High-intensity walking in midlife is associated with improved memory in physically capable older adults

Table 3 Results of multiple linear regression analyses of the associations between walking and cognition (n = 188)

	Total score				Memory score				Non-memory score
	B	SE	\(\beta\)	p	B	SE	\(\beta\)	p	B	SE	\(\beta\)	p
Overall walking (n = 125)
Model 1^a
Walking	7.056	2.223	0.214	0.002	3.647	1.310	0.185	0.006	1.794	1.004	0.127	0.076
None	Reference				Reference				Reference
Model 2^b
Walking	5.477	1.984	0.166	0.006	2.852	1.214	0.144	0.020	0.889	0.063	0.063	0.339
None	Reference				Reference				Reference
Midlife-initiated walking (n = 103)
Model 1^a
Walking	7.496	2.425	0.226	0.002	4.016	1.412	0.207	0.005	1.931	1.080	0.137	0.076
None	Reference				Reference				Reference
Model 2^b
Walking	5.439	2.156	0.164	0.013	2.936	1.313	0.152	0.027	0.848	0.990	0.060	0.393
None	Reference				Reference				Reference
Late life-initiated walking (n = 22)
Model 1^a
Walking	5.541	3.255	0.177	0.093	2.418	2.213	0.114	0.278	1.672	1.556	0.116	0.286
None	Reference				Reference				Reference
Model 2^b
Walking	5.927	3.120	0.189	0.062	3.173	2.176	0.149	0.150	1.072	1.586	0.075	0.501
None	Reference				Reference				Reference

For these analyses, we used the no walking (n = 63) as the reference
APOE4 apolipoprotein ε4, GDS geriatric depression scale, VRS vascular risk score, BMI body mass index
^aAdjusted for age, sex, and APOE4
^bAdjusted for covariates in model 1 plus, education, clinical diagnosis, GDS, annual income, alcohol intake, smoking, VRS, BMI, dietary pattern including food types (such as protein and fruit or vegetables), serum nutritional markers (such as hemoglobin, albumin, glucose, and HDL-/LDL-cholesterol), and overall physical activity score

ISSN: 1758-9193