Temporal incidence and impact of dementia in rheumatoid arthritis: a cohort study

The objective of this study is to address conflicting evidence that chronic inflammation may increase the risk of dementia in patients with rheumatoid arthritis (RA).

Retrospective population-based study using longitudinally linked administrative health data over a 30-year period for ever hospitalised patients with RA (n = 14,041, age 64 years, 67.2% female) and controls (n = 33,785, age 65 years, 65.6% female). Dementia was defined by the International Classification of Diseases codes for Alzheimer's disease (AD), vascular dementia, and nonspecific dementia subtypes. Dementia incidence rate (IR) and mortality rate (MR) per 1000 person-years and comorbidities are reported.

During 9.6 years of follow-up, 1463 (10.4%) of patients with RA and 3701 (11%) of controls were diagnosed with dementia at respective age of 83 vs 84 years (P = .01). The IR was 12.07 (95% CI: 11.15-12.71) in patients with RA and 11.59 (95% CI: 11.22-11.97) in controls corresponding to an IR ratio of 1.04 (95% CI: 0.98-1.11, P = .21), which did not change significantly over 3 decades. Traditional risk factors for dementia were equal in both groups, but patients with RA with dementia were less likely to be classified as AD (odds ratio = 0.59, 95% CI: 0.48-0.73, P < .001). Hospitalisation rates after dementia diagnosis were higher for patients with RA, and the crude MR (overall 91.8 vs 90.4, P = .64) remained similar before or after 2000 for both groups.

There was no difference in temporal incidence and MRs for dementia between patients with RA and matched controls. These data suggest that there is minimal impact of RA on the frequency and outcome of dementia.