Table 2.
n | Univariable OR (95% CI) |
Multivariable OR (95% CI) |
P value |
33 960 | 31 615 | ||
VKA | 1.00 (–) | 1.00 (–) | |
Dabigatran | 0.53 (0.47 to 0.60) | 0.54 (0.48 to 0.62) | <0.001 |
Rivaroxaban | 0.61 (0.58 to 0.65) | 0.76 (0.71 to 0.82) | |
Apixaban | 0.57 (0.54 to 0.61) | 0.77 (0.71 to 0.84) | |
CHA2DS2VASc | |||
0–1 | 1.00 (–) | ||
2 | 1.00 (0.93 to 1.08) | <0.001 | |
3–4 | 0.94 (0.88 to 1.01) | ||
5–9 | 0.81 (0.74 to 0.89) | ||
HASBLED-1 | |||
0–2 | 1.00 (–) | ||
3–9 | 0.98 (0.93 to 1.05) | 0.62 | |
Number of drugs | |||
Continuous/Linear | 0.99 (0.98 to 1.00) | 0.067 | |
Townsend quintile | |||
1 | 1.00 (–) | ||
2 | 0.93 (0.87 to 0.99) | <0.001 | |
3 | 0.86 (0.80 to 0.91) | ||
4 | 0.91 (0.85 to 0.97) | ||
5 | 0.86 (0.80 to 0.94) | ||
Date of first prescription* (years after 1 January 2011) | |||
Continuous/Linear | 1.29 (1.22 to 1.37) | <0.001 | |
Continuous/Quadratic | 0.94 (0.93 to 0.95) | ||
BIC | 46 263.63 | 42 880.41 |
*Time difference (in years) between the date of the first ever OAC prescription for each patient and the start date of the study (1 January 2011). This suggests that the maximum effect of calendar time occurs at −ln(1.29)(2×/ln(0.94))=2.13 years.
BIC, Bayes information criterion; CHA2DS2-VASc, Congestive heart failure, Hypertension, Age≥75 years, Diabetes mellitus, Stroke, Vascular disease, Age 65-74 years, Sex category; HASBLED-1, Hypertension; Abnormal renal/liver function; Stroke; Bleeding; Labile INR; Elderly; Drugs or alcohol; OAC, oral anticoagulant; VKA, vitamin K antagonist.