126 Development of a Falls Prediction Tool for People with Diabetes Mellitus. (20th December 2019)

• Record Type: Journal Article
• Authors: Wettasinghe, Asha; Dissanayake, Dilshani; Allet, Lara; Katulanda, Prasad; Lord, Stephen
• Is Part Of: Age and ageing. Volume 48(2019)Supplement 4
• Page Start: iv28
• Page End: iv33
• Publication Date: 2019-12-20
• Subjects: Diabetes mellitus -- fall risk assessment -- falls risk factors; Aging -- Periodicals; Geriatrics -- Periodicals; 618.97
• DOI: 10.1093/ageing/afz164.126
• Languages: English
• ISSNs: 0002-0729

Abstract

Abstract: Introduction: Diabetes mellitus (DM), aging and falls have been recognized as a growing and a challenging triad. Despite many tools to assess risk of falls, they may not be applicable for fall risk assessment in DM patients. The aim of the study was to develop a low- cost tool to predict the faller status of DM people. Methods: People with DM (n=103) were recruited from diabetic clinics in Sri Lanka. Demographic, neuropathy status, contrast sensitivity (Melbourne Edge test) lower limb (LL) sensation, cognitive functions, fear of falling (Icon-FES), LL strength (maximal isometric quadriceps strength), hand reaction time, balance abilities (postural sway, maximal balance range, coordinated stability, unipedal stand time (UST) and tandem and near tandem standing ability), mobility (Timed up and go test-TUG) and gait parameters were assessed. Falls were prospectively recorded over six months. Data were analyzed using SPSS and STATA with negative binominal regression. Results: Fall rates were significantly associated with DM symptoms (p=0.001) and examination scores (p=0.026), HbA1c (p=0.012), contrast sensitivity (p=0.033), TUG (p=0.006), quadriceps strength (p=0.015), sway path on floor eyes closed (p=0.002), sway on foam eyes opened (p=0.009), near tandem (p=0.006) and tandem (p=0.008) balance, UST (p=0.027), stride length (p=0.009), TUG times with cognitive task (p=0.007), dual task backward walking (p=0.005). A multivariate negative binomial regression model comprised two significant and independent variables influencing falls: quadriceps strength and tandem balance ability (likelihood ratio Chi square=12.43, p=0.002). For 1kg increase in quadriceps strength fall rates decreased by 0.94% and the presence of poor tandem balance increased fall rates by a factor of 2.48. Discussion and Conclusion: We identified multiple factors that elucidate why people with DM fall. A simple, easy to use tool comprising two independent risk factors: tests of quadriceps strength and tandem balance ability are suggested as a simple screen for fall risk in this population. Such patients may then warrant further detailed assessment and a tailored intervention.