Protocol: A two-wave cross-sectional study in England investigating suicidal behaviour and self-harm amongst healthcare workers during the Covid-19 pandemic

Introduction

There have been longstanding concerns regarding an increased risk of suicide amongst healthcare workers. The Covid-19 pandemic has placed an additional burden on staff, yet few studies have investigated the impact of the pandemic on their risk of suicide and self-harm. We aimed to investigate the cumulative incidence, prevalence, and correlates of suicidal ideation, suicide attempts, and non-suicidal self-injury amongst healthcare workers during the Covid-19 pandemic.

Methods and Analysis

NHS Check is an online survey that was distributed to all staff (clinical and non-clinical), students, and volunteers in 18 NHS Trusts across England during the Covid-19 pandemic. Data collected in wave 1 (collected between April 2020 and January 2021) and wave 2 (collected 6 months after wave 1) will be analysed. The full cohort of wave 1 participants will be weighted to represent the age, sex, ethnicity, and roles profile of the workforce at each Trust, and the weighted prevalence and cumulative incidence of suicidal ideation, suicide attempts, and non-suicidal self-injury will be described. Two-level random effects logistic regression models will be used to investigate the relationship between suicidal behaviour and self-harm, and demographic characteristics (age, sex, ethnicity) and workplace factors (concerns regarding access to personal protective equipment, re-deployment status, moral injury, confidence around raising and the management of safety concerns, support by supervisors or managers, satisfaction with standard of care provided). Results will be stratified by role (clinical/non-clinical).

Changes in this protocol compared with the original version

• All variables describing workplace factors will be analysed as binary variables for consistency.

• The responses to two questions on raising, and the management of, safety concerns will be analysed separately rather than combined to minimise loss of information.

• Binary suicide-related outcomes will be used in the logistic regression analyses, where the presence of an outcome represents its occurrence within the previous one (wave 2) or two months (wave 1), specifically. This was previously not specified.