Incidence of intracerebral haemorrhage over the past three decades is reported as stable. This disappointing finding is questionable and suggests that any reduction in intracerebral haemorrhage incidence associated with improvements in primary prevention, namely, better control of blood pressure, might have been offset by an increase in cases of intracerebral haemorrhage owing to other factors, including the use of antithrombotic drugs in the ageing population. Therefore, we aimed to analyse trends in intracerebral haemorrhage incidence from 1985 to 2008 in the population-based registry of Dijon, France, taking into consideration the intracerebral haemorrhage location, the effect of age and the changes in the distribution of risk factors and premorbid treatments. Incidence rates were calculated and temporal trends were analysed by age groups (<60, 60-74 and ≥75 years) and intracerebral haemorrhage location (lobar or deep) according to study periods 1985-92, 1993-2000 and 2001-08. Over the 24 years of the study, 3948 patients with first-ever stroke were recorded. Among these, 441 had intracerebral haemorrhage (48.3% male), including 49% lobar, 37% deep, 9% infratentorial and 5% of undetermined location. Mean age at onset increased from 67.3 ± 15.9 years to 74.7 ± 16.7 years over the study period (P < 0.001). Overall crude incidence was 12.4/100,000/year (95% confidence interval: 11.2-13.6) and remained stable over time. However, an ∼80% increase in intracerebral haemorrhage incidence among people aged ≥75 years was observed between the first and both second and third study periods, contrasting with a 50% decrease in that in individuals aged <60 years, and stable incidence in those aged 60-74 years. This result was attributed to a 2-fold increase in lobar intracerebral haemorrhage in the elderly, concomitantly with an observed rise in the premorbid use of antithrombotics at this age, whatever the intracerebral haemorrhage location considered. In conclusion, intracerebral haemorrhage profiles have changed in the past 20 years, suggesting that some bleeding-prone vasculopathies in the elderly are more likely to bleed when antithrombotic drugs are used, as illustrated by the rise in the incidence of lobar intracerebral haemorrhage in the elderly, in which cerebral amyloid angiopathy may be strongly implicated. Future research should focus on the impact and management of antithrombotics in patients with intracerebral haemorrhage, which may differ according to the underlying vessel disease.