The interventions were reported clearly. They were valid interventions for the setting, but it was not clear whether they included all the options.
The clinical and effectiveness data were from a small, non-randomised study and two trials identified by a systematic review. The pilot study included only 30 patients (with five patients in the anastrozole group) and its results should be treated with caution. The authors combined these results with those from two trials, with a total of 364 patients. The meta-analysis combined data based on the number of cycles. The authors appropriately reported the methods of their systematic review and it is likely that all relevant studies were included. A quality assessment of these trials was not reported, making it difficult to assess their validity and hence the validity of the model inputs. The measure of benefit was stated to be quality-adjusted life-years, but the utility derivation and values were not reported, severely limiting the analysis.
The perspective was reported to be that of the hospital, but indirect costs, such as out-of-pocket expenses, work loss, and transport costs, were included and it appears that a societal perspective was adopted. For this societal perspective, all the major relevant cost categories were analysed. The authors appropriately reported the sources for the unit costs and resource use. The price year, time horizon, and currency were appropriately reported. As the pilot study supplied the resource use, the generalisability of the findings may be limited.
Analysis and results:
The costs and effectiveness data were synthesised using a decision-analytic Markov model. Appropriate details of this model were reported, including a diagram. The authors’ conclusion that clomiphene citrate was cost-effective, compared with anastrozole, was based on the average cost-utility ratios, rather than the incremental cost-utility ratio. The incremental ratio for anastrozole, compared with clomiphene citrate, was $52 per QALY gained, suggesting that anastrozole was highly cost-effective. The uncertainty in the model was assessed in a limited one-way sensitivity analysis. This went some way towards evaluating the uncertainty, but a probabilistic sensitivity analysis could have captured the overall model uncertainty. The authors reported that the main limitation to their study was the small number of patients included in their pilot study.
The conclusions of this study should be treated with extreme caution. The authors’ conclusion that clomiphene citrate was cost-effective, compared with anastrozole, was based on average cost-utility ratios, rather than the incremental cost-utility ratio, and this was inappropriate.