We agree with Dr. Kim that cilostazol and pentoxifylline, the only 2 U.S. Food and Drug Administration–approved drugs for treating walking impairment in patients with PAD, have modest effects on walking performance in this patient population. Therefore, we repeated the analyses that we reported in our manuscript with additional statistical adjustment for use of pentoxifylline or cilostazol. Our results (Table) did not change substantially from those reported in our manuscript. Therefore, self-directed walking exercise is associated with slower rates of functional decline in persons with PAD, independent of the use of pentoxifylline and cilostazol therapy or other confounders.

We appreciate the comments of Mr. Johnson and Dr. Bell. Unfortunately, we did not measure the intensity of walking exercise of participants in our study and therefore are not able to comment on associations between intensity of walking exercise and functional decline in our cohort. However, Gardner and colleagues (1) demonstrated similar improvement in walking performance in patients with PAD who were randomly assigned to a low-intensity versus a high-intensity exercise walking rehabilitation program.

We found that participants with PAD who walked for exercise more frequently demonstrated less functional decline in the 4-meter walk. Although we did not measure walking speed during exercise, our analyses adjusted for performance during the previous year. Consequently, our findings were independent of walking speed as determined during the previous year's visit. To determine whether our findings were similar among patients with PAD who had different levels of performance, we also repeated our analyses after we stratified participants according to baseline performance. Our results suggested that participants with PAD whose baseline performance was poorest, such as those with slowest walking speed, achieved a benefit from the self-directed exercise program that was similar to or greater than that seen by other participants with PAD.