Interventions:
The comparators were appropriately selected, as treatments for asthma in children could be delivered by either MDIs or wet nubulisation.
Effectiveness/benefits:
The clinical data were from a published systematic review of the literature, which should have ensured the inclusion of all relevant trials. Only head-to-head randomised controlled trials were included ensuring high internal validity. These clinical data were varied in the sensitivity analyses. The benefit measure was specific to the disease and will not permit comparisons to be made with other disease interventions. It was chosen because the analysis focused on the reduction in hospitalisation and not long-term quality of life, as acknowledged by the authors.
Costs:
Only those costs relevant to the stated hospital perspective were analysed. The unit costs and resource quantities were presented separately, facilitating the replication of the study for other settings. Appropriate Canadian sources were used for both the unit costs and resource use data. The impact of variations in each cost category on the total costs was reported. The authors stated that physician costs for the treatment of asthma exacerbation were not included as they were assumed to be similar for the two groups. The price year was not explicitly reported, but seems to have been 2009.
Analysis and results:
An incremental analysis was appropriately conducted and showed that the MDI was the dominant option. The total costs and benefits were clearly reported for each strategy. The uncertainty was assessed in both deterministic and probabilistic analyses, varying all the parameters simultaneously. A very simple model was used, with a short time horizon, according to the study’s objectives. The authors stated that their use of clinical trials for the resource use and clinical data might have reduced the external validity of their analysis. They stated that their results should be considered to be specific to Canada. They acknowledged that the main limitation of their analysis was the use of retrospectively reported cost outcomes, instead of prospectively collected resource use.
Concluding remarks:
The methods were appropriate, and the data sources and results were clearly reported. The authors’ conclusions are valid.