768 PREDICTORS OF INDEPENDENCE IN COMMUNITY-DWELLING OLDER PEOPLE

Abstract

Abstract Introduction Many people wish to retain their independence as they age. Therefore, identifying factors that can predict sustainability of independence for older people over time are essential for preventing functional decline and maintaining quality of life. Method Longitudinal, health, social and economic data, from community-dwelling older people aged ≥75 years (collected at baseline (BL), 6, 12, 24 and 48 months), were obtained from the ongoing Community Ageing Research (CARE 75+) cohort study. Linear regression models were used to identify predictors of independence. The Nottingham Extended Activities of Daily Living (NEADL) score (range 0–66) at 12 months from BL was the dependent variable (DV) to indicate independence. Independent variables (IV)s were selected based on interviews with CARE75+ participants and known predictors such as, ethnicity and frailty. Each IV was regressed against the DV in univariable analyses. All IVs with p value <0.1 from univariable analyses, including baseline NEADL score, were included in the multivariable model. Results Data from 1,277 participants (mean (SD) age 84.61 (4.95); 49% male) were analysed. The multivariable model (adjusted R2: 0.71) showed that, in addition to higher BL NEADL (estimated effect 0.49, 95% confidence interval (CI) 0.41 to 0.58), white ethnicity, good sight, lower level of frailty, ability to perform basic activities of daily living, lower depression score, lower cognitive impairment, younger age, living circumstances, fewer hours of informal support, greater physical functioning and lower pain scores were predictors of a higher NEADL at 12 months (p < 0.05 for all). Conclusion As well as physical health, social and psychological variables are important in predicting independence, based on NEADL score, over time. Further research into the mechanisms behind these relationships will be conducted.