Abstract and Introduction
Abstract
Background: A physical performance evaluation can inform fall risk in older people, however, the predictiveness of a one-time assessment is limited. The trajectory of physical performance over time has not been well characterized and might improve fall prediction. We aimed to characterize trajectories in physical performance and determine if fall prediction improves using trajectories of performance.
Methods: This was a cohort design using data from the National Health and Aging Trends Study. Physical performance was measured by the short physical performance battery (SPPB) with scores ranging from 0 (worst) to 12 (best). The trajectory of SPPB was categorized using latent class modeling and slope-based multilevel linear regression. We used Cox proportional hazards models with an outcome of time to ≥2 falls from annual self-report to assess predictiveness after adding SPPB trajectories to models of baseline SPPB and established non-physical-performance-based variables.
Results: The sample was 5969 community-dwelling Medicare beneficiaries aged ≥65 years. The median number of annual SPPB evaluations was 4 (IQR, 3–7). Mean baseline SPPB was 9.2 (SD, 3.0). The latent class model defined SPPB trajectories over a range of two to nineteen categories. The mean slope from the slope-based model was −0.01 SPPB points/year (SD, 0.14). Discrimination of the baseline SPPB model to predict time to ≥2 falls was fair (Harrell's C, 0.65) and increased after adding the non-performance-based predictors (Harrell's C, 0.70). Discrimination slightly improved with the SPPB trajectory category variable that had the best fit (Harrell's C, 0.71) but did not improve with the SPPB linear slope. Calibration with and without the trajectory categories was similar.
Conclusions: We found that the trajectory of physical performance did not meaningfully improve upon fall prediction from a baseline physical performance assessment and established non-performance-based information. These results do not support longitudinal SPPB assessments for fall prediction.
Introduction
Falls are the leading cause of fatal and nonfatal injuries among adults aged ≥65 years old.[1] A priority in healthcare is therefore to identify individuals at increased risk for falls and initiate fall prevention strategies.[2,3] A bedside physical performance assessment – walking speed, standing balance, and chair rise – has been shown to be associated with falls and a predictor of future falls.[4–6] The evaluation is recommended by the Centers for Disease Control and Prevention as part of the process to assess fall risk in community-dwelling older adults.[7] However, the discriminatory performance of a single assessment with falls was only fair (c-statistic, 0.65–0.68).[4,5]
Physical performance can change over time, and at different rates over time, based on a variety of factors such as aging, co-morbidities, injuries, and therapies. As a result, longitudinal assessments of physical performance and the associated trajectories in performance may improve the ability to predict future falls. A few previous studies of older adults have applied latent class modeling to describe the trajectory of physical performance.[8–10] These studies all identified three trajectories which were generally characterized as either a good baseline performance with minimal decline over time, an intermediate-good baseline performance with a mild–moderate decline, or an intermediate-low baseline performance with a substantial decline. A limitation of the latent class modeling, however, is that the results often lack much granularity because only a small number of trajectory categories are identified. The prior studies were also limited by having only 3–4 performance assessments available to define the trajectories. Finally, the prior studies did not evaluate whether the trajectory data improved fall prediction.
In this study, we sought to describe and define trajectories in physical performance using up to 8 years of data from a national sample of older adults in the National Health and Aging Trends Study (NHATS). We aimed to compare a latent class modeling approach to define trajectory categories with an approach that uses multi-level linear regression to determine a more granular slope at the individual level. To estimate the potential clinical utility of trajectories in physical performance, we then evaluated the marginal accuracy of predicting time to ≥2 falls by adding the trajectory data to models of baseline SPPB and established non-performance-based fall predictors (e.g., self-reported fear of falling, problems with balance, and use of an assistive device). We hypothesized that trajectory data would improve predictiveness and that slope trajectories would meaningfully outperform latent class trajectory categories.
J Am Geriatr Soc. 2022;70(12):3413-3423. © 2022 Blackwell Publishing
