Trajectories of Objectively Measured Physical Activity in Free-Living Older Men

BACKGROUND: The steep decline in physical activity (PA) among the oldest old is not well understood; there is little information about the patterns of change in PA and sedentary behaviour (SB) in older people. Longitudinal data on objectively measured PA data can give insights about how PA and SB change with age. METHODS: Men age 70–90 yr, from a United Kingdom population-based cohort wore a GT3X accelerometer over the hip annually on up to three occasions (56%, 50%, and 51% response rates) spanning 2 yr. Multilevel models were used to estimate change in activity. Men were grouped according to achieving ≥150 min·wk(−1) of MVPA in bouts of ≥10 min (current guidelines) at two or three time points. RESULTS: A total of 1419 ambulatory men had ≥600 min wear time on ≥3 d at ≥2 time points. At baseline, men took 4806 steps per day and spent 72.5% of their day in SB, 23.1% in light PA, and 4.1% in moderate-to-vigorous PA (MVPA). Mean change per year was −341 steps, +1.1% SB, −0.7% light PA, and −0.4% MVPA each day (all P < 0.001). A total of 76.3% (n = 1083) never met guidelines (“stable low”), 7.9% (n = 112) consistently met guidelines (“stable high”), 8.2% (n = 116) stopped meeting guidelines by the last occasion (“decreasers”), and 4.9% (n = 69) started meeting guidelines by the last occasion (“increasers”). “Decreasers” spent 69.3% of each day in SB at baseline, which increased by 2% per year (P < 0.005), light activity remained at 23.3% (change, −0.2% per year; P = 0.4), and total MVPA decreased from 7.1% by −1.7% per year, (P < 0.001). The number of sedentary bouts >30 min increased from 5.1 by 0.1 per year (P = 0.02). CONCLUSIONS: Among older adults, the steep decline in total PA occurred because of reductions in MVPA, while light PA is relatively spared and sedentary time and long sedentary bouts increase.