No description available.

Physiotherapy for people with stroke has been found to be beneficial but details of the most effective interventions are unclear. Further development of the evidence base for stroke physiotherapy is limited by a lack of clinical practice models, sensitive clinically based outcome measures and effective stratification techniques to characterise homogenous groups of subjects. These issues are addressed here with regard to balance and posture. These aspects were chosen because they form a cornerstone of stroke physiotherapy as they are thought essential for the rehabilitation of functional activities. A systematic review of assessment methods in the literature revealed a lack of measurement tools which met the utility criteria: reliability, validity, sensitivity to short-term change, suitability for a wide range of abilities, ease of use and suitability for different settings. This prompted the development of a new measurement tool. Firstly, a model of the clinical assessment process was developed using an adapted focus group method with neurological physiotherapists. This informed the content of a new measurement tool which combined an ordinal scale with functional performance tests- the Brunel Balance Assessment. The tool was evaluated in a series of studies involving 92 stroke patients. It was hierarchical (coefficient of reproducibility= 0.99, coefficient of scalability = 0.69), reliable (100% agreement) and valid as a measure of balance disability (r=0.58-0.97). The psychometric properties of the individual functional performance tests were also tested and found to be reliable (ICCs =0.88-1) and valid (r=0.32-0.63). Measurement error ranged 0-40% and the minimum change needed to detect true clinical change was calculated for each test. Balance disability, measured with the Brunel Balance Assessment, is heterogeneous with sitting, standing and stepping balance forming distinct levels of ability (p<0.027). Consequently, the BBA could be used to stratify people with stroke according to balance ability. Weakness, sensation and age were significant independent contributors to balance disability (r2=82.7%). Balance ability was a strong contributor to independence in ADL (p<0.0001). The findings of this thesis address the issues that have limited research into stroke physiotherapy with regard to balance disability. In relation to clinical practice, a robust measurement and stratification tool has been developed.