. 2024 Feb 29;11:1348918. doi: 10.3389/fnut.2024.1348918

Table 4.

Logistic regression analysis of soybean product consumption and executive function among Tibetan children and adolescents in high altitude areas of Tibet, China.

Executive dysfunction	OR(95% CI)
Executive dysfunction	Model 1	Model 2	Model 3
Inhibit function dysfunction
≥6 t/w	1.000	1.000	1.000
2–5 t/w	1.570 (1.063, 2.319)^a	1.587 (1.073, 2.347)^a	1.584 (1.066, 2.353)^a
≤1 t/w	1.821 (1.155, 2.871)^a	1.868 (1.177, 2.963)^a	1.844 (1.152, 2.951)^a
p for trend	<0.001	<0.001	<0.001
1back dysfunction
≥6 t/w	1.000	1.000	1.000
2–5 t/w	1.146 (0.785, 1.672)	1.146 (0.783, 1.678)	1.002 (0.674, 1.489)
≤1 t/w	1.181 (0.748, 1.865)	1.191 (0.748, 1.896)	0.891 (0.547, 1.450)
p for trend	<0.001	<0.001	<0.001
2back dysfunction
≥6 t/w	1.000	1.000	1.000
2–5 t/w	1.229 (0.828, 1.824)	1.257 (0.845, 1.869)	1.155 (0.762, 1.749)
≤1 t/w	1.507 (0.949, 2.394)	1.633 (1.020, 2.613)^a	1.198 (0.725, 1.979)
p for trend	<0.001	<0.001	<0.001
Conversion function dysfunction
≥6 t/w	1.000	1.000	1.000
2–5 t/w	1.833 (1.157, 2.906)^a	1.873 (1.176, 2.981)^a	1.483 (0.88, 2.498)
≤1 t/w	2.792 (1.675, 4.655)^b	3.012 (1.787, 5.076)^b	2.008 (1.106, 3.646)^a
p for trend	<0.001	<0.001	<0.001

Model 1 is a crude model. Model 2 controls for age, gender, father’s education, and mother’s education. Model 3 controls for waist circumference, daily moderate to high intensity physical activity, and maximal oxygen uptake (VO_2max) on the basis of model 2. a indicates p < 0.05, b indicates p < 0.01.