sedentary time is high in this Icelandic cohort, which has high life-expectancy and is living north of 60° northern latitude.

Background Sedentary behavior is associated with adverse health effects. To prevent sedentary behavior and limit health risks, insights into associated determinants are essential. Sedentary behavior should be viewed as a distinct health behavior, therefore its determinants should be independently identified. Purpose This study examines the prospective associations between a wide-range of midlife determinants and objectively measured sedentary time in old age. Methods Data from 565 participants (aged 73–92 years) of the AGESII-Reykjavik Study were used. Participants wore an accelerometer (ActiGraph GT3X) on the right hip for 7 consecutive days. On average 31 years earlier (during midlife) demographic, socioeconomic, lifestyle and biomedical factors were collected. Linear regression models were used to examine prospective associations between midlife determinants and sedentary time (<100 counts per minute) in old age. Results After adjustment for sex, age, follow-up time, minutes of moderate to vigorous physical activity, BMI, health status, mobility limitation and joint pain in old age, the midlife determinants not being married, primary education, living in a duplex or living in an apartment (vs. villa), being obese and having a heart disease were associated with, respectively, on average 15.3, 12.4, 13.5, 13.3, 21.8, 38.9 sedentary minutes more per day in old age. Conclusions This study shows that demographic, socioeconomic and biomedical determinants in midlife were associated with considerably more sedentary time per day in old age. These results can indicate the possibility of predicting sedentariness in old age, which could be used to identify target groups for prevention programs reducing sedentary time in older adults.

Many studies have examined the hypothesis that greater participation in physical activity (PA) is associated with less brain atrophy. Here we examine, in a sub-sample (n = 352, mean age 79.1 years) of the Age, Gene/Environment Susceptibility-Reykjavik Study cohort, the association of the baseline and 5-year change in magnetic resonance imaging (MRI)-derived volumes of gray matter (GM) and white matter (WM) to active and sedentary behavior (SB) measured at the end of the 5-year period by a hip-worn accelerometer for seven consecutive days. More GM (β = 0.11; p = 0.044) and WM (β = 0.11; p = 0.030) at baseline was associated with more total physical activity (TPA). Also, when adjusting for baseline values, the 5-year change in GM (β = 0.14; p = 0.0037) and WM (β = 0.11; p = 0.030) was associated with TPA. The 5-year change in WM was associated with SB (β= −0.11; p = 0.0007). These data suggest that objectively measured PA and SB late in life are associated with current and prior cross-sectional measures of brain atrophy, and that change over time is associated with PA and SB in expected directions.