Interventions:
The two interventions were relevant as a programme of maternal screening and postpartum management was compared with typical practice in the countries analysed (no screening).
Effectiveness/benefits:
Clinical data were retrieved by a literature review the methods of which were not reported. Most estimates were taken from RCTs and so were likely to have good internal validity. However, no details on these studies were provided and the authors stated that some clinical inputs were based on poor-quality studies. A panel of experts validated the model and the clinical estimates. Cost-effectiveness results were stable to changes in clinical parameters. Use of DALYs as a measure of benefits appeared appropriate given the impact of disease on morbidity and mortality. No information was provided on sources for disutilities.
Costs:
The authors did not state the perspective adopted in the cost analysis; it appeared to be from a third-party payer viewpoint. Resource quantities and unit costs were not presented separately but unit costs for screening were given for each site included in the analysis. Private and public facilities in India and Israel were considered and costs appeared representative for screening and postpartum management. Future costs were taken from published studies conducted in USA and converted to local values, which took account of differences in income among the settings. Other details such as price year and discount rates were provided.
Analysis and results:
The study results were reported clearly. A cost-effectiveness ratio was not needed in the base case given the dominance of the intervention. Only a one-way deterministic sensitivity analysis was conducted. This showed the robustness of the base case findings, but a more comprehensive analysis would have been useful. The authors pointed out that the decision model was elaborated by a group of international experts. The study results appeared to be specific to the two countries analysed and might not be transferable to other settings.
Concluding remarks:
The study was based on a validated model and appropriate methodology, although some clinical sources were not fully described. The authors’ conclusions appear robust.