Analytical approach:
A decision-analytic Markov model was used to synthesise the data from published studies, a hospital database, and expert opinion. The time horizon was 11 days. The authors stated that they took a Danish hospital perspective.
Effectiveness data:
The key clinical outcomes were the days in hospital and the pain level when arterial and venous punctures were performed. Pain was measured on a visual analogue scale and the mean values were used from a published study (Giner, et al. 1996, see ‘Other Publications of Related Interest’ below for bibliographic details). Danish hospital statistics were used to estimate the average length of stay and the average number of arterial punctures per patients.
Monetary benefit and utility valuations:
The pain scores from the Giner, et al. (1996) study were transformed into utility data using Danish tariffs for the European Quality of life (EQ-5D) questionnaire. It was assumed that venous conversion produced a gain in quality of life of 0.062 for three minutes, compared with arterial puncture.
Measure of benefit:
The measure of benefit was quality-adjusted life-years (QALYs).
Cost data:
The direct medical costs included the education of nurses, the product license for the blood gas analyser, staff, materials, and complications. Published sources were used to measure the complication costs and extra materials while expert opinion and the hospital database provided data for the remaining resources. Costs were calculated in Danish kroner for 2008 to 2009 and converted to UK pounds sterling (£) using the 2010 exchange rate of £1 equals 8.6376 kroner.
Analysis of uncertainty:
The model parameters were examined in a probabilistic sensitivity analysis of 10,000 Monte Carlo simulations. Beta distributions were assigned to the probabilities, while gamma distributions were assigned to the costs and utilities. The results were illustrated in a cost-effectiveness acceptability curve. Scenario analyses were performed to assess other pulmonary departments with different patient volumes.