Table 3.
Competing risk models for thrombosis after PV diagnosis by treatments and patient characteristics (n = 820)
| Characteristic* | Model 1 | Model 2 | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| Phlebotomy | ||||||
| No | 1.00 | |||||
| Yes | 0.52 | 0.42-0.66 | <.01 | |||
| Phlebotomy intensity, times per year | 0.46 | 0.29-0.74 | <.01 | |||
| HU PDC, every 10% | 0.92 | 0.89-0.96 | <.01 | 0.92 | 0.88-0.95 | <.01 |
| Age, y | 1.01 | 0.99-1.02 | .55 | 1.01 | 0.99-1.03 | .27 |
| Sex | ||||||
| Female | 1.00 | 1.00 | ||||
| Male | 0.94 | 0.74-1.20 | .61 | 0.98 | 0.77-1.24 | .86 |
| Race | ||||||
| White | 1.00 | 1.00 | ||||
| Nonwhite | 0.74 | 0.46-1.20 | .23 | 0.85 | 0.53-1.38 | .52 |
| Modified Elixhauser score | ||||||
| 0 | 1.00 | 1.00 | ||||
| 1 | 1.28 | 0.94-1.75 | .11 | 1.30 | 0.95-1.77 | .10 |
| ≥2 | 1.37 | 1.01-1.86 | .05 | 1.38 | 1.03-1.86 | .03 |
| Prior thrombosis | ||||||
| No | 1.00 | 1.00 | ||||
| Yes | 1.17 | 0.82-1.67 | .40 | 1.01 | 0.71-1.43 | .98 |
| Disability | ||||||
| No | 1.00 | 1.00 | ||||
| Yes | 0.98 | 0.66-1.45 | .92 | .01 | 0.69-1.47 | .98 |
| Low-income subsidy | ||||||
| No | 1.00 | 1.00 | ||||
| Yes | 1.48 | 1.10-1.99 | .01 | 1.38 | 1.03-1.85 | .03 |
| Influenza vaccination within 12 mo prior to PV diagnosis | ||||||
| No | 1.00 | 1.00 | ||||
| Yes | 0.99 | 0.78-1.26 | .96 | 0.98 | 0.78-1.25 | .88 |
*
All variables in the table were simultaneously included in the same model. The only difference between models 1 and 2 was that model 1 included phlebotomy as a binary variable and model 2 included the intensity of phlebotomy (number of phlebotomies each year).