Objectives: To compare the ability of the Short Physical Performance Battery (SPPB), 4-m walk test (4mWT), 6-minute walk test (6MWT), and handgrip strength to predict mortality, incident disability, worsening mobility, and falls in older community dwellers.
Design: Cohort study.
Participants: Individuals aged 65 and older n = 561) without prevalent basic activity of daily living (ADL) disability participating.
Measurements: Separate logistic regression models were developed to predict incident ADL disability, worsening mobility, and falls in 3 years, and Cox regression models were used to assess 7-year risk of death as a function of the four tests, adjusting for covariates.
Results: Performance tests were reciprocally correlated at baseline. After 3 years, 33 (7.3%) of 453 participants reexamined were disabled in ADLs, 87 (20%) had worsening mobility, and 99 (22%) reported falls. Of the 561 baseline participants, 141 (25%) died over the 7 years. All measures predicted incident ADL disability, with adjusted odds ratios (ORs) per unit increase of 0.85 (95% confidence interval (CI) = 0.77-0.93) for handgrip strength, 0.08 (95% CI = 0.02-0.36) for 4mWT, 0.74 (95% CI = 0.61-0.89) for SPPB, and 0.993 (95% CI = 0.988-0.997) for 6MWT. Handgrip strength (OR = 0.88, 95% CI = 0.83-0.93), 4mWT (OR = 0.33, 95% CI = 0.11-0.94), and SPPB (OR = 0.81, 95%CI = 0.71-0.93) predicted worsening mobility. No measure predicted falls; only SPPB (hazard ratio (HR) = 0.92, 95% CI = 0.85-0.997) and 6MWT (HR = 0.997, 95% CI = 0.995-0.999) predicted death.
Conclusion: Performance measures are independent predictors of relevant health outcomes, with the exception of falls. Because SPPB is easily applied and is the only measure predicting incident ADL disability, worsening mobility, and death, it is preferable to the other tests.
Keywords: cohort studies; performance tests; predictive value.
© 2015, Copyright the Authors Journal compilation © 2015, The American Geriatrics Society.