Effect of Structured Physical Activity on Overall Burden and Transitions Between States of Major Mobility Disability in Older Persons: Secondary Analysis of a Randomized Trial

Abstract

Background: The total time a patient is disabled likely has a greater influence on his or her quality of life than the initial occurrence of disability alone.

Objective: To compare the effect of a long-term, structured physical activity program with that of a health education intervention on the proportion of patient assessments indicating major mobility disability (MMD) (that is, MMD burden) and on the risk for transitions into and out of MMD.

Design: Single-blinded, parallel-group, randomized trial. (ClinicalTrials.gov: NCT01072500).

Setting: 8 U.S. centers between February 2010 and December 2013.

Participants: 1635 sedentary persons, aged 70 to 89 years, who had functional limitations but could walk 400 m.

Intervention: Physical activity (n = 818) and health education (n = 817).

Measurements: MMD, defined as the inability to walk 400 m, was assessed every 6 months for up to 3.5 years.

Results: During a median follow-up of 2.7 years, the proportion of assessments showing MMD was substantially lower in the physical activity (0.13 [95% CI, 0.11 to 0.15]) than the health education (0.17 [CI, 0.15 to 0.19]) group, yielding a risk ratio of 0.75 (CI, 0.64 to 0.89). In a multistate model, the hazard ratios for comparisons of physical activity with health education were 0.87 (CI, 0.73 to 1.03) for the transition from no MMD to MMD; 0.52 (CI, 0.10 to 2.67) for no MMD to death; 1.33 (CI, 0.99 to 1.77) for MMD to no MMD; and 1.92 (CI, 1.15 to 3.20) for MMD to death.

Limitation: The intention-to-treat principle was maintained for MMD burden and first transition out of no MMD, but not for subsequent transitions.

Conclusion: A structured physical activity program reduced the MMD burden for an extended period, in part through enhanced recovery after the onset of disability and diminished risk for subsequent disability episodes.

Primary funding source: National Institute on Aging, National Institutes of Health.

Figure 1

Risk ratios denoting burden of major mobility disability over time for physical activity versus health education according to subgroups.

Figure 2

Six-month probability of transitions between no MMD, MMD and death according to intervention group for all follow-up, Year 1 and post Year 1. Values were estimated from the multi-state model, as described in the Appendix (Section 2.0). MMD denotes major mobility disability. Bars represent 95% confidence intervals.