Exercise Interventions With Trained Home Helpers for Preventing Loss of Autonomy and Falls in Community-Dwelling Older Adults Receiving Home Heath Physical Therapy T4H: A Randomized Controlled Pilot Study

Background and Purpose: Older adults at risk for falls live independently in the community in their own home and have rehabilitation needs. However, little is known about whether home coaching of older adults can decrease falls at home. We sought to determine whether a novel program for preventing falls and a loss of exercise capacity, the T4H program, in which home helpers act as exercise coaches by using an information technology (IT) device, was acceptable and feasible. Methods: Between February 2015 and October 2015, we performed a cluster randomized controlled trial in which home helpers either assisted older adults 75 years and over, to participate in the T4H program, or provided standard home help over 3 months. We assessed levels of acceptability and satisfaction among the older adults and home helpers with regard to the exercise program and the technologies used. To measure efficacy, the main outcome measures for the older adults were the absence of falls requiring medical or paramedical care, unplanned hospitalizations, walking ability in a Timed Up and Go test (TUG), and self-care ability by the Barthel Index at the 3-month follow-up visit. Results and Discussion: Overall, 35 older adults were included, aged 89 years and with 68.6% women. Eighty-five percent of the respondents were pleased or very pleased to have participated in the T4H exercise program, 70% were satisfied with the IT devices, and 92% were satisfied with their home helper's level of involvement. Two of the 4 home helper respondents were satisfied or very satisfied with the exercise program, and 2 were moderately satisfied. The proportions of older adult participants with no falls or no unplanned hospitalizations were higher in the T4H group (92.3% and 85.7%, respectively) than in the control group (81.8% and 71.4%, respectively), although these intergroup differences were not statistically significant. The T4H and control groups did not differ significantly with regard to the TUG time (median [IQR]: 27.6 seconds [17.9-58.6] vs 30.7 seconds [19.7-57.2], respectively) or the Barthel Index (median [IQR]: 90 [75-95] and 90 [75-95], respectively). Conclusions: The novel T4H home help model was feasible and was associated with a high level of participant satisfaction. We observed a trend toward fewer falls and hospitalizations and better quality of life in the older adults.