| dc.description.abstract | Purpose: To assess the association between muscle strength and risk of developing Alzheimer’s disease (AD) among older adults aged 70-77 years in Norway.
Methods: Grip strength (GS) and muscle mass of 1031 apparently healthy older adults (48% men, age of 72.2±1.9 years old) who participated in the randomized controlled trial, Generation 100 Study was measured at baseline and after one year of exercise intervention. GS was measured using a hydraulic hand dynamometer and the lean muscle mass of the dominant arm was obtained using bioelectrical impedance. To identify AD cases over the next 9 years follow-up, hospital records were reviewed by a neurologist. To assess the hazard ratio (HR) at both baseline and year-one we used continuous variables adjusting for sex and level of education. To assess the hazard ratio according to the change in GS we categorized them into, unfit to unfit (below mean at both baseline and year-one and we used that as the reference group), unfit to fit (below mean at baseline and above mean at year-one), fit to unfit (above mean at baseline and below mean at year-one), and fit to fit (above mean at both baseline and year-one).
Result: The mean GS was 21.6% (p<0.001) higher in men (18.5±3.1 kg-0.67) than in women (15.2±2.6 kg-0.67). Over a mean follow-up period of 7.7±0.8 years, 26 participants (2.5%, 14 men) developed AD. Those who developed AD had a mean follow-up of 3.2±1.7 years and had decreased GS by -0.4±1.3 kg-0.67 (p=0.08) over one year of intervention, not different from those who remained healthy till the end of follow-up (-0.1±2.0 kg-0.67, p=0.11). The GS from baseline to year-one for women who later developed AD decreased by 4% (-0.7±0.8 kg-0.67, p<0.01) compared to a 2% decrease (-0.3±1.8 kg-0.67, p<0.001) among women who did not develop AD during follow-up. Changes in GS for men who later developed AD and those who did not develop AD were -0.2±1.6 kg-0.67, p<0.59 and +0.1±2.2 kg-0.67, p<0.07. There was no significant association between GS at baseline or at year-one and risk of developing AD (HR 1.05, 95% confidence interval (CI) of 0.91–1.19, and HR 0.99, 95% CI 0.87–1.13, respectively). There was no significant association between change in GS from baseline to year-one and the incidence of AD. Compared with those who remained unfit at both timepoints, changing from unfit to fit (HR 0.36, 95% CI 0.04-2.88) or remaining fit at both timepoints showed no reduction in risk of AD (HR 1.14, 95% CI 0.62-3.18).
Conclusion: We observed no associations between GS and AD either at baseline or according to change in GS during one year of intervention and individuals with higher GS did not have a lower risk of developing AD. | |
