Background : Some people with autistic spectrum disorders (ASD) are detained within hospitals because of the risk of violence, but we know little about the relationship between autism and this risk, nor do we fully understand the differences within this group and the implications for clinical care. There is little information about the most appropriate care-pathway for this population. There is a lack of evidence linking risk of future violence and ASD, and there is evidence that diagnosis alone is not an appropriate predictor of outcome from secure hospitals. It is more likely that comorbid factors, related to neurocognitive functioning and personality explain the relationship between violence and autism. Considering that secure beds are expensive, it is important to clarify the relationships between these variables to help manage risk carefully, target resources correctly, and ensure care pathways are appropriate. We have developed a sub-typology of people with ASD who have been detained in hospital. While these subtypes have face validity, they have not been examined thoroughly for people with ASD detained in hospital either in respect to treatment needs or outcome within the hospital care pathway. Aims: This study has three aims, 1) to further develop a proposed sub-typology for people with autistic spectrum disorders (ASD) detained within hospital, 2) to test the validity of these subtypes, by examining the relationship between these subtypes, clinical data, and neurocognitive variables, and 3) to examine the relationship between these subtypes and patient outcome in order to understand the most appropriate care pathway. Methods : The aims will be investigated within three related Workstreams: 1) Workstream I will involve using focus group work and consensus methods with clinicians and services users to refine our sub-typology. Clinicians will then be asked to rate all their current inpatients with ASD according to the subtypes, 2) Workstream II will involve examining the validity of our subtypes within two sub-streams: i) Workstream IIa will involve collecting data about behaviour, resource use and risk from clinical records and staff members, describing factors related to care-pathways, and then making comparisons between subtypes on a minimum of 150 patients, and ii) Workstream IIb will involve asking 100 of these patients to complete a battery of neurocognitive tests to allow us to test the validity of our subtypes further. It is anticipated that differences between the subtypes can be characterised by key indicators (e.g. aggression and psychopathy) and 3) Workstream III involves those patients from Workstream IIa being followed up in 12 months, allowing us to collect behavioural data over time. This will allow us to examine how outcome differs between our subtypes, and whether the care pathway differs between subtypes. Potential Benefits to Patients : This study has direct implications for hospital care pathways. Importantly, the study will also allow us to examine the differences between patients, describe their care needs more effectively, and design better hospital care pathways. The Autism Act (2009) directed the government to publish a strategy document which was subsequently titled, “Fulfilling and rewarding lives: the strategy for adults with autism in England.” This document states that people with autism should be able to access services and derive benefit from them, while the Equality Act (2010) states that all services need to make reasonable adjustments to enable access to services for people with disabilities. This project will help the NHS to understand patients with ASD who are detained within hospital effectively, consider risk more appropriately and design better in-patient services and care pathways to the community. This will directly benefit patients by minimising the risk of them being in restrictive hospital settings for longer than necessary.