Abstract
BACKGROUND: Despite evidence of associations between glucocorticoid treatment and adverse psychiatric and suicidal behaviour outcomes, large-scale observational evidence for serious outcomes is lacking. AIMS: To assess the risk of psychiatric and suicidal behaviour outcomes during glucocorticoid treatment. METHOD: Using Swedish population registers, we identified 1 105 964 individuals aged 15-54 years who collected a glucocorticoid prescription in oral form between 2006 and 2020. We investigated associations with a range of psychiatric outcomes: unplanned specialist healthcare contacts due to depressive, bipolar, anxiety or schizophrenia-spectrum disorders; and deaths by suicide or unplanned specialist healthcare contacts due to self-harm ('suicidal behaviour'). We estimated hazard ratios from Cox proportional hazards models in a medication-only cohort by comparing outcome rates during and outside treated periods within individuals. We further identified individuals with an autoimmune or gastrointestinal autoimmune disorder diagnosis and compared hazards of the outcomes between those who did and did not initiate a glucocorticoid using a target trial emulation approach. RESULTS: We found increased risks for psychiatric outcomes, with within-individual hazard ratios ranging from 1.08 (95% CI, 1.00-1.16) for depressive disorders to 1.23 (95% CI, 1.12-1.36) for bipolar disorder and 1.25 (95% CI, 1.20-1.31) for anxiety disorders. We found no clear association with suicidal behaviour (hazard ratio: 1.06; 95% CI, 0.96-1.17). These findings were similar when stratified by age and gender. Within-individual associations were attenuated in those diagnosed with an autoimmune disorder. The risk of anxiety and bipolar disorder outcomes appeared particularly elevated in the first weeks of treatment. Absolute rates were modestly elevated during treatment, and higher in those with a history of psychiatric disorders. CONCLUSIONS: Glucocorticoid treatment is associated with elevated risks of serious psychiatric outcomes, including the onset and relapse of common psychiatric disorders. Individuals with psychiatric histories may require additional monitoring during glucocorticoid treatment.