Abstract
Improving physical activity is crucial for management of type 2 diabetes (diabetes) particularly in older persons. Poor balance confidence is a common concern in older population that can lead to restriction of physical activities. We identified differences in physical activity levels between older persons with and without diabetes and examined low balance confidence as a potential determinant of reduced physical activity levels in those with diabetes. 60 persons aged ≥65 years were recruited; 46 with diabetes for ≥5 years and 14 without diabetes. Physical activity level was measured using the Human Activity Profile Adjusted Activity Score (HAP-AAS) that takes into consideration the activities that the person has stopped performing. Balance confidence was measured by the Activity Specific Balance Confidence (ABC) scale. Potential covariates to be included in the multiple regression analysis were selected using bivariate correlation analysis of demographics, knee extensor strength, ankle proprioception, fall history in the previous year, balance performance and depression with HAP-AAS scores at p<0.05. Age and BMI were significantly different between those with and without diabetes. ANCOVA adjusted for age and BMI demonstrated HAP-AAS scores were significantly lower in the elderly with diabetes (p<0.05). The fully adjusted multiple regression model explained 50% variance in HAP-AAS scores and demonstrated ABC scores as the only significant determinant of HAP-AAS scores (p<0.05). Balance confidence is a strong determinant of physical activity levels in older persons with diabetes. Monitoring of and intervention for balance confidence may be required to improve physical activity levels in older persons with diabetes.