To determine the risk of dementia in patients with type 1 or type 2 diabetes and in individuals with glycosylated haemoglobin, type A1C (HbA1c) of ⩾48 mmol/mol, which is the diagnostic limit for diabetes.

We included the following cohorts: all incident diabetes cases aged 15 or above registered in the National Diabetes Registry (NDR) from January 2000 through December 2012 (n = 148 036) and a reference population, adult participants from the Glostrup cohort (n = 16 801), the ADDITION Study (n = 26 586) and Copenhagen Aging and Midlife Biobank (CAMB) (n = 5408). Using these cohorts, we analysed if a diagnosis of type 1 or type 2 diabetes in the NDR or HbA1c level of ⩾ 6.5% (48 mmol/mol) in the cohorts increased risk of dementia in the Danish National Patient Registry or cognitive performance assessed by the Intelligenz-Struktur-Test 2000R (IST2000R).

A diagnosis of type 1 or type 2 diabetes in the NDR was associated with increased risk of dementia diagnosed both before or after age 65 as well as across different subtypes of dementia. Self-reported diabetes or high HbA1c levels were associated with lower cognitive performance (p = 0.004), while high HbA1c was associated with increased risk of dementia (HR 1.94 (1.10–3.44) in the Glostrup cohort but not in the ADDITION Study (HR 0.96 (0.57–1.61)).

Both type 1 and type 2 diabetes are associated with an increased risk of dementia, while the importance of screening-detected elevated HbA1c remains less clear.