Initiation of antidepressant medicines in Swedish patients diagnosed with COVID-19: a registry-based cohort study

Soc Psychiatry Psychiatr Epidemiol. 2026 Jul 17. doi: 10.1007/s00127-026-03169-2. Online ahead of print.

ABSTRACT

PURPOSE: While antidepressant therapy is a common treatment for post-COVID mental health symptoms, little is known about initiations of antidepressants in patients after COVID-19 and how sociodemographic factors modify this risk. This study investigated whether adults infected by COVID-19 were more likely to initiate antidepressant therapy compared with non-infected individuals, and examined the influence of socioeconomic factors on initiation patterns.

METHODS: We conducted a nationwide registry-based cohort study using data from the SCIFI-PEARL project, including all adults (≥ 18 years of age) in Sweden from 1 March 2020 to 31 March 2022. Exposed individuals had a COVID-19 infection, and unexposed individuals were matched 1:1 by sex and birth year. Stratified Cox regression models were used to estimate the risk of antidepressant (ATC: N06A) initiation, adjusting for socioeconomic status, and Charlson Comorbidity Index.

RESULTS: The cohort included 2,666,510 individuals (1,333,255 exposed and unexposed). Exposed individuals had a 16% higher risk of initiating antidepressants compared with unexposed individuals after full adjustment (HR: 1.16, 95% CI:1.15-1.17). The risk reduced by roughly 43% when we adjusted for post-index healthcare contacts as potential mediator. Higher risks were observed for unmarried (HR: 1.14, 95% CI:1.13-1.16), or unemployed/retired (HR 1.14, 95% CI: 1.13-1.16) individuals. Conversely, higher educational level and being born outside Sweden were associated with lower risk of initiation of antidepressant, (HR:0.70, 95% CI:0.68-0.71) and (HR:0.92, 95% CI: 0.91-0.94), respectively.

CONCLUSIONS: COVID-19 exposure was associated with an increased risk of initiating antidepressant therapy, particularly among socioeconomically vulnerable groups. Differences by education and immigrant status suggest disparities in access to healthcare.

PMID:42467082 | DOI:10.1007/s00127-026-03169-2