Interventions:
The comparators appear to have been the possible strategies for the prevention of cryptococcosis in a high endemic area. A clear description of each strategy was provided.
Effectiveness/benefits:
: A literature review to identify the clinical inputs was not reported and limited details of the methods and other characteristics of the primary studies were given. This makes it difficult to judge the validity of the clinical sources. Issues around the use of data from sources that might not be homogeneous in their characteristics (types of intervention or patient populations) were not addressed. Most of the data were from local sources or studies conducted by international organisations in Cambodia and they are likely to have reflected the epidemiological situation in Cambodia. The benefit measure was appropriate for capturing the impact of the disease on survival, which was the relevant dimension of health for these patients.
Costs:
The analysis appears to have been consistent with the perspective stated, as only the direct medical costs, borne by the health care payer, were considered. A list of cost items was provided, with some unit costs. The treatment costs were presented as category totals. The price year was reported, allowing reflation exercises in other time periods. The cost estimates were treated deterministically and no variation in the economic inputs was considered in the sensitivity analysis, except for the cost of fluconazole, which affected the cost-effectiveness of prophylaxis as expected.
Analysis and results:
The results were clearly presented and the costs and benefits were appropriately synthesised, using an incremental approach, which allowed the identification of the most cost-effective strategy. The uncertainty was investigated, using a partial approach that considered only variations in single selected inputs. The authors acknowledged that the results were applicable to the setting used for the clinical data and will not be easily transferable to low endemic areas. The study had a short time horizon, but the authors stated that one year was sufficient to capture the important effects of these strategies for HIV patients.
Concluding remarks:
Some aspects of the analysis were not extensively presented, but the methods appear to have been valid and the authors’ conclusions are robust.