TABLE 3.

Logistic regression analysis and sensitivity analysis for different frequencies of physical activity.

	Model 1		Model 2		Model 3		Model 4
	OR (95% CI)	P	OR (95% CI)	P	OR (95% CI)	P	OR (95% CI)	P
A	0.275 (0.136–0.558)	<0.001	0.336 (0.158–0.714)	0.005	0.313 (0.141–0.696)	0.004	0.337 (0.143–0.795)	0.013
B	0.419 (0.191–0.919)	0.030	0.473 (0.203–1.103)	0.083	0.521 (0.206–1.315)	0.167	0.529 (0.197–1.419)	0.206
C	0.300 (0.086–1.053)	0.060	0.455 (0.116–1.782)	0.258	0.337 (0.077–1.481)	0.150	0.362 (0.079–1.657)	0.190
D	0.357 (0.169–0.753)	0.007	0.579 (0.256–1.310)	0.190	0.595 (0.244–1.451)	0.253	0.704 (0.279–1.775)	0.457

A: participate in physical activity daily.

B: participate in physical activity multiple times a week.

C: participate in physical activity once a week.

D: occasionally participate in physical activity.

Model 1: the unadjusted logistic regression.

Model 2: adjusted more covariates logistic regression including Aβ42, T-tau, P-tau, Aβ42/T-tau, and Aβ42/P-tau.

Model 3: first sensitivity analysis was based on more covariates including gender, years of education, and MMSE.

Model 4: second sensitivity analysis was based on more covariates including gender, years of education, MMSE, history of alcohol consumption and smoking, diabetes, hypertension, and coronary heart disease. The bold values are all p ≤ 0.05, statistically significant parameters.