Table 4

Hazard ratios (95% CI) for thromboembolism and death according to weekly alcohol intake for women with incident atrial fibrillation

Alcohol intake, drinks* per week
Abstainers<7 (reference)7 to 1314 to 20>20
Subjects, n45601229110123
Thromboembolism or death
VKA adjusted†0.90 (0.46 to 1.76)10.94 (0.67 to 1.32)1.20 (0.78 to 1.83)1.08 (0.68 to 1.70)
 Multi adjusted‡0.76 (0.38 to 1.50)11.06 (0.74 to 1.50)1.25 (0.82 to 1.92)1.23 (0.78 to 1.96)
 Frequency adjusted§0.70 (0.35 to 1.40)11.30 (0.85 to 2.01)1.66 (0.96 to 2.86)1.71 (0.93 to 3.14)
Thromboembolism
VKA adjusted0.75 (0.18 to 3.13)11.03 (0.56 to 1.91)0.77 (0.30 to 1.97)1.53 (0.73 to 3.17)
 Multi adjusted0.66 (0.16 to 2.79)11.14 (0.61 to 2.13)0.77 (0.301.98) to1.71 (0.81 to 3.60)
 Frequency adjusted0.68 (0.16 to 2.94)11.45 (0.68 to 3.11)1.11 (0.37 to 3.28)2.78 (1.02 to 7.60)
Death
VKA adjusted0.84 (0.39 to 1.81)10.93 (0.63 to 1.37)1.24 (0.77 to 1.98)0.97 (0.57 to 1.64)
 Multi adjusted0.72 (0.33 to 1.55)11.06 (0.72 to 1.57)1.29 (0.80 to 2.08)1.10 (0.64 to 1.89)
 Frequency adjusted0.65 (0.30 to 1.42)11.29 (0.79 to 2.10)1.65 (0.89 to 3.03)1.44 (0.72 to 2.87)
  • Thromboembolism: includes ischaemic stroke or peripheral arterial thromboembolism. Measures of association in bold indicates statistical significance.

  • *One drink equates to 12 g of pure alcohol.

  • †Calculated using Cox's regression with time since atrial fibrillation diagnosis as underlying time variable and adjusted for VKA treatment.

  • ‡Further adjustment for CHA2DS2-VASc components.

  • §Also adjusted for frequency of intake (not weekly, non-daily or daily).

  • VKA, vitamin K antagonist.