Background: Atopic dermatitis is a chronic inflammatory skin disease and inflammation has been implicated in development of other chronic diseases, but few studies have examined the relationship with dementia.

Objectives: This study examines associations of atopic dermatitis (AD) and systemic inflammation in adolescence measured using erythrocyte sedimentation rate (ESR), as well as AD diagnosed in adulthood, with dementia risk.

Methods: We used three Swedish register-based cohorts. Cohort I (N = 795,680) comprised men, born in 1951–1968, who participated in the military conscription examinations with physician-assessed AD and ESR; Cohort II (N = 1,757,600) included men and women, born in 1951–1968; and Cohort III (N = 3,988,783) included all individuals in Sweden, born in 1930–1968. We used Cox regression, estimating hazard ratios (HR), with the follow-up from 50 years of age to dementia diagnosis, date of emigration, death, or 31 December 2018, whichever occurred first. Further, we used a sibling comparison design to adjust for unmeasured confounders shared among siblings.

Results: Cohort I: 1466 dementia events were accrued during follow-up of 7.8 years, with a crude rate of 21.6 [95% confidence interval (CI): 20.6, 22.8] per 100,000 person-years. Cohort II: 3549 dementia events were accrued during follow-up of 7.4 years, with a crude rate of 23.7 (95% CI: 22.9, 24.5) per 100,000 person-years. Cohort III: 120,303 dementia events were accrued during follow-up of 23.7 years, with a crude rate of 180.3 (95% CI: 179.3, 181.3) per 100,000 person-years. In multivariable analysis using Cohort I, there was no association between AD and dementia [HR 0.68 (95% CI 0.32, 1.43)], nor with moderate [HR 0.71 (95% CI: 0.46, 1.10)] or high [HR 1.23 (95% CI: 0.87, 1.75)] ESR. AD was not associated with dementia risk in Cohort II [HR 1.28 (0.97, 1.71)] or Cohort III [HR 1.01 (0.92, 1.11)].

Conclusions: AD was not associated with dementia risk, neither was systemic inflammation measured by ESR in adolescence.