Case‐control co‐twin study on the associations of dementia risk factors with episodic memory in nonagenarians

Abstract

BackgroundSeveral risk and protective factors of dementia are known, but the importance of genetic and environmental influences and the antecedents of cognitive functioning in 90‐year‐olds are not well known. Genetic and environmental confounding can be controlled better by using family data compared to samples of unrelated individuals. We examined dementia risk factors in 90‐year‐olds by conducting a case‐control co‐twin study on the associations of physical and psychological well‐being with episodic memory (EM).MethodWe invited 90‐year‐old twins from a population‐based older Finnish Twin Cohort study to participate in the ongoing NONAGINTA –study that includes questionnaires (risk factors) and telephone interview (cognitive functioning). To date, eight full same‐sex pairs have cognitive data available (6 female, 6 monozygotic). We calculated a risk score for each twin and co‐twin, based on the following factors: depressive symptoms (measured with the Center for Epidemiological Studies – Depression, CES‐D), body mass index (BMI), physical activity (MET hours), life satisfaction, neuroticism, and years of education. Within each twin pair, co‐twins were assigned as having greater versus lower risk score. We ran a paired t‐test to compare EM performance (total number of recalled words in three trials of 10‐word list) in co‐twins with greater versus lower risk score. Associations of individual factors with EM were studied with exact McNemar test.ResultMean difference in EM within twin pairs was 3.38 (SD = 2.50). Pairwise correlation of EM was r = 0.75 and seven twin pairs were discordant for EM. Co‐twins with lower risk score remembered more words (M = 13.25, SD = 5.59) than their co‐twins with greater risk score (M = 12.88, SD = 5.82), but this difference was not significant (p = 0.82). Examining each risk factor individually, the strongest – non‐significant – association was that co‐twins with greater life satisfaction had better EM in 5/6 pairs (p = 0.10).ConclusionPreliminary data did not indicate any significant within‐twin pair associations of dementia risk and episodic memory. As expected, co‐twins were highly similar in EM. Ongoing data collection will result in more twin pairs and greater statistical power to see if the suggestive association of greater life satisfaction (or other factors) with better EM is evident in nonagenarians.