Role of Gait Speed in the Assessment of Older Patients

Over the last 3 decades, a large and increasing amount of evidence has documented the importance of physical function in elderly individuals, both as a crucial component of clinical assessment as well as a specific outcome for interventions.¹ However, the evaluation of physical function is still not considered as relevant as that of other clinical or biochemical parameters. Insufficient time, inadequate space, and the need for special equipment are some of the obstacles to the routine assessment of physical function in the geriatric clinical settings.² To overcome these limitations, gait speed has repeatedly and increasingly been proposed as an unique measure of physical performance and as a potential screening tool,^{3 - 4} but adoption has remained inconsistent.

In this issue of JAMA, the study by Studenski et al⁵ fills an important research gap and paves the way to a broader adoption of gait speed assessment. Their findings from a pooled analysis of 9 major cohort studies confirm gait speed as a predictor of mortality in older persons^{3 - 4 ,6} and also provide the statistical foundations to estimate expected survival at different ages based only on gait speed. Several barriers limiting the clinical use of physical performance measures might thus be considered overcome. First, the study establishes the validity of a test readily adoptable to clinical use that, differently from other functional assessment tools, is inexpensive, objective, and easy to interpret. Second, the study standardizes the methods to assess gait speed to the 4-meter-long track starting from a still, standing position. This is not a trivial accomplishment because gait speed has often been measured testing individuals over tracks of different lengths (eg, 8 ft or 4 or 6 m), sometimes after exclusion of the initial acceleration time, leading to difficulties in interpretation, comparability, or both. Third, the study by Studenski et al⁵ assesses survival of older persons associated with various gait speed results.

The increase in life expectancy at a population level has further highlighted the heterogeneity of individuals, making it increasingly difficult to distinguish merely old (chronologically aged) from geriatric (biologically aged) patients. This very ability to screen frail geriatric patients vs old individuals is an urgent matter in geriatrics but increasingly urgent in other specialties. For example, oncologists seek objective methods to identify patients eligible for standard and more aggressive interventions because they are chronologically, but not biologically, old.⁷ Also, cardiac surgeons have proposed gait speed as a means to identify which older patients undergoing surgery may be at increased risk for adverse outcomes.^{8 - 9} Assessment of gait speed may serve as a single-item screening tool to determine which patients need a geriatric multidisciplinary approach to care. At the same time, gait speed should be regarded as an important component of the comprehensive geriatric assessment, provided it facilitates clinical decisions on the basis of a “hard” outcome such as life expectancy.

The study by Studenski et al⁵ may also be of interest for investigators involved in clinical research on aging. Researchers and clinicians seek results that are clinically, not just statistically, significant. Because of the established association of gait speed with survival, it could be used as a surrogate for survival outcomes in clinical trials of older persons.

Gait speed should not be regarded solely as a measure of lower extremity function. Gait speed has been associated with clinical (eg, comorbidities)¹⁰ as well as subclinical conditions (eg, atherosclerosis¹¹ or inflammatory status¹² ) and is able to predict several health-related events even apparently unrelated to physical function (eg, cognitive impairment, hospitalization, institutionalization).^{3 - 4} Gait speed may serve as a marker of physiological reserve and potentially could quantify overall health status. Indeed, a close relationship between mobility performance and the aging process has been documented in humans³ and animals.^{13 - 15} This convergence may indicate the existence of shared pathophysiological mechanisms; thus, at some point, gait speed may be considered a new “vital sign,” specifically sensitive for older persons.

Results by Studenski et al⁵ showed that gait speed had only a statistically fair accuracy for predicting survival for gait speed (pooled area under the receiver operator characteristic curve equal to 0.717 and 0.737 for 5-year and 10-year survival, respectively). However, very similar results were also reported from models testing the overall predictive value of multiple clinical data (ie, prevalent diseases, body mass index, systolic blood pressure, and history of prior hospitalizations). Therefore, although the gait speed assessment may present some limitations in accurately estimating the expected survival, it still matches the predictive value obtained from an extensive and more complex clinical evaluation. Moreover, it is also noteworthy that overall mortality is a composite outcome, subject to a wide range of unmeasurable confounders, and thus may limit the ability to identify a highly accurate predictive model.

It is possible that the study sample considered in the analyses by Studenski et al⁵ may not be representative of the general older population, limiting the direct applicability of the findings. Nevertheless, the study illustrates the importance of transforming gait speed into a clinically meaningful parameter and demonstrating its predictive ability above and beyond classical comorbidities. Clinical practice is based on assumptions and, thus, never perfect. In this context, the categorization of a continuous variable (and this study greatly facilitates this task) represents the easiest way to quickly identify those at risk.

The simplicity of the gait speed assessment does not exclude possible issues from its routine implementation in clinical settings, especially in primary care practice. Functional limitations, cognitive impairment, hearing or visual problems, and balance or postural disorders may render the test more time-consuming than expected. Furthermore, clinicians will be required to spend some additional time explaining the significance of the novel assessment to their patients and relatives. However, the logistical, methodological, and conceptual familiarization of the clinical staff with the test might speed up its assessment over time. Moreover, the gait speed assessment does not need to be performed in all older patients because it is not as informative when functional disability is already present.

The prevention of physical disability is a primary goal of geriatric medicine and a major public health priority. In the scenario shaped by the present study, future analyses should consider physical disability as a primary outcome, expanding and completing the understanding of the role and relevance of gait speed as a means to identify patients at risk for functional decline. This task requires a consensus around the definitions of disability to be most widely generalizable.

Because no evidence definitively supports the hypothesis that gait speed improvements are associated with better health-related outcomes, gait speed should not be considered as a primary target for interventions at this time. It represents a global marker of health status, and an optimal secondary and complementary outcome to support research findings, clinical decisions, or both aimed at modifying more pragmatic end points.

In conclusion, assessing gait speed in older persons is likely to be a useful research tool and may have a clinical role. Future research will be needed to determine whether gait speed has the potential to change the way in which a patient is defined as geriatric.