Background Depression is associated with an increased risk of a series of cardiovascular diseases and with increased symptom burden in patients with atrial fibrillation. The aim of this study was to determine the association between depression as well as antidepressant treatment and the risk of incident atrial fibrillation. Design A nationwide register-based study comparing the atrial fibrillation risk in all Danes initiating antidepressant treatment from 2000 to 2013 ( N = 785,254) with that in a 1:5-matched sample from the general population. Methods Cox regression was used to estimate adjusted hazard ratios (aHRs) and associated 95% confidence intervals (95% CIs), both after initiation of treatment and in the month before when patients were assumed to have medically untreated depression. Results Antidepressant treatment was associated with a three-fold higher risk of atrial fibrillation during the first month (aHR = 3.18 (95% CI: 2.98–3.39)). This association gradually attenuated over the following year (aHR = 1.37 (95% CI: 1.31–1.44) 2–6 months after antidepressant therapy initiation, and aHR = 1.11 (95% CI: 1.06–1.16) 6–12 months after). However, the associated atrial fibrillation risk was even higher in the month before starting antidepressant treatment (aHR = 7.65 (95% CI: 7.05–8.30) from 30 to 15 days before, and aHR = 4.29 (95% CI: 3.94–4.67) the last 15 days before). Overall, 0.4% of patients were diagnosed with atrial fibrillation from 30 days before to 30 days after antidepressant treatment. Conclusions Antidepressant users had a substantially increased atrial fibrillation risk, particularly before treatment initiation. Whether this mirrors a causal relation between depression and atrial fibrillation may have large consequences for public health and should be discussed.