Rehabilitation in care homes (RICH-T): a cluster-randomized controlled trial
Article first published online: 21 NOV 2007
Copyright © 2007 John Wiley & Sons, Ltd.
Physiotherapy Research International
Volume 12, Issue 4, page 205, December 2007
How to Cite
Sackley, C., Patel, S. and Wright, C. (2007), Rehabilitation in care homes (RICH-T): a cluster-randomized controlled trial. Physiother. Res. Int., 12: 205. doi: 10.1002/pri.388
- Issue published online: 23 NOV 2007
- Article first published online: 21 NOV 2007
- Cited By
Introduction. In a care home population, loss of independence in mobility is strongly associated with poor quality of life and mortality. The present study assessed the effects of targeted physiotherapy (PT) and occupational therapy (OT) interventions on mobility. A cluster-randomized controlled trial was set in 24 care homes in Birmingham. Each home was independently randomized to an intervention group (12 care homes, 128 residents) or a control group (12 care homes, 121 residents). Residents receiving end-of-life care were excluded. Ethical review was favourable. Method. The interventions comprised PT and OT targeted to mobility, delivered to individual residents, including equipment provision and carer education. The control group received standard care. Researchers masked to group allocation assessed residents with the Rivermead Motility Index (RMI) at baseline and three months (pre- and post-intervention), and at six months. Data were analysed by use of SAS, with all statistical hypothesis tests performed at 5% level of significance. Intention-to-treat analyses were conducted on the RMI, accounting for clustering and baseline values. One hundred and eighty-five participants were female; their ages ranged from 48 to 108 years (mean 85 years) and 168 (67%) were cognitively impaired. Result. Mean RMI scores at baseline were 5.8 (standard deviation [SD] 4.1) and 6.9 (SD 3.8); at three months 5.1 (SD 3.5) and 6.7 (SD 3.7); and at six months 5.2 (SD 3.8) and 6.5 (SD 2.8) in the intervention and control groups, respectively. The corresponding mean scores for the Timed Up-and-Go (TUG) test were 60 (SD 42) and 57 (SD 54), 68 (SD 50) and 59 (SD 56), and 59 (SD 37) and 55 (SD 37), respectively. Conclusion. There were no statistically significant differences in mean scores between groups on either outcome. The intervention was well tolerated but no between-group differences were seen. The sample size estimate in future studies should allow for the variability in participants and homes. Copyright © 2007 John Wiley & Sons, Ltd.