FSIQ < 70), ‘Borderline’ (FSIQ 70–79), ‘Low Average’ (FSIQ 80–89), ‘Average’ (FSIQ 90–109), ‘High Average/Superior’ (FSIQ ≥ 110). Possible ID was categorised as FSIQ < 70, and possible BIF was IQ 70–79.
Analysis Data were analysed using SPSS statistics package version 19 (SPSS IBM Corp 2010) and SAS system
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research
volume 58 part 11 november 2014
1007 L. Haysom et al. • ID in young people in custody
version 9.2 (SAS 2007). Only participants with complete cognitive and psychological assessments were included in the analysis (n = 295). Descriptive statistics (frequencies and chi-squared analyses) were used to examine the sample characteristics, FSIQ and sub-scale IQ scores by Aboriginality. The significance of associations between demographic, offending and psychological risk factors and possible ID (FSIQ < 70) were also examined using logistic regression analyses. Regression analyses were stratified by Aboriginality and adjusted for variables found to be significant in the crude odds ratio. Significance was at a P-value of less than 0.05.
Results Figure 1 shows a total of 361 young people participated in the survey, representing 80% of all young people in custody. Of these participants, 295 (82%) completed the cognitive and psychological assessments, which was 65% of all young people in custody. This sample was 87% male, 50% Aboriginal, with an average age of 17 years. Reasons for non-participation included unavoidable court and transfer commitments (15%), and a small number of young people (5%) refused consent. Cognitive assessments were not administered in a further 15% of young people because of time constraints, other study measurements taking priority, or inability to complete the assessment. The socio-demographic characteristics were not significantly different for those with intellectual functioning data, versus those without. Table 1 describes the socio-demographic characteristics, stratified by Aboriginality. One half (50%) of all young people came from areas of greater disadvantage, and 41% reported ever attending a special school or class. Almost one half (46%) of all young people had been excluded or expelled from school, and 36% were not attending school, Technical and Further Education (TAFE) or work in the 6 months prior to their incarceration. Compared with non-Aboriginal young people, Aboriginal young people were twice as likely to have been placed in care before 16 years of age (37% versus 18%, P < 0.001), to have ever had a parent incarcerated (63% versus 31%, P < 0.001), and to have left school before 14 years of age (34% versus 16%,
P < 0.05). There were no differences in rates of specialised education between Aboriginal and nonAboriginal young people. When compared with non-Aboriginal young people, Aboriginal young people were more likely to have had a previous incarceration (83% versus 64%, P < 0.001), and to have had three or more incarcerations (68% versus 47%, P < 0.001), were more than twice as likely to have been first incarcerated before 14 years of age (52% versus 21%, P < 0.001), and were more likely to be re-incarcerated in the 18 months of follow-up (56% versus 42%, P < 0.05). Despite this, Aboriginal young people were less likely to be incarcerated for a violent offence-type compared with non-Aboriginal young people (68% versus 91%, P < 0.001). Overall lifetime rates of psychiatric disorders were high (87%). Compared with non-Aboriginal young people, Aboriginal young people were more likely to have ever been diagnosed with a psychiatric disorder (93% versus 81%, P < 0.05). There were no significant differences between Aboriginal and nonAboriginal young people for specific mental health disorders. Table 2 describes the full scale and sub-scale IQ assessments, stratified by Aboriginality. Almost one half (46%) of young people had a FSIQ score that was in the borderline range or lower. Compared with non-Aboriginal young people, Aboriginal young people had significantly lower FSIQ scores (P < 0.001), and were three times more likely than non-Aboriginal young people to have an IQ in the extremely low range (FSIQ < 70, 20% versus 7%, P < 0.001). Aboriginal young people had significantly lower IQ scores across all sub-scale indices compared with non-Aboriginal young people (P < 0.05). In particular, Aboriginal young people were twice as likely to have extremely low range scores in verbal comprehension (Verbal Comprehension Index score < 70, 26% versus 13%, P < 0.001), borderline range perceptual reasoning (Perceptual Reasoning Index score 70–79, 25% versus 14%, P < 0.05), borderline range working memory (Working Memory Index score 70–79, 23% versus 10%, P < 0.05) and borderline range processing speed (Processing Speed Index score 70–79, 24% versus 12%, P < 0.05). Table 3 identifies significant associations between risk factors and possible ID (FSIQ < 70) in the total
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1008 L. Haysom et al. • ID in young people in custody
N=452 (100%) NSW Juvenile Custodial Population August – October 2009
Not invited N=70 (15%) Discharged Excluded (health) Unavailable (court, work) Transferred Unable to be seen
n=21 (30%) n=7 (10%) n=5 (7%) n=7 (10%) n=30 (43%)
Invited N=382 (85%) Female centre Urban centres Rural centres Maximum security centre
n=46 (12%) n=221 (58%) n=96 (25%) n=19 (5%)
Not consented (refused) N=21 (5%)
Consented N=361 (80%) Male Female Aboriginal Non-Aboriginal
n=319 (88%) n=42 (12%) n=174 (48%) n=187 (52%)
Completed psychological and cognitive assessments N=295 (65%) Male Female Aboriginal Non-Aboriginal
n=257 (87%) n=38 (13%) n=148 (50%) n=147 (50%)
Figure 1 Participant flowchart. NSW, New South Wales.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1009 L. Haysom et al. • ID in young people in custody
Table 1 Characteristics by Aboriginality
Characteristics Socio-demographic Male Age less than 16 years SE Disadvantage Score* Placed in care before 16 years of age Left school before 14 years age (if left) Ever attended special school or class Ever excluded/expelled from school Not attending school, TAFE, work in 6 months before incarceration History of parental incarceration Offending Any previous incarcerations First incarceration before 14 years age Three or more previous incarcerations Most serious offence-type ‘violent’ Any re-incarceration at 18-month follow-up† Psychological Any lifetime psychiatric disorder Major depressive episode at baseline Any anxiety disorder at baseline ADHD at baseline Conduct disorder at baseline Any psychotic disorder at baseline High/very high psychological distress (K10 22+) at baseline Ever self-harmed Ever attempted suicide
Non-Aboriginal (N = 147) n (%)
Aboriginal (N = 148) n (%)
Total (N = 295) n (%)
P-value
128 (87.1) 31 (21.1) 65 (46.8) 26 (18.4) 15 (16.0) 59 (42.4) 72 (51.1) 44 (31.7)
129 (87.2) 44 (29.7) 74 (54.0) 51 (37.0) 31 (34.1) 52 (38.8) 54 (41.2) 55 (40.1)
257 (87.1) 75 (25.4) 139 (50.4) 77 (27.6) 46 (24.9) 111 (40.7) 128 (46.2) 99 (35.9)
NS NS NS P < 0.001 P < 0.05 NS NS NS
42 (31.1)
82 (62.6)
124 (46.6)
P < 0.001
90 (63.8) 29 (20.6) 69 (46.9) 133 (90.5) 59 (41.5)
113 (82.5) 70 (51.5) 101 (68.2) 101 (68.2) 81 (56.3)
203 (73.0) 99 (35.7) 170 (57.6) 234 (79.3) 140 (49.0)
P < 0.001 P < 0.001 P < 0.001 P < 0.001 P < 0.05
117 (81.3) 26 (18.1) 31 (21.5) 39 (27.1) 77 (53.5) 1 (0.7) 40 (28.8)
127 (92.7) 22 (16.1) 30 (21.9) 44 (32.1) 90 (65.7) 7 (5.1) 39 (27.7)
244 (86.8) 48 (17.1) 61 (21.7) 83 (29.5) 167 (59.4) 8 (2.8) 79 (28.2)
P < 0.05 NS NS NS NS NS NS
20 (14.3) 13 (9.3)
26 (18.3) 16 (11.3)
46 (16.3) 29 (10.3)
NS NS
* Socio-Economic Index For Areas (SEIFA) score in quintile 1 (high disadvantage) or 2 (mid-disadvantage). † Excludes n = 9 young people continuously in custody over the 18-month follow-up period. ADHD, attention deficit hyperactivity disorder; K10, Kessler 10; NS, not significant; SE, socio-economic; TAFE, Technical and Further Education.
cohort, stratified by Aboriginality. After adjustment for variables found to be significant in the crude odds ratio, significant markers of possible ID in the overall group included first incarceration before 14 years of age [adjusted odds ratio (AOR) 5.28, 95% confidence interval (CI) 1.33–29.93, P < 0.05], and any psychotic disorder at baseline (AOR 9.43, 95% CI 1.94–45.84, P < 0.001). After stratification, these markers of possible ID were only significant for Aboriginal young people. Aboriginal young people first incarcerated before 14 years of age were almost 10 times more likely to have a possible ID compared with Aboriginal young people first incarcer-
ated at a later age (AOR 9.59, 95% CI 1.39–66.24, P < 0.05). Aboriginal young people diagnosed with a psychotic disorder at baseline were 20 times more likely to have a possible ID compared with Aboriginal young people with no psychosis (AOR 20.48, 95% CI 2.69–155.69, P < 0.001).
Discussion This study has shown that almost one half of all young people in custody in NSW have borderline or lower intellectual functioning (by IQ assessment),
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1010 L. Haysom et al. • ID in young people in custody
Table 2 Cognitive assessments (IQ) by Aboriginality
Intellectual Quotient (IQ) Full Scale IQ Extremely low (
Journal of Intellectual Disability Research 1004
doi: 10.1111/jir.12109
volume 58 part 11 pp 1004–1014 november 2014
Intellectual disability in young people in custody in New South Wales, Australia – prevalence and markers L. Haysom,1 D. Indig,2 E. Moore3 & C. Gaskin1 1 Adolescent Health, Justice Health & Forensic Mental Health Network, Sydney, NSW, Australia 2 University of New South Wales, Sydney, NSW, Australia 3 Centre for Research into Criminal Justice, Justice Health & Forensic Mental Health Network, Sydney, NSW, Australia
Abstract Background Intellectual disability (ID) is known to be more common in incarcerated groups, especially incarcerated youth. Aboriginal young people have higher rates of ID, and make up half of all youth in juvenile custody in New South Wales (NSW), Australia. We aimed to describe the prevalence of possible ID and borderline intellectual functioning (BIF) in young people in NSW custody, and to describe the association between possible ID and Aboriginality after adjusting for the inequalities in social disadvantage. Methods Baseline study of all youth in NSW Custodial Centres between August and October 2009, with 18-month follow-up. Using Wechsler Intelligence Scale for Children – Fourth Edition (WISCIV) and Wechsler Adult Intelligence Scale – Fourth Edition (WAIS-IV) cognitive assessments, possible ID was defined as Extremely Low Intellectual Quotient range (Full Scale Intellectual Quotient, FSIQ < 70), and possible BIF was defined as Correspondence: Dr Leigh Haysom, Adolescent Health, Justice Health & Forensic Mental Health Network, Suite 305, 152 Bunnerong Road, Pagewood, Sydney, NSW 2035, Australia (e-mail: [email protected]).
Borderline IQ range (FSIQ < 80). Risk factors for possible ID and BIF included age, gender, Aboriginality, socio-economic disadvantage, offending history and psychological disorders. Results N = 295 (65%) of all young people in NSW custody completed cognitive and psychological assessments (87% male, 50% Aboriginal, average age 17 years). Almost one half (45.8%) of young people had borderline or lower intellectual functioning (by IQ assessment), and 14% had an IQ in the extremely low range (FSIQ < 70), indicating a possible ID. Aboriginal participants were three times more likely than non-Aboriginal participants to have a possible ID, but after accounting for the excess disadvantage in the Aboriginal group, Aboriginality was no longer a marker of ID. Incarceration from a young age and psychosis were significantly associated with possible ID in Aboriginal participants, compared with Aboriginal participants first incarcerated at a later age, and Aboriginal participants without psychosis. Conclusion The inequalities in criminal justice between Aboriginal and non-Aboriginal youth may exacerbate or contribute to the intellectual impairment of those incarcerated from a young age. Aboriginal young people with psychosis are also at high
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research
volume 58 part 11 november 2014
1005 L. Haysom et al. • ID in young people in custody
risk of cognitive impairments that might indicate a possible co-morbid ID, and these patients should be diverted at court into community assessment services, rather than incarcerated. These results highlight a need for better and earlier identification of young people (particularly Aboriginal youth) at risk of ID and other co-morbidities in the juvenile justice system. Keywords ADHD, intellectual disability, mental health, psychiatric disorders
Introduction Intellectual disability (ID) in people involved with the criminal justice system has been described in a number of settings, with varying rates of frequency. A large meta-analysis from 2008 described rates of ID in prisoner groups that were similar to the community of 0.5% to 1.5% (Fazel et al. 2008). An Australian study also described rates of ID in adult prisoners that were similar to the community prevalence of 1.3% (Holland & Persson 2011), but in another study, over 10% of Australians presenting to local magistrates courts had possible ID, with a further 20% in the borderline intellectual functioning (BIF) range (Vanny et al. 2009). A review of UK studies showed a prevalence of ID or BIF of 7% to over 50% (Mason & Murphy 2002; Herrington 2009). This wide disparity in results reflects the different age and socio-economic (SE) background of offenders (Hayes 2006), their co-morbidities (Raina & Lunsky 2010; Lindsay 2011), offence types (Sondenaa et al. 2008), the range of custodial settings (court and community versus custodial, and high versus low security environments) (Hogue et al. 2006), and the application of screening tools (versus diagnostic instruments) and non-standardised diagnostic systems. Young Australian offenders on community orders had rates of possible ID of 15% (Frize et al. 2008), which is comparable to rates of 10% to 17% among young people entering UK and Australian juvenile detention (Allerton et al. 2003; Herrington 2009). The reasons for this over-representation in younger offenders have not been properly explored. Incarcerated young people have very high rates of mental illness (Allerton et al. 2003), a known association with ID as the ‘Dual Diagnosis’ (Einfeld et al. 2011;
Matson & Shoemaker 2011), and previously described in adults in contact with the criminal justice system (Riches et al. 2006; Vanny et al. 2009). Almost one half (48%) of all young people in custody in New South Wales (NSW) are Aboriginal, a 17 times over-representation (Juvenile Justice NSW 2012). According to census figures, 7% of Indigenous Australians aged 15 years and over have an ID (ABS Australian Social Trends 2006), which is three times the rates in the Australian population. Aboriginal children have lower rates of literacy and numeracy than non-Aboriginal children (Ministerial Council for Education, Early Childhood Development and Youth Affairs 2008; D’Aprano et al. 2011), and risk for developmental delay is higher because of factors such as lower birth weights (Sayer et al. 2003), a higher prevalence of foetal alcohol syndrome (Burns et al. 2009), social disadvantage and limited access to specialised education (Penman 2009). Intellectual disability and its co-morbidities affect the outcomes for young people at every stage of the criminal justice process, including offence-type classification, length of sentence, presentation in the courtroom, behaviour in custody, and meeting conditions of parole (Smith et al. 1990; Riches et al. 2006; Vanny et al. 2009; Baldry et al. 2012). Accurately identifying young people with ID presenting to court or entering custody is therefore crucial in establishing opportunities for diversion and referral (Hayes 2006; Lindsay et al. 2010; Mulder et al. 2010), but has been fraught with difficulties in the Australian setting because of limited access to young people in these environments, interagency issues, and the cultural appropriateness of using screening tools and diagnostic instruments that have only been validated in nonIndigenous populations (Wechsler 2003, 2008; McKenzie et al. 2012). This study is the first to describe the prevalence of possible ID and possible BIF in Aboriginal and non-Aboriginal young people in custody in NSW, Australia, and the association of possible ID with demographic, offending and psychosocial risk factors. This knowledge will assist in the development of age and culturally appropriate screening tools within the NSW juvenile justice system to better identify those young people at risk of an ID.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1006 L. Haysom et al. • ID in young people in custody
Methods
Risk factors
Study design
At baseline, participants were asked to complete a questionnaire (administered face-to-face) which included information on demographic characteristics. Indigenous status was self-reported, and determined by asking ‘are you of Aboriginal and/or Torres Strait Islander origin?’. Postcode of residence was used as a proxy measure of SE status based on the Australian Bureau of Statistics’ Index of Relative Disadvantage (Australian Bureau of Statistics SEIFA 2011), which includes indices on income, educational attainment, unemployment and proportion of people in unskilled occupations. This SE Disadvantage Score is an ordinal measure (based on a mean score of 1000 with a standard deviation of 100), with lower scores indicating lower SE status and higher disadvantage. Lifetime psychological disorders, including major depressive episode, anxiety disorders, psychotic disorders, attention deficit hyperactivity disorder (ADHD) and conduct disorder were assessed at baseline using the Kiddie Schedule for Affective Disorders and Schizophrenia – Present and Lifetime Version (KSADS-PL) 2009 Working Draft (Axelson et al. 2009). The 2009 version was adapted from the KSADS-PL, and this has demonstrated good reliability and validity (Kaufman et al. 1997). The Kessler Psychological Distress Scale (K10) was used to screen for psychological distress, with strong associations between high or very high scores (K10 > 22), and a current diagnosis of an affective or anxiety disorder (Kessler et al. 2002). Participants self-reported self-harm and suicide attempts, experience of out-of-home care placements, age of the first placement, exclusions and expulsions from school and any parental incarcerations. Attendance at a special school or class for specific educational purposes was also self-reported. Information on incarcerations and offence types at baseline, and re-incarcerations at 18-month follow-up were determined through data linkage to the NSW Juvenile Justice and Correctives Services’ databases.
Data were collected at baseline and at 18-month follow-up as part of the NSW Young People in Custody Health Survey (Indig et al. 2011), conducted between August and October 2009 in eight Juvenile Justice Centres, and one high security Juvenile Correctional Centre. Informed consent by participants or their carers was a requirement for participation. Ethics approvals were obtained from the Justice Health Human Research and Ethics Committee, the Juvenile Justice Research Committee, the Corrective Services NSW Ethics Committee and the Aboriginal Health and Medical Research Council Ethics Committee.
Participants All young people who were in custody between August and October 2009 were approached for participation. Exclusion criteria included inability to speak sufficient English or being unavailable at the time of the survey because of work or court commitments.
Cognitive assessments Cognitive functioning was measured at baseline using the Wechsler Adult Intelligence Scale – Fourth Edition (WAIS-IV) Australian and New Zealand Language Adaptation for young people aged 17 years and over (Wechsler 2008). All young people under the age of 16 years received the Wechsler Intelligence Scale for Children – Fourth Edition (WISC-IV) Australian Standardised Edition (Wechsler 2003). Participants who were 16 years old were able to complete either test, because of the overlap allowed in the age brackets for the tests, and was determined by the psychologists. Based upon Australian normative data, WISC-IV and WAIS-IV intelligence scores were categorised as ‘Extremely Low’ (Full Scale Intellectual Quotient, FSIQ < 70), ‘Borderline’ (FSIQ 70–79), ‘Low Average’ (FSIQ 80–89), ‘Average’ (FSIQ 90–109), ‘High Average/Superior’ (FSIQ ≥ 110). Possible ID was categorised as FSIQ < 70, and possible BIF was IQ 70–79.
Analysis Data were analysed using SPSS statistics package version 19 (SPSS IBM Corp 2010) and SAS system
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research
volume 58 part 11 november 2014
1007 L. Haysom et al. • ID in young people in custody
version 9.2 (SAS 2007). Only participants with complete cognitive and psychological assessments were included in the analysis (n = 295). Descriptive statistics (frequencies and chi-squared analyses) were used to examine the sample characteristics, FSIQ and sub-scale IQ scores by Aboriginality. The significance of associations between demographic, offending and psychological risk factors and possible ID (FSIQ < 70) were also examined using logistic regression analyses. Regression analyses were stratified by Aboriginality and adjusted for variables found to be significant in the crude odds ratio. Significance was at a P-value of less than 0.05.
Results Figure 1 shows a total of 361 young people participated in the survey, representing 80% of all young people in custody. Of these participants, 295 (82%) completed the cognitive and psychological assessments, which was 65% of all young people in custody. This sample was 87% male, 50% Aboriginal, with an average age of 17 years. Reasons for non-participation included unavoidable court and transfer commitments (15%), and a small number of young people (5%) refused consent. Cognitive assessments were not administered in a further 15% of young people because of time constraints, other study measurements taking priority, or inability to complete the assessment. The socio-demographic characteristics were not significantly different for those with intellectual functioning data, versus those without. Table 1 describes the socio-demographic characteristics, stratified by Aboriginality. One half (50%) of all young people came from areas of greater disadvantage, and 41% reported ever attending a special school or class. Almost one half (46%) of all young people had been excluded or expelled from school, and 36% were not attending school, Technical and Further Education (TAFE) or work in the 6 months prior to their incarceration. Compared with non-Aboriginal young people, Aboriginal young people were twice as likely to have been placed in care before 16 years of age (37% versus 18%, P < 0.001), to have ever had a parent incarcerated (63% versus 31%, P < 0.001), and to have left school before 14 years of age (34% versus 16%,
P < 0.05). There were no differences in rates of specialised education between Aboriginal and nonAboriginal young people. When compared with non-Aboriginal young people, Aboriginal young people were more likely to have had a previous incarceration (83% versus 64%, P < 0.001), and to have had three or more incarcerations (68% versus 47%, P < 0.001), were more than twice as likely to have been first incarcerated before 14 years of age (52% versus 21%, P < 0.001), and were more likely to be re-incarcerated in the 18 months of follow-up (56% versus 42%, P < 0.05). Despite this, Aboriginal young people were less likely to be incarcerated for a violent offence-type compared with non-Aboriginal young people (68% versus 91%, P < 0.001). Overall lifetime rates of psychiatric disorders were high (87%). Compared with non-Aboriginal young people, Aboriginal young people were more likely to have ever been diagnosed with a psychiatric disorder (93% versus 81%, P < 0.05). There were no significant differences between Aboriginal and nonAboriginal young people for specific mental health disorders. Table 2 describes the full scale and sub-scale IQ assessments, stratified by Aboriginality. Almost one half (46%) of young people had a FSIQ score that was in the borderline range or lower. Compared with non-Aboriginal young people, Aboriginal young people had significantly lower FSIQ scores (P < 0.001), and were three times more likely than non-Aboriginal young people to have an IQ in the extremely low range (FSIQ < 70, 20% versus 7%, P < 0.001). Aboriginal young people had significantly lower IQ scores across all sub-scale indices compared with non-Aboriginal young people (P < 0.05). In particular, Aboriginal young people were twice as likely to have extremely low range scores in verbal comprehension (Verbal Comprehension Index score < 70, 26% versus 13%, P < 0.001), borderline range perceptual reasoning (Perceptual Reasoning Index score 70–79, 25% versus 14%, P < 0.05), borderline range working memory (Working Memory Index score 70–79, 23% versus 10%, P < 0.05) and borderline range processing speed (Processing Speed Index score 70–79, 24% versus 12%, P < 0.05). Table 3 identifies significant associations between risk factors and possible ID (FSIQ < 70) in the total
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1008 L. Haysom et al. • ID in young people in custody
N=452 (100%) NSW Juvenile Custodial Population August – October 2009
Not invited N=70 (15%) Discharged Excluded (health) Unavailable (court, work) Transferred Unable to be seen
n=21 (30%) n=7 (10%) n=5 (7%) n=7 (10%) n=30 (43%)
Invited N=382 (85%) Female centre Urban centres Rural centres Maximum security centre
n=46 (12%) n=221 (58%) n=96 (25%) n=19 (5%)
Not consented (refused) N=21 (5%)
Consented N=361 (80%) Male Female Aboriginal Non-Aboriginal
n=319 (88%) n=42 (12%) n=174 (48%) n=187 (52%)
Completed psychological and cognitive assessments N=295 (65%) Male Female Aboriginal Non-Aboriginal
n=257 (87%) n=38 (13%) n=148 (50%) n=147 (50%)
Figure 1 Participant flowchart. NSW, New South Wales.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1009 L. Haysom et al. • ID in young people in custody
Table 1 Characteristics by Aboriginality
Characteristics Socio-demographic Male Age less than 16 years SE Disadvantage Score* Placed in care before 16 years of age Left school before 14 years age (if left) Ever attended special school or class Ever excluded/expelled from school Not attending school, TAFE, work in 6 months before incarceration History of parental incarceration Offending Any previous incarcerations First incarceration before 14 years age Three or more previous incarcerations Most serious offence-type ‘violent’ Any re-incarceration at 18-month follow-up† Psychological Any lifetime psychiatric disorder Major depressive episode at baseline Any anxiety disorder at baseline ADHD at baseline Conduct disorder at baseline Any psychotic disorder at baseline High/very high psychological distress (K10 22+) at baseline Ever self-harmed Ever attempted suicide
Non-Aboriginal (N = 147) n (%)
Aboriginal (N = 148) n (%)
Total (N = 295) n (%)
P-value
128 (87.1) 31 (21.1) 65 (46.8) 26 (18.4) 15 (16.0) 59 (42.4) 72 (51.1) 44 (31.7)
129 (87.2) 44 (29.7) 74 (54.0) 51 (37.0) 31 (34.1) 52 (38.8) 54 (41.2) 55 (40.1)
257 (87.1) 75 (25.4) 139 (50.4) 77 (27.6) 46 (24.9) 111 (40.7) 128 (46.2) 99 (35.9)
NS NS NS P < 0.001 P < 0.05 NS NS NS
42 (31.1)
82 (62.6)
124 (46.6)
P < 0.001
90 (63.8) 29 (20.6) 69 (46.9) 133 (90.5) 59 (41.5)
113 (82.5) 70 (51.5) 101 (68.2) 101 (68.2) 81 (56.3)
203 (73.0) 99 (35.7) 170 (57.6) 234 (79.3) 140 (49.0)
P < 0.001 P < 0.001 P < 0.001 P < 0.001 P < 0.05
117 (81.3) 26 (18.1) 31 (21.5) 39 (27.1) 77 (53.5) 1 (0.7) 40 (28.8)
127 (92.7) 22 (16.1) 30 (21.9) 44 (32.1) 90 (65.7) 7 (5.1) 39 (27.7)
244 (86.8) 48 (17.1) 61 (21.7) 83 (29.5) 167 (59.4) 8 (2.8) 79 (28.2)
P < 0.05 NS NS NS NS NS NS
20 (14.3) 13 (9.3)
26 (18.3) 16 (11.3)
46 (16.3) 29 (10.3)
NS NS
* Socio-Economic Index For Areas (SEIFA) score in quintile 1 (high disadvantage) or 2 (mid-disadvantage). † Excludes n = 9 young people continuously in custody over the 18-month follow-up period. ADHD, attention deficit hyperactivity disorder; K10, Kessler 10; NS, not significant; SE, socio-economic; TAFE, Technical and Further Education.
cohort, stratified by Aboriginality. After adjustment for variables found to be significant in the crude odds ratio, significant markers of possible ID in the overall group included first incarceration before 14 years of age [adjusted odds ratio (AOR) 5.28, 95% confidence interval (CI) 1.33–29.93, P < 0.05], and any psychotic disorder at baseline (AOR 9.43, 95% CI 1.94–45.84, P < 0.001). After stratification, these markers of possible ID were only significant for Aboriginal young people. Aboriginal young people first incarcerated before 14 years of age were almost 10 times more likely to have a possible ID compared with Aboriginal young people first incarcer-
ated at a later age (AOR 9.59, 95% CI 1.39–66.24, P < 0.05). Aboriginal young people diagnosed with a psychotic disorder at baseline were 20 times more likely to have a possible ID compared with Aboriginal young people with no psychosis (AOR 20.48, 95% CI 2.69–155.69, P < 0.001).
Discussion This study has shown that almost one half of all young people in custody in NSW have borderline or lower intellectual functioning (by IQ assessment),
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 58 part 11 november 2014
Journal of Intellectual Disability Research 1010 L. Haysom et al. • ID in young people in custody
Table 2 Cognitive assessments (IQ) by Aboriginality
Intellectual Quotient (IQ) Full Scale IQ Extremely low (
