Standardized Functional Assessments Help Predict Falls and ER Visits in PACE

August 8, 2016

Standardized functional assessments significantly contribute to the prediction of falls and emergency room (ER) utilization within PACE programs, according to a new study by InnovAge in conjunction with the University of Colorado-Denver. The findings were presented at the Annual Research Meeting of AcademyHealth recently in Boston.

Authors of the study, titled “Impaired Physical Function Is a Predictor of Falls and Health Care Utilization for Program of All-Inclusive Care for the Elderly (PACE) Participants,” were Jason Falvey, Lucine Papazian, Allison Gustavson and Jennifer Stevens-Lapsley, of the University of Colorado-Denver, and Lisa Price and Michael Muller, of InnovAge.

They reported that physical therapy assessments may lead to earlier identification of PACE participants at risk for adverse events. Because physical function is a modifiable risk factor, earlier identification could lead to a reduction in falls and ER visits, which may decrease costs for PACE programs and improve the ability of PACE participants to remain in community settings.

Impaired physical function is a biomarker for adverse events in older adult populations, especially those within long-term care settings. Community-based long-term care recipients in PACE often have significant impairments in physical function, yet it is unclear if this is associated with an increased risk of falls or ER visits. Because a barrier to determining this relationship is the variety of functional measures used by physical therapists, the main goal of this quality improvement project was to determine if a standardized functional battery measured by physical therapists helps identify dual Medicare/Medicaid-eligible PACE participants at risk for a fall or ER visit.

The study involved dual-eligible PACE participants across five InnovAge clinics in the Denver metropolitan area. Physical therapists at the facilities adopted two standardized assessments: the Short Physical Performance Battery (SPPB), which assesses overall physical function, and the timed up and go (TUG), which assesses walking ability.

The researchers identified 762 PACE participants across the five facilities who were assessed between March and August 2015 and had three-month follow-up data available. Demographic information was extracted from the electronic medical record for each patient and linked with functional data, facility ER claims and fall reports. SPPB and TUG data were categorized into mild, moderate and severe impairment. Logistic regression was used to calculate odds ratios (OR) for functional measures on the binary outcomes of falls and ER utilization.

Overall, 70 participants reported a fall and 72 had an ER visit within the three-month period following their assessment. Univariable analysis revealed that PACE participants with severe functional impairment (SPPB < 4/12) had more than a two-fold increase in fall risk (OR = 2.70, 95 percent CI 1.42-5.10) and ER visit utilization (OR = 2.07, 95 percent CI 1.11-3.89) compared to those with mild functional impairment (SPPB ≥ 7/12). Moderate functional impairment (SPPB 4/12-6/12) was associated with a significant increase in fall risk (OR = 2.08, 95 percent CI 1.16-3.76) compared to mild functional impairment.

Similarly, participants with severe impairments in walking ability (TUG time >30 seconds) had a significant increase in fall risk (OR = 2.27, 95 percent CI 1.21-4.27) and a non-significant increase in ER visit utilization (OR = 1.65, 95 percent CI 0.84-3.27) compared to participants with mild impairments (TUG time < 20 seconds).

In a multivariable logistic regression model controlling for patient characteristics and clinical site, severe functional impairment remained a significant predictor of falls (OR = 2.15, 95 percent CI 1.03-4.48) and ER visits (OR = 2.03, 95 percent CI 1.05-3.95) compared to mild impairment.

For more information, contact InnovAge.

Back to Top