Table 1.
Selected baseline and time-varying characteristics of buprenorphine-treated, methadone-treated, and untreated pregnancies with opioid use disorder (OUD).
| Exposure early in pregnancy (0-20 gestational weeks)a,b | Exposure late in pregnancy (>20 gestational weeks)a,b | |||||||
|---|---|---|---|---|---|---|---|---|
| Characteristics | Untreated (N=178) | Buprenorphine (N=85) | Methadone (N=137) | p-value | Untreated (N=184) | Buprenorphine (N=72) | Methadone (N=144) | p-value |
| Maternal Age, years (mean, SD) | 28.04 (5.26) | 30.01 (5.31) | 29.64 (4.32) | 0.002 | 28.18 (5.50) | 30.18 (4.80) | 29.48 (4.34) | 0.0060 |
| Maternal Age, years (n, %) | ||||||||
| <20 | <11 | <11 | <11 | <.0001 | <11 | <11 | <11 | 0.0373 |
| 20-34 | 141 (79.21) | 59 (69.41) | 113 (82.48) | 143 (77.72) | 52 (72.22) | 118 (81.94) | ||
| >34 | 30 (16.85) | 24 (28.24) | 22 (16.06) | 32 (17.39) | 20 (27.78) | 24 (16.67) | ||
| Race, (n, %) | ||||||||
| Black | 15 (8.43) | <11 | <11 | <.0001 | 14 (7.61) | <11 | <11 | <.0001 |
| Other | 32 (17.98) | <11 | 23 (16.79) | 36 (19.57) | <11 | 22 (15.28) | ||
| White | 131 (73.60) | 72 (84.71) | 109 (79.56) | 134 (72.83) | 62 (86.11) | 116 (80.56) | ||
| Birth year, (n, %) | ||||||||
| 2008-2011 | 63 (35.39) | 13 (15.29) | 40 (29.20) | 0.0035 | 70 (38.04) | <11 | 40 (27.78) | <.0001 |
| 2012-2016 | 115 (64.61) | 72 (84.71) | 97 (70.80) | 0.0035 | 114 (61.96) | 66 (91.67) | 104 (72.22) | <0.001 |
| Multifetal gestation, (n, %) | <11 | <11 | <11 | 0.1098 | <11 | <11 | <11 | 0.0137 |
| Infant gender, boy (n, %) | 85 (47.75) | 44 (51.76) | 67 (48.91) | 0.8506 | 89 (48.37) | 35 (48.61) | 72 (50.00) | 0.9554 |
| Pre-existing comorbidities, (n, %) | ||||||||
| Depression | 56 (31.46) | 31 (36.47) | 32 (23.36) | 0.0923 | 60 (32.61) | 24 (33.33) | 35 (24.31) | 0.2015 |
| Anxiety/PTSD | 62 (34.83) | 32 (37.65) | 38 (27.74) | 0.2446 | 65 (35.33) | 27 (37.50) | 40 (27.78) | 0.2362 |
| Healthcare resource utilization at baseline, (n, %) | ||||||||
| Number of outpatient visits (all-cause), (mean, SD) | 13.15 (18.69) | 12.25 (9.05) | 54.23 (80.26) | <.0001 | 12.93 (15.23) | 12.65 (9.17) | 52.22 (79.67) | <.0001 |
| Number of inpatient visits (all-cause), (mean, SD) | 1.09 (2.14) | 2.61 (3.89) | 2.01 (3.14) | <.0001 | 1.04 (1.85) | 2.89 (4.23) | 2.03 (3.22) | <.0001 |
| Use of substances at 3-month baseline, (n, %) | ||||||||
| Tobacco use disorder/abuse | 20 (11.24) | 11 (12.94) | 20 (14.60) | 0.6735 | 19 (10.33) | 11 (15.28) | 21 (14.58) | 0.4024 |
| Alcohol use disorder/abuse | 14 (7.87) | <11 | <11 | 0.0234 | 14 (7.61) | <11 | <11 | 0.0306 |
| Substance use disorder/abuse | 44 (24.72) | 19 (22.35) | 12 (8.76) | 0.0010 | 48 (26.09) | 14 (19.44) | 13 (9.03) | 0.0004 |
| Use of substances early in pregnancy, (n, %) | ||||||||
| Tobacco use disorder/abuse | 28 (15.22) | 11 (15.28) | 22 (15.28) | 0.9999 | ||||
| Alcohol use disorder/abuse | 14 (7.61) | <11 | <11 | 0.0234 | ||||
| Substance use disorder/abuse | 49 (26.63) | 15 (20.83) | 23 (15.97) | 0.0660 | ||||
| Concomitant medication uses at 3-month baseline, (n, %) | ||||||||
| Benzodiazepines | 33 (18.54) | 20 (23.53) | 33 (24.09) | 0.4327 | 39 (21.20) | 16 (22.22) | 31 (21.53) | 0.9839 |
| Antidepressants | 42 (23.60) | 32 (37.65) | 34 (24.82) | 0.0436 | 45 (24.46) | 27 (37.50) | 36 (25.00) | 0.0853 |
| Opioid analgesics | 45 (25.28) | 15 (17.65) | 23 (16.79) | 0.0019 | 48 (26.09) | 12 (16.67) | 23 (15.97) | 0.0008 |
| Anticonvulsants | 20 (11.24) | <11 | <11 | 0.4975 | 18 (9.78) | <11 | 13 (9.03) | 0.9711 |
| Concomitant medication uses early in pregnancy, (n, %) | ||||||||
| Benzodiazepines | 35 (19.02) | 19 (26.39) | 33 (22.92) | 0.4005 | ||||
| Antidepressants | 42 (22.83) | 27 (37.50) | 34 (23.61) | 0.0415 | ||||
| Opioid analgesics | 31 (16.85) | <11 | 12 (8.33) | 0.0006 | ||||
| Anticonvulsants | 17 (9.24) | <11 | <11 | 0.6992 | ||||
Abbreviation: SD = standard deviation.
a
Small cell count <11 was suppressed.
b
Markers of severity of OUD (including injection drug use-related infection, opioid-related overdose, and hepatitis C virus infection) at baseline and early in pregnancy were included as time-varying covariates in models for inverse probability treatment weighting; however, descriptive statistics were not reported due to small counts (i.e., <11).