Pilot comparative effectiveness study of surface perturbation treadmill training to prevent falls in older adults

Background: Falls are the leading cause of fatal and non-fatal injuries among older adults. Exercise programs appear to reduce fall risk, but the optimal type, frequency, and duration of exercise is unknown. External perturbations such as tripping and slipping are a major contributor to falls, and task-specific perturbation training to enhance dynamic stability has emerged as a promising approach to modifying fall risk. The purpose of this pilot study was 1) to determine the feasibility of conducting a large pragmatic randomized trial comparing a multidimensional exercise program inclusive of the surface perturbation treadmill training (SPTT) to multidimensional exercise alone (Standard PT); and 2) to assess fall outcomes between the two groups to determine whether an effect size large enough to warrant further study might be present. Methods: A randomized pilot study at two outpatient physical therapy clinics. Participants were over age 64 and referred for gait and balance training. Feasibility for a larger randomized trial was assessed based on the ability of therapists to incorporate the SPTT into their clinical practice and acceptance of study participation by eligible patients. Falls were assessed by telephone interview 3 months after enrollment. Results: Of 83 patients who were screened, 73 met inclusion criteria. SPTT was successfully adapted into clinical practice and 88% of eligible subjects were willing to be randomized, although 10% of the SPTT cohort dropped out prior to treatment. The SPTT group showed fewer subjects having any fall (19.23% vs. 33.33% Standard PT; p < 0.227) and fewer having an injurious fall (7.69% vs. 18.18%; p < 0.243). These results were not statistically significant but this pilot study was not powered for hypothesis testing. Conclusions: Physical therapy inclusive of surface perturbation treadmill training appears clinically feasible, and randomization between these two PT interventions is acceptable to the majority of patients. These results appear to merit longer-term study in an adequately powered trial. Trial registration: clinicaltrials.gov: NCT01006967