Table 3.
Effect of standardized plasma DOAC levels on bleeding outcomes, adjusted for potential confounders
| Characteristic | First model (C-trough), n = 1657 |
Second model (C-peak), n = 1298 |
||
|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | |
| Standardized C-trough DOAC levels, first 3 mo | 1.36 | 1.02-1.78 | - | - |
| Standardized C-trough DOAC levels, >3 mo | 0.97 | 0.66-1.45 | - | - |
| Standardized C-peak DOAC levels, first 3 mo | - | - | 1.10 | 0.70-1.74 |
| Standardized C-peak DOAC levels, >3 mo | - | - | 0.71 | 0.45-1-13 |
| HAS-BLED score | 1.04 | 0.66-1.62 | 1.14 | 0.67-1.96 |
| BMI, kg/m2 | 1.00 | 0.93-1.06 | 1.00 | 0.93-1.07 |
| Glomerular filtration rate, mL/min | 1.00 | 0.99-1.02 | 1.01 | 0.99-1.02 |
| Low- vs standard-dose DOAC, first 3 mo | 3.61 | 1.28-10.2 | 3.30 | 1.6-10.3 |
| Low- vs standard-dose DOAC, >3 mo | 1.40 | 0.61-3.23 | 2.17 | 0.82-5.74 |
| Antiplatelet treatment (yes vs no) | 1.06 | 0.66-1.62 | 1.23 | 0.51-3.00 |
Both models were estimated using time-varying Cox regression. The inclusion of enrollment center as a potential confounder was not significant (P > .9 for both models) and it is not reported because it did not materially change estimates.
BMI, body mass index; CI, confidence interval; HR, hazard ratio.