Abstract
This study assesses the extent to which the association between erythrocyte sedimentation rate, a marker of inflammation, and cognitive function is explained by shared familial factors using within-sibling analyses. Men who were born in Sweden between 1950 and 1965 and recorded in the Swedish Military Conscription Register between 1969 and 1983 were included (N = 632,396). Erythrocyte sedimentation rate and cognitive function were measured at the conscription assessment (median age = 18.3 years, with a range from 15.5 to 28.5 years). Conventional linear regression and multilevel linear regression with a hybrid modeling approach were used, with the latter to obtain within-effect estimation in which unmeasured familial confounding shared by siblings was controlled for. We found that the association between erythrocyte sedimentation rate and cognitive function at conscription assessment was partly accounted for by, but remained independent of, shared familial factors.