B3390 - Predictors and patterns of self-harm thoughts and behaviours - 11/10/2019

B number: 
B3390
Principal applicant name: 
Becky Mars | UOB (United Kingdom)
Co-applicants: 
Title of project: 
Predictors and patterns of self-harm thoughts and behaviours
Proposal summary: 

Self-harm in young people is a major problem. As many as 1-in-6 teenagers have self-harmed, but we know little about what happens to them as they get older. We also know little about how much self-harm thoughts and behaviours (SHTB) change from day-to-day, and what factors help to predict this. This project will look at predictors and patterns of SHTB both over long periods of time (from adolescence to adulthood) and over short periods of time (over days/weeks).
Although selfharm is very common in young people, most do not seek help. This makes it difficult to provide support. In this study, I will find out whether young people who self-harm are either (1) not visiting a GP or (2) visiting a GP for other reasons and not telling them about their self-harm. I will also look for factors that will help GPs to better identify young people who have self-harmed.

Impact of research: 
work steam 1 will considerably enhance knowledge of the epidemiology of self-harm and will facilitate effective targeting of early interventions to those who are most likely to show a chronic course. It will also improve understanding of the role of adolescent experiences in shaping future health. The knowledge gained from workstream 2 will be important in the design and targeting of interventions aimed at improving help seeking and detection rates for self-harm. This workstream will also inform the early identification of young people who are likely to engage in serious self-harm in the future; a group known to be at high risk for suicide.
Date proposal received: 
Monday, 7 October, 2019
Date proposal approved: 
Tuesday, 8 October, 2019
Keywords: 
Epidemiology, Mental health, Statistical methods, Cohort studies - attrition, bias, participant engagement, ethics, Development, Methods - e.g. cross cohort analysis, data mining, mendelian randomisation, etc.