Table 3. Primary Analysis.
| Clinical outcome | Patients, No. (%) | Adjusted OR (95% CI) | P value | ||
|---|---|---|---|---|---|
| Pretreated (n = 59 894) | Not pretreated (n = 4963) | Missing | |||
| Primary end point | |||||
| Death at 30 da,b | 846 (1.4) | 125 (2.5) | 0 | 1.44 (0.78-2.62) | .36 |
| Secondary end point | |||||
| Death at 1 ya,c | 2324 (4.3) | 241 (7.1) | 0 | 1.34 (0.77-2.34) | .30 |
| Definite stent thrombosis at 30 da,d | 243 (0.2) | 19 (0.2) | 0 | 1.17 (0.64-2.16) | .60 |
| In-hospital bleedinga,e | 3562 (6.0) | 380 (7.5) | 11 (0.1) | 1.49 (1.06-2.12) | .02 |
Abbreviation: OR, odds ratio.
Instrumental variable analysis. The following variables were entered into regression in addition to the instrumental variable: age, sex, diabetes, indication for percutaneous coronary intervention, the severity of the coronary disease, smoking status, hypertension, hyperlipidemia, previous myocardial infarction, previous percutaneous coronary intervention, previous coronary artery bypass graft, arterial access site, type of stent, type of P2Y12 antagonist, Killip class, completeness of revascularization, and hospital.
Underidentification test, P = .02; weak identification test (F statistic), 357; overidentification test of all instruments, P = .89.
Underidentification test, P = .006; weak identification test (F statistic), 275; overidentification test of all instruments, P = .51.
Underidentification test, P = .01; weak identification test (F statistics), 356; overidentification test of all instruments, P = .77.
Includes major bleeding (Bleeding Academic Research Consortium type 3) and minor bleeding (Bleeding Academic Research Consortium type 2). Underidentification test, P = .02; weak identification test (F statistic), 372; overidentification test of all instruments, P = .61.