Table 3.

Univariable and multivariable analyses of the effect of intervention on knowledge of medications.

Medication knowledge	Participants obtaining optimal scorea		Univariable	Multivariabled
	Intervention N (%) (n = 72)	Control N (%) (n = 61)	(OR, 95% CI)	(OR, 95% CI)	P-value
Individual item score (optimal score = 1)b
Name of medication	48 (66.7)	38 (62.3)	1.21 (0.59, 2.47)	1.03 (0.46, 2.34)	0.938
Reasons for administration	42 (58.3)	35 (57.4)	1.04 (0.52, 2.07)	0.87 (0.39, 1.97)	0.751
Mechanism of administration	40 (55.6)	33 (54.1)	1.06 (0.53, 2.10)	1.17 (0.49, 2.78)	0.730
Timing of medication	64 (88.9)	57 (93.4)	0.56 (0.16, 1.96)	0.44 (0.10, 1.20)	0.219
Side effect	7 (9.7)	5 (8.2)	1.20 (0.36, 4.01)	1.27 (0.32, 4.98)	0.736
What to do when
Medication side effects occur	60 (83.3)	47 (77.1)	1.48 (0.63, 3.52)	1.58 (0.62, 4.00)	0.336
Dose of medication is missed	56 (77.8)	53 (85.3)	0.61 (0.24, 1.49)	0.71 (0.26, 1.95)	0.511
Composite scorec
Median (Q1, Q3)	5.6 (3.6, 6.0)	5.0 (3.7, 6.7)	0.92 (0.63, 1.85)	0.89 (0.56, 1.41)e	0.612
Optimal (≥5)	47 (65.3)	37 (60.7)	1.22 (0.60, 2.47)	1.10 (0.42, 2.92)	0.841

^aAnalyses were restricted to 133 patients who were taking medications at 12 months.

^bItem score was calculated as the sum of scores for an item divided by the number of medications assessed.

^cComposite score was calculated as the sum of scores for all items divided by the number of categories of medications used.

^dAdjusted for type of stroke, medical history, demographic, socioeconomic, mental and functional status, number of prescribed prevention medications, length of use of prevention medications, method by which medications are packaged/provided by pharmacist, and self-reported method(s) of keeping track of medications.

^eExponentiated coefficient obtained from the stepwise multivariable linear regression analyses.

OR, odds ratio; CI, confidence interval.