Table 2: Predictive accuracy of specific combinations of predictor variables for classification of transition to Alzheimer’s disease (AD) by 3 years of followup in two independent samples (ADNI and QD) of older adults with Mild Cognitive Impairment, and comparisons of three predictor models.

ADNIQD
Model Predictor variablesAUC (SE)Sensitivity at specificity = 80% (90%)Correct classification %AUC (SE)Sensitivity at specificity = 80% (90%)Correct classification %

Age0.49712.74 (5.73)55.670.73952.61 (29.85)73.02
MMSE0.65537.88 (19.20)65.540.77841.41 (26.79)76.00
Hippocampal vol.0.72548.05 (34.42)64.600.75362.07 (41.38)80.51
Entorhinal volume0.71850.65 (35.71)67.160.77367.86 (50.00)80.34
AVLT0.75649.47 (25.16)44.330.84971.63 (53.13)80.00
FAQ0.73849.05 (35.90)44.330.70845.46 (32.83)75.42
Age, MMSE0.65936.94 (18.79)63.480.82172.73 (39.39)76.00
Hippocampal and entorhinal volumes0.74455.84 (35.71)68.980.82467.86 (67.86)88.03
AVLT/SRT and FAQ0.81162.74 (42.68)72.700.87978.13 (59.38)82.05
Model 1Age, MMSE, AVLT/SRT and FAQ0.828 (0.024)73.25 (49.05)73.400.921 (0.027)90.63 (81.25)87.07
Model 2Age, MMSE, Hippocampal and entorhinal volumes0.783 (0.028)57.79 (40.26)73.720.866 (0.046)82.14 (71.43)87.07
Model 3Age, MMSE, AVLT/SRT, FAQ, Hippocampal and entorhinal volumes0.865 (0.022)75.33 (55.20)77.010.940 (0.027)92.59 (88.89)89.72

Model comparisonsAUC difference valueAUC difference value

Model 1 versus Model 20.03960.22710.04280.3618
Model 1 versus Model 30.04280.0035**0.02820.1979
Model 2 versus Model 30.08240.0001**0.07100.0254*

A threshold of 0.5 was used on predicted risk derived from the logistic regression models. Area under the curve (AUC) was derived from receiver operating characteristic (ROC) analyses. (157 converters) in ADNI and (33 converters) in QD. The differences between models in AUCs are slightly different from the direct subtraction of AUCs between models because of missing data that ranged from 1% to 4% for the variables examined in ADNI and 1% to 5% for the variables examined in QD.
* , ** .