Self-harm is complex and can involve varying degrees of suicidal intent. People who have self-harmed, for example by intentionally poisoning or injuring themselves, have particularly high risks of suicide. Furthermore, at least one in six people presenting to an ED following self-harm will have a further self-harm episode within a year. The COVID-19 pandemic has impacted help seeking in primary and secondary care. Population mental health may also have been affected. Therefore, it is important to understand how the pandemic has affected presentation rates and clinical management of self-harm. So far, information has come from people who have chosen to respond to surveys, rather than from the general population. We aim to find out:
how the pandemic has affected presentations rates of self-harm
likelihood of being prescribed medication for mental illness and being referred for mental health treatment by GPs for patients who have self-harmed
risks of further self-harm and suicide
The pandemic has affected people differently so we will examine differences by age group, gender, ethnic group, existing mental or physical illness and social deprivation.
This research will use anonymous health records, linked to information on deaths, providing information broadly representative of the UK. The study will cover the periods before, during and after the Spring 2020 peak of the pandemic, including up to August 2020.
Monthly incidence and event rates of primary care and ED-presenting self-harm and likelihood of GP consultation and the type of care (including referral to mental health services and prescription for psychotropic medication) will be examined, comparing the March to August period in 2010-2019 to the same period in 2020. Predicted values (based on the antecedent period) will be estimated and compared to observed values during the Covid-19 period, with the values compared using a ratio measure (and its 95% CI). The modelling will be conducted using a negative binomial/Poisson type approach. We will examine differences between groups of patients, specifically age group, gender, ethnic group, COVID-19 shielding status, mental illness and area-level social deprivation.
Risks of the following adverse outcomes following primary care-recorded will be compared between the March to August period in 2010-2019 and the same period in 2020: repeat self-harm (presentations to ED or primary care), suicide, external-cause mortality and all-cause mortality. Cox regression models will be used to estimate hazard ratios for repeat self-harm, suicide, external cause mortality and all-cause mortality.
We will also examine clinical management and risks of adverse outcomes separately for ED-presenting self-harm. Using HES A&E and HES APC data, a cohort of people presenting to ED with self-harm, comparing the March to August period in 2010-2019 to the same period in 2020. This cohort will be linked to records in HES OP, CPRD GOLD and Aurum, IMD and ONS Death Registrations. We will only include patient records that are considered to be of an acceptable quality for research. When using GOLD data, we will only include patients who have been registered with a general practice deemed to be up to standard with respect to continuity and standard data completeness criteria.
GP contacts (including face-to-face and telephone consultations), referrals to mental health services, prescriptions for psychotropic medication, psychiatric outpatient appointments attended, repeat self-harm, suicide, external-cause mortality, all-cause mortality.
Darren Ashcroft - Chief Investigator - University of Manchester
Sarah Steeg - Corresponding Applicant - University of Manchester
Brian McMillan - Collaborator - University of Manchester
Emma Nielsen - Collaborator - University of Nottingham
Laszlo Trefan - Collaborator - University of Manchester
Matthew Carr - Collaborator - University of Manchester
Navneet Kapur - Collaborator - University of Manchester
Pearl Mok - Collaborator - University of Manchester
Roger Webb - Collaborator - University of Manchester
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation