Table 2.
Unadjusted and IP-weighted hazard ratios of pregnancy and obstetric outcomes
| Outcome | Primary analyses |
Secondary analyses |
||
|---|---|---|---|---|
| Events occurred 1 week after opioid exposure |
Events occurred within 1 week of opioid exposure |
|||
| Unadjusted HR (95% CI) | IP-weighted HR (95% CI) | Unadjusted HR (95% CI) | IP-weighted HR (95% CI) | |
|
| ||||
| Cesarean delivery, Yes vs No | 2.10 (1.32–3.35) | 1.70 (0.78–3.67) | 1.99 (1.23–3.21) | 1.68 (0.74–3.83) |
| Neuraxial anesthesia during labor vs None/local anesthesiaa | 1.53 (1.14–2.07) | 1.41 (0.93–2.14) | 1.54 (1.14–2.07) | 1.40 (0.92–2.14) |
| Operative vaginal delivery, Yes vs No | 1.67 (1.08–2.59) | 1.68 (0.81–3477) | 1.59 (1.02–2.47) | 1.64 (0.76–3.54) |
| Induction of delivery, Yes vs No | 1.17 (0.43–3.14) | 0.50 (0.16–1.59) | 1.17 (0.44–3.14) | 0.56 (0.17–1.79) |
| Prolonged labor, Yes vs No | 1.60 (0.76–3.40) | 1.13 (0.41–3.15) | 1.61 (0.76–3.41) | 1.15 (0.40–3.30) |
| Placental abruption, Yes vs No | 3.65 (0.90–14.80) | 2.12 (0.49–26.19) | 3.66 (0.90–14.82) | 2.07 (0.48–9.05) |
| Premature rupture of membranes, Yes vs No | 1.05 (0.34–3.26) | 0.80 (0.24–2.73) | 0.70 (0.17–2.81) | 0.47 (0.11–2.05) |
| Antepartum hemorrhage, Yes, vs No | 1.23 (0.39–3.84) | 0.98 (0.34–2.73) | 0.82 (0.20–3.30) | 0.61 (0.14–2.64) |
| Postpartum hemorrhage, Yes vs No | 2.01 (0.51–7.96) | 4.27 (0.69–26.19) | 2.01 (0.51–7.97) | 4.19 (0.64–27.35) |
| Antepartum depression, Yes vs No | 1.95 (0.98–3.88) | 2.73 (0.96–7.68) | 2.20 (1.15–4.20) | 3.19 (1.22–8.33) |
| Cardiac events, Yes vs No | 6.10 (0.84–44.39) | 3.25 (0.43–24.81) | 6.10 (0.84–44.49) | 9.44 (1.19–74.83) |
| Preterm birth, Yes vs No | 0.89 (0.29–2.78) | 0.55 (0.16–1.92) | 0.89 (0.29–2.79) | 0.53 (0.15–1.85) |
Maternal opioid exposure compared with unexposed pregnancies in primary analyses and secondary analyses using time-varying exposure and covariates weighted by propensity score
Adjusted for previous exposure history, baseline covariates (i.e., time interval and polynomial terms, maternal age, quadratic form of maternal age, race, year of birth, multiple gestation, pre-existing comorbid conditions) in the numerator models, and previous exposure history, baseline covariates, time-varying pain indications and comedication use in the denominator models
CI confidence interval, HR hazard ratio, IP inverse probability
Anesthesia method at delivery is missing for 23 (0.2%) and only complete cases were analyzed. Neuraxial anesthesia during labor includes spinal or epidural anesthesia