Depression is a common and recurring mental health condition, affecting around 4 out of 100 people worldwide. People with moderate to severe depression often require treatment with antidepressants. While using antidepressants consistently has been shown to be effective, stopping their use can be challenging and requires careful planning in clinical practice. When treatment is stopped, there is a high risk of depression returning, with suicidal behavior as one early warning sign of relapse. However, understanding the link between stopping antidepressants and suicidal behavior has been challenging in scientific studies because of the serious consequences involved and the limitations of research methods. Since suicidal behavior is rarely reported in clinical trials and is hard to know accurately from individual case reports, we plan to use data from UK primary care to better understand how antidepressants are used, and to answer specific questions about the connection between stopping antidepressants and suicide. This study aims to find the best strategies for maintaining and stopping antidepressants in order to minimise the risk of suicide. This can help reduce the strain on healthcare systems and, on a larger scale, decrease the risk of suicide as well as the emotional and financial costs it brings to individuals, families, and communities.
We aim to examine the patterns of antidepressant discontinuation and its impact on suicidal behaviour in the UK by utilising electronic health records from the Clinical Practice Research Datalink (CPRD) spanning 2001 to 2023. The study population consists of individuals with depression receiving their first antidepressant prescription during study period. We will characterise the classes and individual drugs of the initial antidepressant treatment, and evaluate treatment retention by calculating the duration from antidepressants treatment initiation to discontinuation using survival analyses; with separate analyses for individual drugs and classes. We will investigate the duration and modes (gradual taper, switching, or abrupt stopping) of the discontinuation of initial antidepressant treatment. Subsequently, we will conduct self-controlled case series (SCCS) and cohort analysis to investigate the association of suicide events with antidepressant discontinuation. The SCCS analysis will only include individuals who experienced both antidepressant discontinuation and suicide events during the study period. The incidence rate ratio (IRR) will be estimated by comparing incidence rates during different risk periods, including the period shortly after antidepressant discontinuation, using conditional Poisson regression, adjusting for time-varying factors such age, concurrent use of other psychotropic medications and seasonal effects in the analysis. In cohort analysis, we will estimate the risk of suicide during the antidepressant discontinuation period by emulating hypothetical target trial using clone-censor-weight approach. We will then conduct several subgroup and sensitivity analyses to examine the robustness of the main analysis.
The outcome of interest will be suicide events, composite of suicide attempt and completed suicide.
Kenneth Man - Chief Investigator - University College London ( UCL )
Sohee Park - Corresponding Applicant - University College London ( UCL )
Adrienne Chan - Collaborator - UCL School Of Pharmacy
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