Epidemiology

Epidemiology

The below sections describe key epidemiological trends with illustrative figures. More detailed and additional graphs are available the PowerPoint pack of this needs assessment.

Self-harm
Suicide

Data caveats and limitations

Data on self-harm and suicide is collected by multiple services, each with varied methods and purposes, complicating standardisation and comparison. Data on deaths from suicide is generally reliable, informed by comprehensive coroner inquests and local audits. However, local suicide audit data should still be interpreted cautiously, especially in small sub-categories where random variation may influence reliability, limiting the strength of conclusions about trends or differences between groups. In contrast, data on suicide attempts, suicidal thoughts, and self-harm is limited and less accurate.

  • Self-harm data is especially impacted by under-reporting, with UK figures indicating that 60% of adults and 90% of young people who self-harm do not seek medical or psychological support (7). As such, hospital data serves as a proxy for severe self-harm but captures only a proportion of cases.
  • Those presenting to hospital or primary care with self-harm are more likely to attempt suicide or overdose but less likely to engage in non-suicidal self-harm or self-cutting compared to the wider population engaging in self-harm (28). Hence, healthcare data is a better estimate for the prevalence of suicide attempts rather than non-suicidal self-harm.
  • There can be significant variation in how hospitals across the country record self-harm as opposed to non-intentional injuries and routine hospital data does not identify suicidal intent, so self-harm is often conflated with suicide attempts. However, local audits (e.g., psychiatric liaison reviews) can sometimes provide additional context.
  • The absence of consistent ambulance data on self-harm and suicide attempts further limits hospital data estimates as many individuals seen by ambulance services may not proceed to hospital A&E, leading to gaps in data collection.
  • Data from Police is limited to detentions under section 136 of the Mental Health Act. Currently, there is no capacity for call takers to flag self-harm or suicide attempts in all call-out data.

Recognising these limitations is essential for accurate interpretation. For instance, a drop in A&E presentations for self-harm may not indicate a reduction in self-harm itself but may instead reflect fewer people seeking emergency care due to factors like long wait times or changes in reporting practices. As many individuals may not engage with or present to mental health services when in crisis, “community prevalence” of self-harm ideation, self-harm, suicidal ideation and suicide attempts often relies on national surveys like England’s Adult Psychiatric Morbidity Survey (APMS). While APMS uses a nationally representative sample, it may not accurately reflect local prevalence due to differing local risk factors and its infrequent updates, with the latest report from 2014.

A significant gap is local GP records regarding self-harm and suicidal ideation. This would be valuable in trying to assess what proportion of those who self-harm or have suicidal thoughts seek support from their GP, and how this compares to hospital figures. There is also a significant gap in understanding local inequalities and the impact of suicide and self-harm on specific population group, e.g., lack of consistent ethnicity recording. Continued work is needed to ensure ethnicity is coded, e.g., using primary care records to do this.

Additional resources