Prospective classification of functional dependence: Insights from machine learning and 39,927 participants in the Canadian Longitudinal Study on Aging

Abstract

ABSTRACTObjectiveFunctional dependence is a multifactorial health condition that affects well-being and life expectancy. To better understand the mechanisms underlying functional dependence, we aimed to identify the variables that best prospectively classify adults with and without limitations in basic and instrumental activities of daily living.MethodsA filtering approach was used to select the best predictors of functional status from 4,248 candidate predictors collected in 39,927 participants aged 44 to 88 years old at baseline. Several machine learning models using the selected baseline variables (2010-2015) were compared for their ability to classify participants by functional status (dependent vs. independent) at follow up (2018-2021) on a training dataset (n = 31,941) of participants from the Canadian Longitudinal Study on Aging. The best performing model was then examined on a test dataset (n = 7,986) to confirm its sensitivity, specificity, and accuracy.ResultsEighteen baseline variables were identified as the best predictors of functional status at follow up. Logistic regression was the best performing model for classifying participants by functional status and achieved balanced accuracy of 81.9% on the test dataset. Older age, phycological distress, slow walking speed, perceived health, being retired, having a chronic condition, and never going for walks at baseline were associated with greater odds of being functionally dependent at follow-up. In contrast, the absence of functional limitations, greater grip strength, being a female and free of chronic conditions at baseline were associated with lower odds of being functionally dependent at follow-up.ConclusionFunctional dependence can be best prospectively estimated by age, psychological distress, physical fitness, physical activity, chronic conditions, and sex. These predictors can estimate functional dependence more than 6 years in advance with high accuracy.ImpactSuch early identification of functional dependence allows sufficient time for the implementation of interventions designed to delay or prevent functional decline.Lay SummaryWhether a patient will be dependent in 6 years can be predicted with good accuracy by 18 variables, including age, psychological distress, physical fitness, physical activity, chronic disease, and sex.