Autor: |
Justin Whitten, Bryant Oleary, Dawn Tarabochia, David Graham |
Rok vydání: |
2022 |
Předmět: |
|
Zdroj: |
Innovation in Aging. 6:38-38 |
ISSN: |
2399-5300 |
DOI: |
10.1093/geroni/igac059.144 |
Popis: |
Background Perturbation-based training (PBT) is a balance training method that causes a trip like event requiring a rapid step response to regain balance. There are numerous examples in the literature demonstrating the effectiveness of PBT but the need to use an expensive treadmill in a scientific laboratory limits the general applicability of PBT as a community-based intervention. A possible solution is to rapidly stop the treadmill belt during walking using the e-brake as the perturbation event. Importantly this could be performed on a commercially available, lower cost treadmill. Therefore, the purpose of this study was to evaluate the effectiveness of a commercial treadmill during PBT. Methods Seventeen participants completed either 9 weeks of PBT or conventional balance training based on ACSM guidelines. During an initial and final testing session participants balance recovery performance was evaluated. Participants were released from a forward static lean angle and asked to recover with a single step, during this test their movement was recorded and subsequently used to determine the Margin or Stability pre- and post-training. Participants were tracked for 6 months following the intervention and falls were recorded on a weekly basis. Results and Summary: There was no difference in balance recovery performance between groups following the training intervention and there was no difference in fall rate between groups in the 6-month follow-up period. We conclude that overall using the e-brake of a commercial treadmill is ineffective as a PBT strategy as it elicits no greater benefit than conventional exercise training. |
Databáze: |
OpenAIRE |
Externí odkaz: |
|