Related Experiment Video
Updated: Aug 5, 2026
Assessment of Dependence in Activities of Daily Living Among Older Patients in an Acute Care Unit
Published on: September 30, 2020
Functional Capacity Assessment Predicts Loss of Independence in Older Adults Undergoing Noncardiac Surgery
Lee A Goeddel1, Samuel Peter Bronckers1, Natalie Waldron1
1Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
Background:
Preoperative evaluation is essential for optimizing perioperative care. This study assessed the clinical utility of the Timed Get Up and Go Test (TUG) and the current practice of subjective functional capacity assessment in predicting loss of independence (LOI) in older adults undergoing noncardiac surgery.
Methods:
Elective surgical cases with TUG evaluated at the Johns Hopkins Center for Preoperative Optimization clinic (06/2018-01/2020) were retrospectively identified. Anesthesiologists evaluated subjective functional capacity using metabolic equivalents of tasks (METs < 1, METs 1-4, METs > 4). LOI was classified as discharge to a higher level of care, outside of the home, with new mobility deficit or functional dependence. We used adjusted logistic regression to characterize the relationship between METs and LOI. Kappa statistics were calculated to assess the agreement between METs and TUG. Cross validation was performed to assess how the sequential inclusion of METs then TUG affected prediction of LOI from a base model.
Results:
Among 3302 eligible patients, 225 (6.8%) experienced LOI. In adjusted logistic regression with METs > 4 as the reference category, METs 1-4 was associated with an approximately 3× increase in odds of LOI (aOR = 2.9; 95% CI = 1.1-7.3; p = 0.02) while < 1 METs was associated with a 14X increase in odds of LOI (aOR = 14; 95% CI = 5.2-35; p < 0.0001). METs minimally agreed with TUG. METs improved LOI prediction from a base model (AUC 0.74 to AUC 0.77). Adding TUG (Base model + METs+TUG) further improved LOI prediction (AUC 0.83) reclassifying patients in all METs levels (NRI = 0.704; 95% CI = 0.574-0.834; p < 0.0001), particularly METs 1-4, an otherwise indeterminate-risk group that frequently undergoes additional preoperative testing.
Conclusions:
METs-based subjective functional capacity marginally improves LOI prediction and only showed partial agreement with TUG-based objective functional capacity assessment. TUG improved LOI prediction over current clinical care models that utilize METs for functional capacity assessment alone. TUG may offer additional utility to current clinical care models that evaluate functional capacity by METs alone.
