Table 3.
The effect of GPAs on the risk of gastrointestinal bleeding in NSAID users.
| Multivariable analysis | |||
|---|---|---|---|
| HR | 95% CI | P value | |
| Concomitant GPA use versus non-use | |||
| None | 1.00 | Reference | |
| PPI | 0.30 | 0.20-0.44 | <0.001 |
| H2RA | 0.35 | 0.29-0.43 | <0.001 |
| Misoprostol | 0.47 | 0.23-0.95 | 0.035 |
| Rebamipide | 0.43 | 0.35-0.51 | <0.001 |
| Eupatilin | 0.98 | 0.86-1.12 | 0.74 |
| GPA use versus PPI co-therapy | |||
| PPI | 1.00 | Reference | |
| H2RA | 1.19 | 0.79-1.79 | 0.409 |
| Misoprostol | 1.58 | 0.72-3.46 | 0.257 |
| Rebamipide | 1.44 | 0.96-2.16 | 0.078 |
| Eupatilin | 3.25 | 2.21-4.77 | <0.001 |
The analysis adjusted for sex, age, body mass index, creatinine level, and co-morbidities (hypertension, cerebrovascular disease, diabetes mellitus and ischemic heart disease).