Mortality and adverse events associated with statin use in primary care patients with depression: a real-world, population-based cohort study

Background New National Institute for Health and Care Excellence (NICE) guidance endorses the prescription of statins in larger population groups for the prevention of cardiovascular and cerebrovascular morbidity and mortality, especially in people with severe mental illness. However, the evidence base for their safety and risk/benefit balance in depression is not established. Objectives This study aims to assess the real-world mortality and adverse events of statins in depressive disorders. Methods Population-based, nationwide (England), between-subject, cohort study. We used electronic health records (QResearch database) of people aged 18–100 years with first-episode depression, registered with English primary care practices over January 1998–August 2020 for 12(+) months, divided into statin users versus non-users. Primary safety outcomes included all-cause mortality and any adverse event measured at 2, 6 and 12 months. Multivariable logistic regression was employed to control for several potential confounders and calculate adjusted ORs (aORs) with 99% CIs. Findings From over 1 050 105 patients with depression (42.64% males, mean age 43.23±18.32 years), 21 384 (2.04%) died, while 707 111 (67.34%) experienced at least one adverse event during the 12-month follow-up. Statin use was associated with lower mortality over 12 months (range aOR2–12months 0.66–0.67, range 99% CI 0.60 to 0.73) and with lower adverse events over 6 months (range aOR2–6months 0.90–0.96, range 99% CI 0.91 to 0.99), but not at 1 year (aOR12months 0.99, 99% CI 0.96 to 1.03). No association with any other individual outcome measure (ie, any other neuropsychiatric symptoms) was identified. Conclusions We found no evidence that statin use among people with depression increases mortality or other adverse events. Clinical implications Our findings support the safety of updated NICE guidelines for prescribing statins in people with depressive disorders.


Analysis
Baseline characteristics of the study cohort were reported with descriptive statistics.Outcomes of interest were explored with multivariable logistic regression models, clustered by general practices, to compute odds ratios (ORs) with 99% confidence intervals (99%CIs) for statin users vs statin non-users (i.e., between-subject design), adjusted (aOR) for the above potential confounders.Multiple imputation by chained equations was employed for missing data: for each imputation, 10 imputed datasets were generated, and coefficient estimates across these were pooled using Rubin's rule (16) to calculate results for the primary full-set analysis (FSA, including imputed data).As a sensitivity analysis, we reported results for the complete-case analysis (CCA), which does not include imputed data and only accounts for cases with no missing variables.Further, FSA data for statins were compared with those of the regression analyses for aspirin to probe whether results were non-specifically associated with another medication with comparable indication (i.e., prevention and treatment of CVD) in a similar population.All statistical analyses were conducted on Stata MP v16.0 (17).
Patient and public involvement and engagement People with lived experience of unipolar depression were recruited from across the UK by the Oxford Precision Psychiatry Lab of the Oxford Health Biomedical Research Centre (https://oxfordhealthbrc.nihr.ac.uk/our-work/oxppl/patient-involvement-and-ethics/) and were involved in the design of the study (1).The drafted manuscript was critically revised by two people with lived experience and their feedback was incorporated in the final article.

CHANGES TO THE PROTOCOL
Compared to the original protocol, the current study focusses on the adverse outcomes and reports the outcome "depressive episode remission" in this context, while the other efficacy outcomes (i.e., change in depression score, response) are not described as beyond the purpose of this investigation.In any case, these two efficacy outcomes show overlapping results with the outcome "depressive episode remission" i.e., no effect.

S2. "Read codes" for depressive disorders
"Read codes" are a coded thesaurus of clinical terms used in English primary care.For depression, the relevant "Read code" is 3829.

PHQ-9
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) cur epi severe with psyc symp Recurrent depressive disorder Recurrent depressive disorder, current episode mild Recurrent depressive disorder, current episode moderate Recurrent depressive disorder, unspecified Recurrent episodes of depressive reaction Recurrent episodes of psychogenic depression Recurrent episodes of reactive depression Recurrent severe episodes of psychotic depression Recurrent severe episodes/reactive depressive psychosis Severe depressive episode with psychotic symptoms Severe depressive episode without psychotic symptoms Single episode agitated depressn w'out psychotic symptoms Single episode major depression w'out psychotic symptoms Single episode of depressive reaction Single episode of major depression and psychotic symptoms Single episode of psychogenic depression Single episode of psychogenic depressive psychosis Single episode of psychotic depression Single episode of reactive depression Single episode of reactive depressive psychosis Other Read codes for depression 4 item geriatric depression scale Acute posttrauma stress state Acute stress reaction NOS Assessment using Whooley depression screen Beck depression inventory Beck depression inventory second edition score Depression -enhanced service completed Depressionenhanced services administration Depression annual review Depression anxiety stress scales anxiety score Depression anxiety stress scales depression score Depression anxiety stress scales stress score BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any relianceSupplemental material placed on this supplemental material which has been supplied by the author(s)

Table S8a .
FSA: unadjusted and adjusted odds ratios (statin users vs statin non-users) for adverse events (any adverse event, any psychiatric symptom, anxiety, sleep disturbance, memory impairment, self-harm, suicidality, completed suicide, and all-cause mortality) at 2 months, 6 months, and 12 months For depressive episode remission an aOR>1 favours statin users, while for all outcomes an aOR<1 favours statin users.*Analyses adjusted for age; sex; BMI; type, severity, and year of diagnosis of depression; deprivation status; geographical area; smoking habits; alcohol use; ethnicity; comorbidities and concomitant medications use. Legend.

Table S8b .
CCA: unadjusted and adjusted odds ratios (statin users vs statin non-users) for adverse events (any adverse event, any psychiatric symptom, anxiety, sleep disturbance, memory impairment, self-harm, suicidality, completed suicide, and all-cause mortality) at 2 months, 6 months, and 12 months BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ