We propose undertaking new data collection and analysing existing data to address the following research questions relating to self-harm in ALSPAC: 1.) What are the health and educational outcomes in children reporting self-harm with and without suicidal intent at ages 11 and 16. 2.)What is the influence of a) help-seeking, b) exposure to self-harm in friends and family, c) autobiographical memory, d) personality traits such as impulsiveness e) media/Internet exposure and f) non-suicidal self injury in mediating progression from suicidal thoughts and non-suicidal self-harm to suicide attempts from age 11 to 16 and from age 16 to 18. 3.) What are the differences and similarities between people who attempt suicide and those who carry out acts of NSSI without suicidal intent. 4.)Through the 5% subset of ALSPAC participants linked with the General Practice Research Database (GPRD) and the recent linkage of ALSPAC to Hospital Episode Statistics, determine the reliability of self-report self-harm and investigate the impact of non-response bias on observed associations. We propose undertaking the following analyses: 1.) Prospective analysis to estimate the incidence of self-harm at age 21. 2.) Prospective analysis based on the self-harm questions at age 11, to identify health and educational impacts of self harm at this age and progression of suicidal thoughts and act between ages 11 and 16. 3.) Prospective analysis based on responders to the self-harm questionnaire at age 16/17. We will assess the impact of self-harm reported at age 16/17 on mental health (CIS-R), educational and other social outcomes by age 18-21. We will compare outcomes for those participants who self-harmed with and without suicidal intent. 4.) We will conduct further prospective analysis to identify social, psychological and educational predictors of self-harm at age 16/17 and 21. 5.) We will explore the reliability of self-report self-harm and investigate the impact of non-response bias on observed associations by identifying a) whether episodes of self-harm identified through HES and GPRD linkage were reported by participants, b) rates of self-harmamongst non-responders to clinics/ questionnaires.