Skip to main content
. 2022 Sep 19;182(11):1117–1127. doi: 10.1001/jamainternmed.2022.3680

Table 2. Association Between Perioperative Gabapentin Use and In-Hospital Adverse Clinical Events After Major Surgery Before and After Propensity Score Matching.

Outcome Before propensity score matching After propensity score matchinga
Gabapentin use, No. (%) (n = 119 087) No gabapentin use, No. (%) (n = 848 460) RR (95% CI) RD, per 100 persons (95% CI) Gabapentin use, No. (%) (n = 118 936) No gabapentin use, No. (%) (n = 118 936) RR (95% CI) RD, per 100 persons (95% CI)
Delirium diagnosis 4051 (3.4) 34 342 (4.0) 0.84 (0.81 to 0.87) –0.6 (–0.8 to –0.5) 4040 (3.4) 3148 (2.6) 1.28 (1.23 to 1.34) 0.75 (0.61 to 0.89)
New antipsychotic use 945 (0.8) 9877 (1.2) 0.68 (0.64 to 0.73) –0.4 (–0.4 to –0.3) 944 (0.8) 805 (0.7) 1.17 (1.07 to 1.29) 0.12 (0.05 to 0.19)
Pneumonia 1522 (1.3) 19 902 (2.3) 0.54 (0.52 to 0.57) –1.1 (–1.1 to –1.0) 1521 (1.3) 1368 (1.2) 1.11 (1.03 to 1.20) 0.13 (0.04 to 0.22)
In-hospital death 363 (0.3) 6360 (0.7) 0.41 (0.37 to 0.45) –0.4 (–0.5 to –0.4) 362 (0.3) 354 (0.2) 1.02 (0.88 to 1.18) 0.00 (–0.04 to 0.05)

Abbreviations: RD, risk difference; RR, risk ratio.

a

The propensity score model included demographic information, insurance type, admission characteristics, surgery type, combined comorbidity score, comorbidities, inpatient medication use and procedures before or on postoperative day 2, hospital-level characteristics, geographic region, and calendar year.