ROBot-assisted physical training of older patients during acUte hospitaliSaTion-study protocol for a randomised controlled trial (ROBUST)

During hospitalisation, older patients spend most of their time passive in bed, which increases the risk of functional decline and negative adverse outcomes. Our aim is to examine the impact of robot-assisted physical training on functional status in older geriatric patients during acute hospitalisation. This is a single-centre investigator-blinded placebo-controlled randomised controlled trial including geriatric patients aged ≥ 65 years, able to ambulate before hospitalisation, and with expected length of stay ≥ 2 days. In addition to standard physiotherapy treatment, the intervention group receive active robot-assisted resistance training and the control group passive robot-assisted sham training. Exclusion criteria are as follows: ambulation without assistance at the time of inclusion, known severe dementia, delirium, patients who have received less than three training sessions at discharge, terminal illness, recent major surgery/lower extremity fracture, conditions contradicting the use of training robot, lower extremity metastases, deemed unsuitable for robot-assisted training by a healthcare professional, or weight > 165 kg. The primary outcome is functional status assessed by change in Barthel Index-100 and 30-s chair stand test between inclusion and day of discharge. Secondary outcomes include functional status at 1- and 3-month follow-up, quality of life, depression, concern about falling, falls, cognition, qualitative interviews, need of homecare, discharge destination, readmissions, healthcare costs, sarcopenia, muscle quantity (bioimpedance), and mortality. Clinical meaningful change of the Barthel Index is 5 points. A recent study in geriatric patients reported a 6.9-point change following exercise. With a significance level of 5%, 80% power, and a drop-out rate of 20%, 244 participants per group (n = 488) are needed to detect the same mean difference. With a significance level of 5%, 80% power, and a drop-out rate of 20%, 74 participants per group (n = 148) are needed to detect a minimum clinical change of 2.6 repetitions for 30-s chair stand test. Recruitment started in January 2023 and is expected to continue for 19 months including follow-up. If our study shows that in-hospital robot-assisted training prevents functional decline in older patients, this may have a major impact on the individual patient due to increased wellbeing and a higher level of independency. In addition, society will benefit due to potential decrease in the need of m