Background: Hospital readmissions within 30 days of discharge have gained national attention and account for more than $17 billion in avoidable U.S. Medicare expenditures each year. (1) In 2011 alone, there were approximately 3.3 million adult 30-day all-cause hospital readmissions, costing the US $41.3 billion.(2) As such, efforts have been made to create algorithms to predict patient risk of 30-day readmission based on known risk factors.(3) For medical admissions, patient mobility is considered to be a core component of patient readiness for discharge.(4) However, there have been few validated approaches to quantifiably risk-stratify patient mobility.(5) We were unable to find any practical, quantitative real-time tools identifying patients at risk for readmission based on mobility, frailty, and fall risk currently used as the standard of care in hospitalized settings.
Methods: Our hypothesis was that an instrumented version of the Quantitative Timed Up and Go (QTUG) test (6) can help quantitate mobility and physical activity for hospitalized inpatients ≥60 years of age using previously validated inertial sensors to establish mobility, frailty, and fall risk while hospitalized. The instrumented QTUG test is a digital application on a Samsung tablet that generates a risk score based on 1) a 8-question verbally-administered patient mobility questionnaire and 2) a 30-domain computational algorithm assessing mobility while standing, walking, turning and sitting. The QTUG mobility test consists of rising from a seated position, walking 3 meters, turning 180 degrees, walking 3 meters and retuning to the seated position while wearing a < 2-ounce biosensor positioned along each anterior shin (calf) via a disposable/adjustable leg strap). Inpatients generating QTUG test scores of ≥70 (based on the QTUG mobility testing and patient responses to the questionnaire) may be at an increased risk of falls, hospital readmissions, and ED visits within the first 30 days of hospital discharge. We tested the ability of QTUG to predict post-discharge outcomes. Data sources included post- discharge patient questionnaires administered by phone and electronic health record data.
Results: Outcome data for 155 inpatients (median age= 68; 75% Medicare insured; female (n,%) = 57 (40) demonstrated increased rates of 30-day hospital readmissions (OR=2.1, p= 0.04), 30-day emergency department visits (OR= 1.4, p=0.03), and falls within the first 30 days after discharge (OR= 1.6, p=0.06) among patients with QTUG scores ≥ 70. Most common co-morbidities included hypertension, coronary artery disease, and diabetes mellitus type 2 and no statistically significant differences across participants were noted. Analysis is ongoing and additional univariate and bivariate analysis for demographic data and discharge disposition is pending and expected January 2023.
Conclusions: This digital version of a analytic physical therapy assessment may optimize risk-prediction of important patient outcomes (hospital readmissions, falls, and ED visits) in the first 30 days after discharge. Further study is needed in and ultimately, this work may assist in establishing a standardized approach to determining the most appropriate hospital discharge disposition and fall risk prevention tools for aging inpatient populations in the future.