All items were performed

without assistance. Participants were scored on the best of three performances. The Global Perceived Effect of Treatment was rated separately through questionnaires at Week 4 and Week 6 by the treating physiotherapists and participants (or their carers if the participants did not have the capacity to answer the questions). Assistance was provided to participants (or their carers) as needed by staff not otherwise involved in the study. The treating physiotherapists and participants (or their carers) were initially asked if they thought their wrists were better, the same or worse. Those who stated Transmembrane Transporters modulator that their wrists were better were asked to rate the improvement between 1 (a little better) and 6 (a very great deal better). Those who stated that their wrists were worse were asked to rate the deterioration between 1 (a little worse) and 6 (a very

great deal worse). These data were PD98059 datasheet analysed by combining responses into a single 13-point scale with –6 reflecting a very great deal worse, 0 reflecting no change and +6 reflecting a very great deal better. The minimally important difference was set at 1 point (Schneider and Olin 1996). Perception of treatment credibility was evaluated by the treating physiotherapists and participants (or their carers) at Week 4 using questionnaires which captured their tolerance to the treatment (scored on a 5-point scale), their perceptions of the worth of the treatment (scored on a 5-point scale), their perceptions of the effectiveness of the treatment (scored on a 5-point scale), and their willingness to continue with the same treatment if it were to be provided (scored yes or no). of Assistance was provided to participants (or their carers) as needed by staff not otherwise involved in the study. Treating physiotherapists were also asked to indicate if they would administer the treatment to the participants if further management for wrist contracture was needed (scored yes or no). In addition, participants

and physiotherapists were asked open-ended questions directed at identifying any issues or concerns about the intervention(s). The sample size was calculated a priori. Best estimates indicated that a sample size of 36 participants was required to provide an 80% probability of detecting a between-group difference of 5 degrees for the primary outcome, assuming a standard deviation of 5 degrees ( Bakhtiary and Fatemy 2008) and a 10% drop-out rate. The minimally important difference for the primary outcome was set at 5 degrees in line with a number of previous studies on joint contracture ( Harvey et al 2000, Harvey et al 2003, Horsley et al 2007, Lannin et al 2007, Lannin et al 2003). Linear regression analyses were performed to assess the effect of the intervention on passive wrist extension and strength.