Abstract
Background: Individuals with chronic stroke are less active, which is both a consequence of stroke-related impairments and a risk factor for future health complications. The PROWALKS clinical trial found significant gains in real-world walking activity (steps/day) after 12 weeks of a step activity monitoring behavioral intervention, provided either alone (SAM) or with high-intensity gait training (FAST + SAM), but not after high-intensity gait training alone (FAST). Previous research in individuals after stroke suggests that tailored behavioral counseling may lead to better long-term physical activity participation, but no previous work has focused on post-intervention maintenance of walking activity changes. Objective: To investigate whether steps/day changes after training (POST) were maintained at 6 months (6MO) and 12 months (12MO) after baseline. We hypothesized that SAM and FAST + SAM groups would have better maintenance of steps/day changes than the FAST group. Methods . This analysis included all participants who completed the PROWALKS intervention (n = 200, mean[SD] age: 63.27[12.41], 102 male/98 female, >6 months post-stroke). Analysis outcomes were steps/day change from POST-6MO, and from POST-12MO. Results: All groups significantly decreased in steps/day from POST-6MO ( P = .001, FAST decreased by mean[SE] 160[272], SAM by 1016[270], FAST + SAM by 400[300]), and POST-12MO ( P < .001, FAST decreased by 610[280], SAM by 1072[306], FAST + SAM by 568[313]). There were no significant differences between groups. Conclusions: All intervention groups showed significant declines in steps/day between POST and 6MO and between POST and 12MO. These results add to a growing body of literature suggesting that a behavioral intervention to initiate behavior change may not be sufficient for maintenance of change. Registration: This study is registered at ClinicalTrials.gov, NCT02835313.
Funding Information
  • Eunice Kennedy Shriver National Institute of Child Health and Human Development (NIH R01 HD086362-01A1)

This publication has 45 references indexed in Scilit: