RESUMO
BACKGROUND: Double reading is a widely used criterion standard in breast cancer screening despite a lack of evidence of the cost-effectiveness of the second reading. This study evaluates the incremental cost-effectiveness of such a strategy. DESIGN: Cost-effectiveness analysis: Nationwide population-based semi-annual screening program for women aged 50-59 in Finland. Participation rate was 91%. All mammograms (95,423) performed during 1990-1995 in three screening centers of the Finnish Cancer Society were read by two radiologists with gradings recorded. The effectiveness of the double reading was the difference in cancers detected in the double compared to that of the single reading. Incremental costs of the double reading for the health care and non-health care and the time costs were estimated. The main outcome measure was the incremental cost per additional cancer found as a result of the double-reading strategy. RESULTS: The total number of cancers detected with the double and single reading were 290 and 261, respectively. A significantly higher ratio of carcinoma in situ was the causative pathology in cancers detected only by the second reader. The cost per cancer detected with a single reading was US$ 18,340. The incremental cost of any additional cancer found was US$ 25,523, that is, a 39% higher cost per additional cancer found by double reading. CONCLUSIONS: The additional cost per cancer detected by double reading is not drastically higher than with single reading. However, the additional cost per life year saved may be much higher.
