increase to the next ambulation classification, was significantly associated with increased

activity and participation scoring, particularly for those with the slowest baseline gait

velocity.

 Those who successfully transitioned to the next ambulation classification were

significantly younger then those who did not (p = 0.0023). When age was controlled for,

those with baseline gait speeds of < 0.4 m/s demonstrated the greatest gains in activity

and participation. Among those who walked less than 0.4 m/s, there were significant

differences in SIS ADL/IADL, mobility, participation, and SIS-16 scores found between

those who were successful and who failed to advance to the next ambulation

classification. This may be attributable to the fact that those with greater deficits had

more opportunity for improvement and there were less ceiling effects in the measures.

 After age was controlled for in the less impaired walking group, there was

significant difference in participation, between those who did and did not successfully

progress to the next classification. SIS participation items include higher level home and

community activities. Increased gait velocity was clearly attributable to increased

abilities to complete higher level tasks at home and in the community. A significant

difference in age contributed to mobility and SIS-16 scores. After controlling for age, the

SIS-16 approached significance, but likely was related to the assessment of higher level

ADLs and mobility. The lack of other significant findings is probably due to the

participants' high degree of independence at baseline and subsequent ceiling effects of

the measures.

 Stratifying by gait speed decreased the sample size, possibly decreasing the

detected differences between groups. Other post-stroke changes may have also impacted