Acurácia do teste de velocidade de marcha para predizer mortalidade em sete anos entre idosos residentes na comunidade

Abstract

In older people, the usual gait speed is an important test of physical performance, because the inability to perform the test e slow execution are associated with several negative health outcomes, including mortality. Also, it has other advantages, such as low cost, simplicity in the measurement process, and the possibility of application in different health care spaces. However, despite these characteristics, few studies have focused specifically on its predictive capacity and the impact of its application on the care of the elderly.Obstacles in determining the best cut-off value for classifying individuals based on their gait speed represent another limitation associated with its use in clinical and research settings. This study aimed to evaluate the external validity of the gait speed in predicting seven-year mortality among communitydwelling older people. The linkage was carried out based on the Frailty in Brazilian Older People Study, Rio de Janeiro section, and the Brazilian Mortality Information System database from 2009 until December 2016. The variables of interest of the present study included sex, age (age of 65–79 years and age of 80 years or greater) and gait speed. The measures of diagnostic accuracy included the sensitivity, specificity, positive likelihood ratio and area under the curve. The optimal cut-off points were identified by considering the highest possible sensitivity and specificity values. Two hundred and four older individuals (27.9%) died during the study period. These individuals presented a lower gait speed (0.77 m/s) than the survivors (0.87 m/s) (p<0.001). The accuracy was low in all subgroups. In the total sample, the ‘optimal’ cutoff value was 0.83 with a sensitivity of 63.7%, a specificity of 63.2%, an area under the curve of 0.64 and a positive likelihood ratio of 1.73. The most accurate cut-off value was 0.8 m/s with a sensitivity of 58.3 and a specificity of 68.0%. The findings highlight the obstacles in using the gait speed as a single tool due the low accuracy in detecting individuals at risk of death. Researchers and health professionals should consider using the gait speed in conjunction with other simple tests in the assessment of older individuals.