Interventions:
The interventions were well described and appeared to be appropriate comparators. Current practice in the study setting was appropriately included.
Effectiveness/benefits:
The methods used to identify relevant studies from the published literature were not described in detail; the authors did not describe the methods used to the select studies from which the data was derived. The methods used to analyse data from the disease registry to account for confounding were not presented. No details of the sample size or the follow-up period were provided. Therefore, it was difficult to assess the quality of the data included in the study or whether all the relevant available evidence was included.
The benefit measure seemed appropriate, as QALYs incorporated both the morbidity and mortality of patients. Although the utilities were estimated using a regression equation from a published study, the underlying methodology supporting the estimation of utilities was not provided, so it was difficult to assess the calculation of QALYs.
Costs:
The authors presented the analysis from two perspectives, societal and health care papyer. The study appeared to include all the relevant costs from a health care payer perspective. Productivity costs associated with caregivers were included as part of a societal perspective, but no productivity costs of the patients were included in the analysis, so the study may have underestimated the societal costs. The sources of cost data were provided, but some of the supporting methodology was not fully described which made it difficult to assess the appropriateness of estimates used. The cost data was derived from sources which appeared to be highly relevant to the study setting. The costs were appropriately discounted and adjusted for inflation.
Analysis and results:
The analytical approach appeared appropriate and was adequately described, including a flow chart of the simulation. The use of an incremental analysis was appropriate to explore the relative cost-effectiveness the three treatment/assessment options. Both deterministic and probabilistic sensitivity analyses were conducted, which provided a thorough assessment of the impact on results of uncertainty in the key parameters. The results of the base-case and sensitivity analysis were clearly presented. The authors discussed a number of limitations with their study in detail.
Concluding remarks:
The quality of the study methods was satisfactory. The results were well reported. However, the effectiveness and cost data were poorly reported, making it difficult to assess the appropriateness of the authors’ conclusions.