The study used a Markov model populated with patient-level data from a randomised controlled trial (RCT) of the intervention, and routinely collected data. The Markov model quantified the events and risk over time. The cycle length was one day and the time horizon was 24 weeks. The authors stated that the perspective was that of the health care provider.
The effectiveness data came from one RCT of the intervention (64 patients) compared with usual care (58 patients), conducted over a 24-week period. Full details of the clinical trial were presented in another paper (Courtney, et al. 2009, see 'Other Publications of Related Interest' below for bibliographic details). The main clinical effectiveness measure was the number of emergency re-admissions and the patient’s health-related quality of life, measured by the Short Form (SF)-12 Health Survey. The outcomes were measured at baseline, four, 12, and 24 weeks.
Monetary benefit and utility valuations:
Outcomes from the SF-12 were mapped to the European Quality of life (EQ-5D) scale, using a published algorithm. All time points appear to have been used, reflecting the full profile over the 24 weeks.
Measure of benefit:
The primary measure of benefit was the number of quality-adjusted life-years (QALYs) gained.
The costs included those of equipment (stretchy band and pedometer) from the RCT, staff (nurses and physiotherapists) from published salary schedules, and health service visits and bed days from national benefits schedules, hospital statistics, and published data. The price year was 2008 and all costs were in Australian dollars (AUD).
Analysis of uncertainty:
The authors estimated the impact of uncertainty by performing a probabilistic sensitivity analysis, with 1,000 re-samples. The results were presented in a cost-effectiveness acceptability curve.