The combined effects of diet quality and physical activity on maintenance of muscle strength among diabetic older adults from the NuAge cohort.

Diabetic older adults are at a higher risk of muscle strength (MS) decline than their non-diabetic counterparts. Adequate protein and energy intakes and physical activity (PA) may preserve MS during aging. However, the role of diet quality (DQ) in MS maintenance is still unknown. This study aimed to determine the association between DQ - alone or combined with PA - and changes in MS over 3 years in diabetic participants aged 67 to 84 years at recruitment in a secondary analysis of the longitudinal observational NuAge study. Changes in handgrip, knee extensor and elbow flexor strengths were calculated as the difference between recruitment (T1) and after 3 years (T4) in 156 diabetic older adults. Baseline DQ was calculated from 3 non-consecutive 24-hour dietary recalls collected at T1 using the validated Canadian Healthy Eating Index (C-HEI). Change in PA was calculated from Physical Activity Scale for the Elderly (PASE) as PASE T4-PASE T1. Four combinations of variables were created: C-HEI<70 with PASE change either < or > median and C-HEI ≥ 70 with PASE change either < or > median. The association between these four categories and MS maintenance was evaluated using General Linear Modeling (GLM). Analyses were stratified by sex and controlled for covariates. Baseline DQ alone was not associated with MS maintenance. Baseline DQ combined with PASE change showed associations with crude and baseline adjusted handgrip strength (p=0.031, p=0.018) and crude and baseline adjusted elbow flexor change (p=0.028, p=0.017) in males only; no significant results were found for knee extensor strength in either males or females. While findings for females were inconclusive, results demonstrate that better adherence to dietary guidelines combined with a more active lifestyle may prevent MS decline among diabetic older males. Additional research is needed on a larger sample since generalization of these results is limited by the small sample size.