Table 7 Optimal odor identification score combination from standard version for discriminating normal biomarkers and AD continuum individuals.

From: Odor identification score as an alternative method for early identification of amyloidogenesis in Alzheimer’s disease

Variances

Estimate

Standard error

z value

Pr ( >|z|)

Analysis A. Coefficients of logistic regression model for the standard version (12 items)

 (Intercept)

4.172

1.589

2.625

0.0087**

 A (India ink)

− 1.350

0.702

− 1.924

0.0544*

 B (wood)

− 1.655

0.708

− 2.336

0.0195*

 C (perfume)

− 1.000

0.708

− 1.414

0.1575

 D (menthol)

0.651

0.869

0.750

0.4534

 E (Japanese orange)

0.789

0.865

0.911

0.3622

 F (curry)

− 2.125

1.461

− 1.455

0.1457

 G (gas leak odor)

− 0.416

0.671

− 0.620

0.5354

 H (rose)

− 1.204

0.727

− 1.657

0.0975*

 I (hinoki (Japanese cypress wood))

0.418

0.781

0.535

0.5924

 J (sweaty socks)

− 1.242

0.654

− 1.899

0.0576*

 K (condensed milk)

− 1.016

0.747

− 1.361

0.1737

 L (roasted garlic)

0.606

0.833

0.727

0.4671

 AIC

–

–

–

98.57

Analysis B. Coefficients of the best logistic regression model for the selected version (4 items)

 (Intercept)

3.777

1.244

3.035

0.0024**

 A (India ink)

− 1.254

0.602

− 2.082

0.0373*

 B (wood)

− 1.960

0.598

− 3.277

0.0011**

 F (curry)

− 1.885

1.203

− 1.567

0.1170

 J (sweaty socks)

− 1.197

0.604

− 1.980

0.0477*

AIC

–

–

–

90.189

  1. The table for Analysis A presents the estimated coefficients (Estimate), their respective standard errors, z-values, and p-values for various variables in a logistic regression model. Each variable (A, B, C, etc.) represents different odors or predictors analyzed within the model. The estimates signify the magnitude and direction of the effect each variable has on the outcome being studied. The standard errors provide a measure of the variability or uncertainty around these estimates. The z-values and p-values indicate the statistical significance of each variable; lower p-values generally suggest stronger evidence against the null hypothesis, implying a more significant impact of the variable on the outcome. Lastly, the AIC (Akaike Information Criterion) with a value of 98.57 serves as a measure of model fit, where lower values indicate a better fit of the model to the data.
  2. The table for Analysis B displays estimates, standard errors, z-values, and p-values for variables in a logistic regression model. Variables A, B, F, and J show statistically significant relationships indicated by their associated p-values (* and **). The AIC value of 90.189 assesses model fit. AIC is lower for the selected type compared to the standard version, indicating superior odor identification in the selected version.
  3. AIC Akaike Information Criterion.
  4. *T test; **Wilcoxon rank sum test.
Back to article page