The effect of risk factors on cognitive impairment: a life table approach

Mieke Reuser, Netherlands Interdisciplinary Demographic Institute (NIDI)
Luc Bonneux, Netherlands Interdisciplinary Demographic Institute (NIDI)
Frans Willekens, Netherlands Interdisciplinary Demographic Institute (NIDI)

The prevalence of dementia increases exponentially with age and is estimated an impressive 40.7% for people aged 90 and over. Ageing and life extension of the baby boom cohorts will cause numbers of demented people to increase rapidly. Smoking, increased BMI and low educational attainment may increase the incidence of dementia, although the results are conflicting. We use multi-state life tables to estimate the incidence and duration of cognitive disability, dependent on these risk factors. Age specific transition probabilities are from a selection of white non-Hispanics from the longitudinal United States Health and Retirement Survey (HRS), 1992-2004. As risk factors we select self-reported BMI, smoking and education. Cognitive functioning is covered by a modified version of the TICS (Telephone Interview Cognitive Screen), itself being a modified version of the MMSE (Mini Mental State Examination). Comparable to other studies, we define a score of 8 or less as cognitively impaired. Lifetime probability of cognitive impairment at age 55 is then 19% for males and 28% for females. BMI and smoking status make no significant difference in life years lived with cognitive impairment, but educational attainment does. A Cox model, controlling for BMI and smoking, shows proportional hazard ratios of 3.5 [3.1:3.9] and 2.2 [1.9:2.6] for lowly educated men and women compared to the highly educated. Highly educated men at age 55 may expect to live on average 0.4 [0.2:0.6] years with cognitive impairment while lowly educated men can expect 2.3 [1.6:3.0] years with cognitive impairment. Women live longer with cognitive impairment, a consequence of their higher life expectancy: 1.4 [0.7:2.0] for the higher educated and 2.8 [2.2:3.4] years for the lower. Higher educated persons may have a greater cognitive reserve that can postpone dementia. Better education may alleviate the expected increase of dementia.