Table 2. Details of component studies included in the systematic review by McGettigan and Henry [18].
| Study, Year | Setting | Study Design | Data Type | Number of Participants | Risk Measure | Risk Estimate † (95% CI) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Celecoxib | Rofecoxib | Meloxicam | Naproxen | Diclofenac | Ibuprofen | Indomethacin | Piroxicam | ||||||
| Bak et al. [39], 2003 | Denmark | Case–control | Admin | Cases = 4,765, controls = 40,000 | OR | — | — | — | 0.70 (0.40–1.22) | 1.10 (0.70–1.73) | 1.30 (1.00–1.69) | 1.40 (0.80–2.45) | 0.50 (0.20–1.25) |
| Curtis et al. [40], 2003 | US | Cohort | Admin/MR | 3,577 users, 6,673 non-users | HR | — | — | — | — | 0.84 (0.70–1.01) | — | — | |
| Fischer et al. [41], 2005 | UK | Case–control | EMR | Cases = 8,688, controls = 33,923 | OR | — | — | — | 0.96 (0.66–1.38) | 1.23 (1.00–1.51) | 1.16 (0.92–1.46) | 1.36 (0.82–2.25) | 0.95 (0.53–1.69) |
| Garcia Rodriquez et al. [42], 2000 | UK | Case–control | EMR | Cases = 1,013, controls = 5,000 | OR | — | — | — | — | — | — | — | — |
| Garcia Rodriquez et al. [43], 2004 | UK | Case–control | EMR | Cases, = 4,975, controls = 20,000 | OR | — | — | 0.97 (0.60–1.56) | 0.89 (0.64–1.24) | 1.18 (0.99–1.40) | 1.06 (0.87–1.29) | 0.86 (0.56–1.32) | 1.25 (0.69–2.25) |
| Gislason et al. [44], 2006 | Denmark | Cohort | Admin | 29,362 users, 29,070 non-users | OR | 2.06 (1.73–2.45) | 2.29 (1.99–2.65) | — | — | 2.19 (1.93–2.49) | 1.39 (1.27–1.53) | — | — |
| Graham et al. [45], 2005 | US | Case–control | Admin | Cases = 8,134, controls = 31,496 | OR | 0.84 (0.67–1.04) | 1.34 (0.98–1.82) | 1.14 (1.00–1.30) | — | 1.06 (0.96–1.17) | — | — | |
| Hippisley-Cox and Coupland [46], 2005 | UK | Case–control | EMR | Cases = 9,128, controls = 86,349 | OR | 1.21 (0.96–1.54) | 1.32 (1.09–1.61) | — | 1.27 (1.01–1.60) | 1.55 (1.39–1.72) | 1.24 (1.11–1.39) | — | — |
| Johnsen et al. [47], 2005 | Denmark | Case–control | Admin | Cases = 10,280, controls = 102,797 | OR | 1.25 (0.97–1.62) | 1.80 (1.47–2.21) | — | 1.50 (0.99–2.29) | — | — | — | — |
| Kimmel et al. [48], 2004 | US | Case–control | Ad hoc | Cases = 1,055, controls = 4,153 | OR | — | — | — | 0.48 (0.28–0.82) | — | 0.52 (0.39–0.69) | — | — |
| Kimmel et al. [49], 2005 | US | Case–control | Ad hoc | Cases = 1,718, controls = 6,800 | OR | 0.43 (0.23–0.79) | 1.16 (0.70–1.93) | — | — | — | — | — | — |
| Lévesque et al. [50], 2005 | Canada | Case–control | Admin | Cases = 2,844, controls = 56,880 | RR | 0.99 (0.85–1.16) | 1.24 (1.05–1.46) | 1.06 (0.49–2.30) | 1.17 (0.75–1.84) | — | — | — | — |
| MacDonald and Wei [51], 2003 | UK | Cohort | Admin | Cases = 822, controls = 6,285 | HR | — | — | — | — | 0.80 (0.49–1.31) | 1.73 (1.05–2.84) | — | — |
| Mamdani et al. [52], 2003 | Canada | Case–control | Admin | 66,964 users, 100,000 non-users | RR | 0.90 (0.70–1.16) | 1.00 (0.80–1.25) | 1.00 (0.60–1.67) | |||||
| McGettigan et al. [53], 2006 | Australia | Case–control | Ad hoc | Cases = 328, controls = 487 | OR | 1.11 (0.59–2.11) | 0.63 (0.31–1.28) | — | — | — | 0.98 (0.53–1.81) | — | — |
| Ray [54], 2002 | US | Cohort | Admin | 181,441 users, 181,441 non-users | RR | — | — | — | 0.95 (0.82–1.09) | — | 1.15 (1.02–1.28) | — | — |
| Ray [55], 2002 | US | Cohort | Admin | 151,728 users, 202,916 non-users | RR | 0.96 (0.76–1.21) | — | — | 0.93 (0.82–1.06) | — | 0.91 (0.78–1.06) | — | — |
| Schlienger et al. [56], 2002 | UK | Case–control | EMR | Cases = 3,315, controls = 13,139 | OR | — | — | — | 0.68 (0.42–1.13) | 1.38 (1.08–1.77) | 1.17 (0.87–1.58) | 1.03 (0.58–1.85) | 1.65 (0.78–3.49) |
| Solomon et al. [57], 2002 | US | Case–control | Admin | Cases = 4,452, controls = 17,700 | ReR | — | — | — | 0.84 (0.72–0.98) | — | 1.02 (0.88–1.18) | — | — |
| Solomon et al. [58], 2004 | US | Case–control | Admin | Cases = 10,895, controls = 49,044 | OR | 0.93 (0.84–1.02) | 1.14 (1.00–1.31) | — | — | — | — | — | — |
| Watson et al. [59], 2002 | UK | Case–control | EMR | Cases = 809, controls = 2,285 | OR | — | — | — | 0.57 (0.31–1.06) | 1.68 (1.14–4.29) | 0.74 (0.35–1.55) | — | — |
†Relative risk of COX-2 inhibitor compared with no-use or remote exposure; accompanying 95% confidence intervals replicated to the second decimal using RevMan 5.3.
Admin, administrative data; EMR, electronic medical records; HR, hazard ratio; MR, medical records; OR, odds ratio; ReR, relative risk; RR, rate ratio.