Table 2.
Number of patients exposed to the drug during | Crude | Adjusted | |||||||
study period | case period but not control period | control period but not case period | OR* | 95%CI | OR+ | 95%CI | OR$ | 95%CI | |
Oral | |||||||||
celecoxib | 210 | 81 | 55 | 1.47 | 1.05 - 2.07 | 1.42 | 0.99 - 2.04 | 1.36 | 0.95 - 1.96 |
non-selective NSAIDs | |||||||||
ketorolac | 76 | 43 | 11 | 3.91 | 2.02 - 7.58 | 2.02 | 1.00 - 4.09 | ||
flurbiprofen | 129 | 54 | 29 | 1.86 | 1.19 - 2.92 | 1.71 | 1.06 - 2.74 | ||
ibuprofen | 755 | 278 | 178 | 1.56 | 1.29 - 1.89 | 1.45 | 1.19 - 1.76 | ||
sulindac | 242 | 93 | 61 | 1.53 | 1.10 - 2.11 | 1.44 | 1.02 - 2.03 | ||
diclofenac | 1,731 | 624 | 421 | 1.48 | 1.31 - 1.68 | 1.29 | 1.13 - 1.47 | ||
acemetacin | 349 | 136 | 93 | 1.46 | 1.12 - 1.90 | 1.28 | 0.96 - 1.70 | ||
naproxen | 231 | 78 | 56 | 1.39 | 0.99 - 1.96 | 1.26 | 0.88 - 1.81 | ||
piroxicam | 259 | 94 | 76 | 1.24 | 0.91 - 1.67 | 1.25 | 0.92 - 1.70 | ||
indomethacin | 184 | 67 | 44 | 1.52 | 1.04 - 2.23 | 1.22 | 0.82 - 1.83 | ||
ketoprofen | 70 | 28 | 22 | 1.27 | 0.73 - 2.23 | 1.17 | 0.64 - 2.11 | ||
mefenamic acid | 1,070 | 370 | 294 | 1.23 | 1.08 - 1.47 | 1.16 | 0.98 - 1.36 | ||
meloxicam | 327 | 106 | 96 | 1.10 | 0.84 - 1.46 | 0.97 | 0.73 - 1.30 | ||
tiaprofenic acid | 92 | 31 | 28 | 1.11 | 0.66 - 1.85 | 0.91 | 0.52 - 1.57 | ||
ns-NSAIDs overall | 3,573 | 1,175 | 768 | 1.53 | 1.40 - 1.68 | 1.42 | 1.29 - 1.56 | ||
Parenteral | |||||||||
ketorolac | 364 | 245 | 35 | 6.99 | 4.91 - 9.96 | 4.27 | 2.90 - 6.29 | ||
ketoprofen | 125 | 69 | 20 | 3.45 | 2.10 - 5.68 | 2.34 | 1.31 - 4.19 | ||
diclofenac | 70 | 34 | 12 | 2.83 | 1.47 - 5.46 | 1.88 | 0.95 - 3.75 | ||
Parenteral NSAIDs overall | 455 | 330 | 65 | 5.08 | 3.89 - 6.62 | 3.35 | 2.50 - 4.47 |
OR = odds ratio; 95%CI = 95% confidence interval; ns-NSAIDs = non-selective NSAIDs
* Crude analysis for oral drug users excluded those ever used parenteral drugs during study period; same exclusion of oral drugs use for parenteral users.
+ Conditional logistic regression for all patients of celecoxib, oral ns-NSAIDs overall, and parenteral ns-NSAIDs overall and adjusted for potential time-varying confounding variables of all discordant use of antihypertensive agents, angiotensin converting enzyme inhibitors or aldosterone receptor blockers, calcium channel blockers, statins, insulin, sulfonylurea, thiazolidinedione, and aspirin between case and control period.
$ Conditional logistic regression for all 17 drugs included in one model with adjustment of above mentioned time-varying confounding variables.