Table 4.

Relationship Between Cognitive Functioning and Alcohol Consumption During Midlife and Late Life.^a,b

	$\hat{\mathit{\beta }}$	SE	F Value/t Value	P Value		$\hat{\mathit{\beta }}$	SE	F Value/t Value	P Value
Episodic memory
Midlife			0.33	.81	Late life			2.99	.03
Abstainer	–	–	–	–	Abstainer	–	–	–	–
Light	−0.06	0.08	−0.69	.49	Light	0.19	0.07	2.79	<.01
Moderate	−0.03	0.09	−0.34	.73	Moderate	0.04	0.06	0.64	.52
Heavy	−0.10	0.11	−0.88	.38	Heavy	0.18	0.12	1.49	.14
Executive function
Midlife			0.99	.40	Late life			0.96	.41
Abstainer	–	–	–	–	Abstainer	–	–	–	–
Light	−0.07	0.09	−0.79	.43	Light	0.04	0.08	0.54	.59
Moderate	0.004	0.11	0.04	.97	Moderate	0.03	0.07	0.46	.65
Heavy	−0.15	0.13	−1.20	.23	Heavy	0.23	0.14	1.68	.09
Global cognition
Midlife			1.21	.31	Late life			2.03	.11
Abstainer	–	–	–	–	Abstainer	–	–	–	–
Light	−0.07	0.06	−1.09	.28	Light	0.10	0.05	1.90	.06
Moderate	−0.04	0.07	−0.57	.57	Moderate	0.04	0.05	0.84	.40
Heavy	−0.15	0.09	−1.75	.08	Heavy	0.18	0.09	1.90	.06

Abbreviations: APOE, apolipoprotein E; SE, standard error; WRAT, Wide Range Achievement Test.

^aBold values indicate statistical significance P < .05.

^bAll models adjusted for age, gender, educational attainment, WRAT-3, smoking, APOE e4, and total brain volume. Models for late life alcohol consumption included years between neuropsychological battery and alcohol assessment as a covariate.