Reliability and responsiveness of two physical performance measures examined in the context of a functional training intervention

Background and Purpose. The reliability and responsiveness of 2 physical pe rformance measures were assessed in this nonrandomized, controlled pilot ex ercise intervention. Subjects. Forty-five older individuals with mobility i mpairment (mean age=77.9 years, SD=5.9, range = 70-92) were sequentially as signed to participate in an exercise program (intervention group) or to a c ontrol group. Methods. The intervention group performed exercise 3 times a week for 12 weeks that targeted muscle force, endurance, balance, and flexi bility. Outcome measures were the 8-item Physical Performance Test (PPT-8) and the 6-minute walk test. Test-retest reliability and responsiveness inde xes were determined for both: tests; interrater reliability was measured fo r the PPT-8. Results. The intraclass correlation coefficient for interrater reliability for the PPT-8 was .96. Intraclass correlation coefficients for test-retest reliability were .88 for the PPT-8 and .93 for the 6-minute wa lk test. The intervention group improved 2.4 points and the control group i mproved 0.7 point on the PPT-8, as compared with baseline measurements. The re was no change in 6-minute walk test distance in the: intervention group; when compared with the control group. The responsiveness index was .8 for the PPT-8 and .6 fbr the Ci-minute walk test. Conclusion and Discussion. Me asurements for both the PPT-8 and the 6-minute walk:test appeared to be hig hly reliable. The PPT-8 was more responsive than the 6-minute walk test to change in performance expected with this functional training intervention.