Table 4: Discriminant indices of Savonix test scores, to detect AD from Normal. Summary of logistic regression for logistic regressions with the predictor tests adjusted education years and ages.
Max/Mix were determined by the AUC values of logistic regression models of1000 of 20 Normal subjects and 21 AD subjects. AUC: Area Under the Receiver Operating Characteristic Curve. CI: Confidence Interval. PPV: Positive Predictive Values. NPV: Negative Predictive Values. B: regression coefficients. SD: Standard Deviations for the regression coefficients. Wald Chisq: Wald Chi square statistics. P value: P values for Wald Chi square statistics. Selected 2 subtests: Trail Making 2; Maze 2: and Complex Figure.
Savonix Test composites AUC (95% CI) Sensitivity Specificity PPV NPV B SD Wald Chisq P value
11 Subtests total score
Max 0.988 1 0.95 0.955 1 Test score -1.107 0.47 5.563 p<0.05
(0.967-1) Education 0.972 0.539 3.25 p =0.071 n.s.
Age 0.287 0.144 3.944 p<0.05
Min 0.921 0.857 0.95 0.947 0.864 Test score -1.371 0.47 8.51 p<0.01
(0.831-1) Education 0.143 0.172 0.689 p = 0.406 n.s.
Age 0.136 0.066 4.203 p < 0.05
Selected 2 subtests total score
Max 0.995 1 0.95 0.955 1 Test score -7.538 2.893 6.787 p < 0.01
(0.984-1) Education 0.25 0.343 0.532 p = 0.466 n.s.
Age 0.326 0.192 2.881 p = 0.090 n.s.
Min 0.902 0.81 0.8 0.81 0.8 Test score -4.035 1.355 8.874 p < 0.01
(0.814-0.99) Education 0.211 0.184 1.311 p = 0.252 n.s.
Age 0.068 0.076 0.796 p = 0.372 n.s.