Interventions:
The selection of the comparators appears to have been appropriate and the screening strategies were likely to be applicable in other health care settings.
Effectiveness/benefits:
The clinical data were from a large sample of women enrolled in UK centres, but only a few cancer cases were detected and there was high uncertainty around the additional cases detected with clinical breast examination, as acknowledged by the authors. The patients’ characteristics and types of cancer were described in detail. The impact of variations in the clinical parameters was not evaluated. The authors stated that QALYs were the summary benefit measure, but no quality adjustments were mentioned, so life-years appear to have been used. These were an appropriate measure, given the impact of breast cancer on life expectancy.
Costs:
The perspective was not clearly stated, but costs relevant to the UK NHS appear to have been included. The unit costs were presented for all items in the analysis. The quantities of resources reflected the screening patterns as observed in the patients enrolled in the clinical study. The price year was not reported, making reflation exercises impossible. The data sources were reported and were typical NHS sources, for most items. The impact of variations in the cost estimates was not considered.
Analysis and results:
The results were not clearly reported, as the calculations that led to the estimation of the cost per QALY were not clear. It was unclear how the QALYs were calculated, and it seems that the cost per life-year saved was presented. The uncertainty was not investigated; no sensitivity analyses were carried out. The analysis had important drawbacks, especially in the simplistic calculation of the cost-effectiveness ratios. The findings were specific to the UK and cannot be transferred to other settings.
Concluding remarks:
The methods of the analysis were not clearly described, especially for the cost-effectiveness calculations. Caution is required when interpreting the authors’ conclusions.