|
Abstract
|
BACKGROUND: Diabetes is associated with gait deficits, future falls and disability, however it is unclear if associations remain after controlling for relevant confounders. This study investigated (i) the effects of type II diabetes on spatiotemporal gait parameters in community-dwelling older adults and (ii) if diabetes status was independently associated with future falls and disability, after controlling for gait and other confounders.
METHODS: Baseline data were obtained from 2,608 community-dwelling adults (≥60 years) participating in The Irish Longitudinal Study on Ageing (TILDA). Diabetes was identified from self-reported doctors' diagnosis, medications and glycated haemoglobin levels. Gait characteristics were obtained during single and dual task walking using a GAITRite® mat (n=2560). Incident falls and disability were collected over four years follow-up (n=2473). Associations between diabetes status and gait (cross-sectional) and falls and disability (longitudinal) were investigated using regression analysis, adjusting for medications, cardiovascular health, neuropsychological function and fall-related factors.
RESULTS: Diabetes (prevalence = 9.1%) was cross-sectionally associated with shorter dual task step length after adjusting for covariates (β=-1.59, 95% CI: -3.10, -0.08, p<0.05). Diabetes was independently associated with increased risk of future IADL difficulty in those with no prior difficulty (IRR=1.51, 95% CI: 1.08 2.11, p<0.05) although dual task step length was an important confounder in all disability models. No independent associations between diabetes and falls were observed.
CONCLUSIONS: Diabetes was independently associated with shorter dual task step length and increased risk of future IADL difficulty. Multidimensional interventions addressing poor health and function in those with diabetes may help reduce the risk of gait deficits and future disability.
Language: en |