Table 1.

Multivariate multinomial logistic regression of factors associated with inappropriate dosing of direct oral anticoagulants in patients with atrial fibrillation (n = 1020)^a

Model including age	Adjusted OR (95% CI) for under-dosing	P-value^b	Adjusted OR (95% CI) for over-dosing	P-value^b
Patient age (years)	1.03 (1.02 to 1.05)	<0.001	1.05 (0.99 to 1.10)	0.084

DOAC molecule		<0.001		0.223
Apixaban	3.93 (2.29 to 6.74)		3.14 (0.82 to 12.06)
Dabigatran	1.55 (1.17 to 2.06)		1.02 (0.43 to 2.42)
Rivaroxaban	1 (ref)		1 (ref)

Kidney function (Cockcroft–Gault) (ml/min)		0.002		0.010
Normal, ≥60	1 (ref)		1 (ref)
Moderate/severe/terminal failure, <60	0.59 (0.42 to 0.83)		3.28 (1.34 to 8.08)

↵a Akaike information criterion = 1455.14; Bayesian information criterion =1504.42.
↵b Adjusted P-value for all the reported variables, obtained from the Wald test using multinomial multivariate logistic regression. CI = confidence interval. DOAC = direct oral anticoagulant. OR = odds ratio.