• Louise Robinson, Michael D Spencer, Lindsay D G Thomson, Andrew C Stanfield, David G C Owens, Jeremy Hall, and Eve C Johnstone.
• Division of Psychiatry, School of Molecular and Clinical Medicine, University of Edinburgh, Edinburgh, United Kingdom. Louise.Robinson@ed.ac.uk
• Plos One. 2012 Jan 1; 7 (5): e36078.

UnlabelledThere have been concerns that individuals with autism spectrum disorders (ASDs) are over-represented but not recognised in prison populations. A screening tool for ASDs in prisons has therefore been developed.AimsWe aimed to evaluate this tool in Scottish prisoners by comparing scores with standard measures of autistic traits (Autism Quotient (AQ)), neurodevelopmental history (Asperger Syndrome (and High-Functioning Autism) Diagnostic Interview (ASDI)), and social cognition (Ekman 60 Faces test).MethodsPrison officers across all 12 publicly-run closed prisons in Scotland assessed convicted prisoners using the screening tool. This sample included male and female prisoners and both adult and young offenders. Prisoners with high scores, along with an equal number of age and sex-matched controls, were invited to take part in interviews. Prisoners' relatives were contacted to complete a neurodevelopmental assessment.Results2458 prisoners were screened using the tool, and 4% scored above the cut-off. 126 prisoners were further assessed using standardised measures. 7 of those 126 assessed scored 32 or above (cut-off) on the AQ. 44 interviews were completed with prisoners' relatives, no prisoner reached the cut-off score on the ASDI. Scores on the screening tool correlated significantly with AQ and ASDI scores, and not with the Ekman 60 Faces Test or IQ. Sensitivity was 28.6% and specificity 75.6%; AUC was 59.6%.ConclusionsAlthough this screening tool measures autistic traits in this population, sensitivity for scores of 32 or above on the AQ is poor. We consider that this limits its usefulness and do not recommend that the tool is routinely used to screen for ASDs in prisons.

21 keywords...

hide…