Interventions:
The interventions were well reported and appear to have included a range of possible regimens and no treatment. It is likely that these interventions will be relevant in other settings.
Effectiveness/benefits:
The methods used to identify and select the source studies were not described and it is unclear if all the best available evidence was used. The description of the sources and the clinical estimates was poor. A number of assumptions were made, which might have been appropriate, but no justifications were given; these may have been available in the model publication (Holland, et al. 2009). The instrument used to estimate the utility values and whose preferences were estimated were not reported and might not have been appropriate. QALYs were accrued over the lifetime of the population modelled, but no discounting was reported.
Costs:
The authors did not report the perspective of the analysis. The abstract suggested that it was societal, but the cost categories were consistent with a health care payer perspective; only the direct costs were included. The identification of the cost and resource use data was not reported in detail, but the sources for the unit costs were provided. The authors did not discuss discounting of costs, but it might not have been necessary as the costs were primarily incurred early in the first year of the model.
Analysis and results:
The authors provided brief details of the model and stated that full details were given elsewhere (Holland, et al. 2009). An incremental analysis was appropriately performed to compare the relative cost-effectiveness of the treatment regimens. The impact of uncertainty was explored in one- and two-way sensitivity analyses. A more comprehensive probabilistic sensitivity analysis could have captured the full impact of overall parameter uncertainty on the results. The reporting was brief and this reduced the transparency of the analysis. The authors discussed a number of limitations to their study, such as the need for assumptions for several parameters.
Concluding remarks:
There were a few limitations to the study especially in the reporting of the methods and results. This makes it difficult to assess the authors’ conclusions, which may reflect the evidence available, but should be considered with caution.