Background:Self-harm is a major public health concern and is a leading cause of death from injury. Reaching participants for self-harm research raises a number of challenges, however an opportunity exists in the use of both the internet for data collection and in the use of routinely collected healthcare data.Aims and objectives:The aim of this project was to explore the potential of both online and routinely collected healthcare data for self-harm research and the way in which these data sources can be brought together.Methods:This thesis represents a series of projects exploring the use of various data sources for self-harm research. The first was the development and piloting of an online platform (SHARE UK) for self-harm research. This website incorporated multiple functions: hosting questionnaires; sign-up for a research register; sign-up for linkage with routinely collected data and uploads to a media databank. Next a national survey was conducted to explore young people’s perspectives on the use of both online and healthcare data for self-harm research. Lastly a population level electronic health record cohort study analysing trends over time and contacts across healthcare services was conducted.Results:Participants engaged well with research online: 498 participants signed up to the SHARE UK platform; of whom 85% signed up for the research register. Sixty-two participants uploaded 95 items to the media databank. Alternative formats are discussed. Only 15% of participants consented for linkage with healthcare data. A total of 2,733 young people aged 10-24 who self-harm completed the national survey. Results demonstrated that the necessity for participants to give their address for linkage poses a significant barrier. Opinions around the use of Big Data, encompassing social media, marketing and health data are explored.A total of 937,697 individuals aged 10-24 provided 5,269,794 person years of data from 01.01.2003 to 20.09.2015 to the electronic health record cohort study. Self-harm incidence was highest in primary care. Males preferentially present to emergency departments. Male are less likely than females to be admitted following attendance. This difference persists in the youngest age groups and for self-poisoning. Analysis supports the importance of non-specialist services.Conclusions:This thesis has explored both online and routinely collected healthcare data and their utility for self-harm research, exploring participant views and issues via a national survey. An online platform for self-harm research was successfully piloted and issues identified. This series of projects explores possibilities for future self-harm research. The use of multiple data sources allows research to represent both those in the community and those presenting to healthcare settings, lowering many of the barriers to participating in self-harm research. The future utility of the SHARE UK platform through its collaboration with the Adolescent Mental Health Data Platform (ADP) is discussed. Results of this series of projects will be used to inform the development of this platform with lessons learnt from the pilot addressed and findings from both the national survey and the electronic health record cohort study informing and shaping future research.
Stratifying risk for dementia onset using large‐scale electronic health record data: A retrospective cohort study
