Table 3

The association of NSAID use with risk of atrial fibrillation, additionally adjusted for left ventricular end-diastolic dimension at baseline (N=4919)

Model 1*Model 2†
nHR95% CIHR95% CI
Never use241(Ref.)1(Ref.)
Current use (days)
 ≤140NA(NA)NA(NA)
 15–3053.61(1.37 to 9.50)3.21(1.22 to 8.48)
 >3031.97(0.59 to 6.56)1.71(0.51 to 5.70)
Past use (days)
 ≤3061.61(0.66 to 3.94)1.47(0.60 to 3.62)
 31–180131.14(0.58 to 2.24)1.03(0.52 to 2.03)
 >1801281.42(0.92 to 2.20)1.32(0.85 to 2.06)
  • *Adjusted for age at index date, sex and left ventricular end-diastolic dimension.

  • †Additionally adjusted for baseline values of: systolic blood pressure, diastolic blood pressure, body mass index, total and HDL cholesterol, smoking status and left ventricular end-diastolic dimension. Time dependent covariates include: blood pressure lowering drugs, history of myocardial infarction, presence of heart failure, prevalent diabetes.

  • HDL, high-density lipoprotein; n, number of atrial fibrillation cases; NA, not applicable; NSAID, non-steroidal anti-inflammatory drug.