Do Strength, Power, & Body Composition Predict Function in Older Adults? Effects of Strength Training

Abstract

PURPOSE: To determine the extent to which muscle strength, muscle power, and body composition predict physical functioning in older adults at baseline and with strength training (ST). METHODS: Healthy, inactive adults (n = 68) aged > 65 yrs were tested to determine knee extensor (KE) strength (1-RM), KE peak power, and body composition (KE muscle volume, KE intermuscular fat, total body % fat, and total body fat free mass (FFM)) before 22 wk of ST. The first 10 wk consisted of unilateral ST of the KE and the last 12 wk consisted of whole body ST. The 1-RM strength test, peak power, muscle volume and intermuscular fat were assessed before and after the unilateral ST program. Strength, physical functioning, total body % fat and FFM were assessed at baseline and after the whole body ST program. Muscle volume and intermuscular fat were assessed by computed tomography and % fat and FFM were assessed by dual energy x-ray absorptiometry. Physical functioning was assessed by the time required to perform five chair stands, a stair climb, and two walking agility tests. The total time of all tasks was summed and used as an indicator of functional abilities. Linear regression was used to test all predictive models. RESULTS: 1-RM strength, peak force during a power test, FFM, and % body fat were included as the baseline independent variables and physical functioning total time as the baseline dependent variable. Peak power (P=0.317) was included in the analysis despite being non-significant due to its effect on R2. Muscle volume and intermuscular fat were not included in the baseline model because their association with physical functioning was not significant. The baseline model (P < 0.001) had a R2 of 0.424 and an adjusted R2 of 0.377. FFM was the only variable that was significant for change with the ST model (P=0.003). All other variables were not significant and had no impact R2. FFM had an R2 of 0.179 and an adjusted R2 of 0.161. CONCLUSION: The combined associations of strength, peak power, % body fat, and FFM have a moderate predictive power for physical functioning. Changes in these measures with ST fail to predict changes in physical functioning, but FFM by itself accounts for a small but significant portion of the total variance.