OBJECTIVES: The effect of a direct-acting oral anticoagulant (DOAC) dose on intracerebral hemorrhage (ICH) severity and outcome remains unclear. The aim of this study is to clarify the frequency of off-label dosing of DOAC treatments in ICH patients and compare clinical characteristics. PATIENTS AND METHODS: We studied 43 patients with ICH who were treated with DOAC for nonvalvular atrial fibrillation before the onset of ICH. DOAC treatments were categorized into three groups based on the following doses: optimal dose, under-dose, and overdose. RESULTS: -VASc scores were the highest in the overdose group (median, 4, 6, respectively) and the lowest in the optimal dose group (median, 2, 4, respectively) (p = 0.006, p = 0.005, respectively). ICH severity measured using the National Institutes of Health Stroke Scale scores was the highest in the overdose group (median, 26.5) and the lowest in the under-dose group (median, 6.5) (p = 0.244). Larger initial hematoma volume was observed in the overdose group. The ratio of good outcome (modified Rankin Scale score ≤ 2) was higher in the under-dose group (40%) than the other groups, but this difference was not significant. CONCLUSION: Our study shows only a few patients received overdosing of a DOAC before the onset of ICH, and they were associated with poorer functional outcomes. Conversely, under-dosing was associated with better functional outcomes than the other groups.