Table 3.

Comparison of IICV components to IICV. Used Akaike Information Criterion (AIC) stepwise backward selection to examine contribution of IICV and individual components of IICV to model fit.

		Predictors and order dropped, or P-Value of retained
Baseline Diagnostic Status	Selection criteria	IICV	Trails A	Trials B	ANART	RAVLT
MCI or Cognitively Healthy N=1307	AIC	0.0153	1st	0.0001	0.0028	<0.0001
MCI N=814	AIC	0.0060	1st	<0.0001	0.0003	<0.0001

Note: P-values provided by AIC models indicate whether predictors offer predictive value, not the magnitude of added value.

Abbreviations:

MCI: Mild Cognitive Impairment

IICV: Intra-individual cognitive variability

ANART: American National Adult Reading Test

RAVLT: Rey Auditory Verbal Learning Test

AIC: Akaike Information Criterion