Conveyance and non-conveyance to the emergency department after self-harm: Prevalence and ambulance service staff perspectives.

Relatively little is known about people who self-harm and are not conveyed to the emergency department, or the experiences of ambulance service staff in working with people who self-harm and the conveyance decisions that they make. This research, with two linked studies, was conducted using a sequential mixed methods design. In Study 1, quantitative data was collected about episodes of self-harm that received an ambulance crew decision following a 999 call made in the Yorkshire region. The data collected included details of the episode of self-harm, demographic information, the care provided by ambulance staff, clinical outcomes (including conveyance rates), and explanations for care and conveyance decisions. In the sample there was a proportion of conveyance of 87% and only 13% non-conveyance. Method of self-harm was related to conveyance, with people who had cut themselves significantly less likely to be conveyed than those using other methods. Non-conveyance was associated with a longer duration of ambulance visit. The findings from Study 1 informed Study 2, which was a qualitative interview-based study with staff from the ambulance service. Six ambulance service staff were interviewed about their experiences of working with people who have self-harmed and about the decision-making around non-conveyance. There were six major themes identified using thematic analysis and the themes were presented as if they were clinicians’ thoughts to demonstrate the decision-making process around whether or not to convey the person who has self-harmed: ‘I’ll do my best to help’ but ‘I worry about getting it wrong’ because ‘I’m not sure what I’m doing’ and ‘I’m not supported’ so ‘It’s more than your job’s worth’, which contributes to an overall ‘conveyance culture’. There are a number of recommendations for future research and improving clinical practice, and the results are presented in relation to existing literature.