Skip to main content
. 2019 Dec 13;93(1108):20190580. doi: 10.1259/bjr.20190580

Table 1.

Meta-analysis of pooled odds ratios of increase in cancer detection rate per 1000 females screened and the increase in recall rate obtained by comparing single reading with CAD or double reading to single reading

Single reading with CAD Double reading
Matched
N = 5
Unmatched
N = 5
Unilateral
N = 6
Mixed
N = 3
Arbitration
N = 8
Odds ratio of increase in cancer detection rate 1.09 1.02 1.13 1.07 1.08
Odds ratio of increase in recall rate 1.12 1.10 1.31 1.21 0.94

CAD, computer-aided diagnosis.

Matched studies: the assessment before and after using CAD was on the same mammogram.

Unmatched studies: the performance of mammography facilities was compared before and after the introduction of CAD. Different mammograms are interpreted in the two conditions.

N is the number of studies included in each group.9