. Author manuscript; available in PMC: 2018 Jun 1.

Published in final edited form as: J Am Geriatr Soc. 2017 Mar 17;65(6):1152–1158. doi: 10.1111/jgs.14835

Table 4.

Associations between ECog domain and total scores at baseline, and diagnostic conversion to MCI at follow-up.

	ECog models^a		ECog and cognition models^b

Independent variable	HR (95% CI)	p-value	HR (95% CI)	p-value
Everyday Memory - informant	2.0 (1.4–2.9)	<.001	1.8 (1.2–2.9)	.006
Everyday Language - informant	1.5 (.9–2.4)	.12	1.3 (.7–2.4)	.46
Everyday Visuospatial - informant	1.4 (.8–2.4)	.18	1.1 (.6–2.2)	.69
Everyday Planning - informant	1.8 (1.1–3.1)	.02	2.0 (1.1–3.6)	.03
Everyday Organization - informant	1.9 (1.3–2.8)	.001	1.9 (1.1–3.0)	.01
Everyday Divided Attention - informant	1.7 (1.2–2.4)^c	.001	1.7 (1.1–2.7)	.02
ECog Total Score - informant	2.0 (1.3–3.2)	.002	1.9 (1.1–3.2)	.02
Everyday Memory - self	1.9 (1.3–2.6)	<.001	1.6 (1.1–2.4)	.02
Everyday Language - self	1.8 (1.2–2.7)	.003	1.6 (.9–2.6)	.10
Everyday Visuospatial - self	1.8 (1.1–2.9)	.03	1.6 (.9–2.8)	.10
Everyday Planning - self	1.8 (1.1–3.0)	.01	1.6 (.8–3.0)	.17
Everyday Organization - self	1.9 (1.2–2.9)	.003	1.4 (.8–2.4)	.21
Everyday Divided Attention - self	1.6 (1.1–2.3)	.006	1.7 (1.1–2.5)	.01
ECog Total Score - self	2.3 (1.4–3.6)	<.001	1.9 (1.1–3.4)	.03

Note: ECog = Everyday Cognition Scales; CI = confident interval; HR = hazard ratio.

Model adjusted for age, education and baseline ECog domain or total score.

Model adjusted for age, education, baseline ECog domain or total score, and cognitive functioning (memory and executive functions)

The assumption of proportional hazards is not met in this model (p=.03). When an interaction between baseline Everyday Divided Attention and logarithm of time is included in the model, the estimated hazard ratio is higher at early time points and lower at later time points (i.e., the hazard ratio decreases over time).