Table 3

Associations among experience-related variables in the 21 radiologists routinely interpreting mammograms and sensitivity, specificity and overall accuracy (multivariate analysis).


Sensitivity
Specificity
Accuracy†





Adj. OR
95% CI
P
Adj. OR
95% CI
P
Adj. OR
95% CI
P

Annual volume of mammogram Reading









     More than 5,000 mammograms
0.93
(0.56 to 1.54)
0.780
1.15
(0.98 to 1.35)
0.090
1.06
(0.90 to 1.25)
0.450
     Less than 5,000 mammograms
1.00
(reference)

1.00
(reference)

1.00
(reference)

Feedback









     Yes
0.98
(0.43 to 2.27)
0.957
1.37
(1.03 to 1.85)
0.029
1.19
(0.89 to 1.56)
0.243
     No
1.00
(reference)

1.00
(reference)

1.00
(reference)

Percentage of time spent on breast radiology









     Less than 25%
0.70
(0.36 to 1.37)
0.295
1.49
(1.18 to 1.89)
0.001
1.10
(0.87 to 1.38)
0.438
     More than 25%
1.00
(reference)

1.00
(reference)

1.00
(reference)

Years of practice reading mammograms









     Less than 10 years
1.40
(0.84 to 2.34)
0.193
1.04
(0.89 to 1.21)
0.640
1.14
(0.97 to 1.34)
0.102
     More than 10 years
1.00
(reference)

1.00
(reference)

1.00
(reference)

Do the radiologist usually consult with other radiologists when reading?









     Yes
0.61
(0.29 to 1.30)
0.202
1.04
(0.84 to 1.28)
0.717
0.89
(0.71 to 1.11)
0.300
     No
1.00
(reference)

1.00
(reference)

1.00
(reference)

Radiologist' age









     More than 45
0.56
(0.31 to 1.01)
0.053
1.33
(1.12 to 1.59)
0.001
1.00
(0.83 to 1.19)
0.971
     Less than 45
1.00
(reference)

1.00
(reference)

1.00
(reference)


Adj. OR: Adjusted odds-ratio; 95% CI: 95% confidence interval; P: Chi-square probability

† Correct diagnosis of both cancer and absence of cancer on recalling women for additional investigations.

Molins et al. BMC Health Services Research 2008 8:91   doi:10.1186/1472-6963-8-91