Abstract
OBJECTIVE: To evaluate temporal trends in medication-assisted treatment use among pregnant women with opioid use disorder. METHODS: We conducted a retrospective cohort study using Pennsylvania Medicaid administrative data. Trends in medication-assisted treatment use, opioid pharmacotherapy (methadone and buprenorphine) and behavioral health counselling, were calculated using pharmacy records and procedure codes. Cochrane-Armitage tests evaluated linear trends in characteristics of pregnant women using methadone compared with buprenorphine. RESULTS: In total, we evaluated 12,587 pregnancies among 10,741 women with opioid use disorder who had a live birth between 2009 and 2015. Across all years, 44.1% of pregnant women received no opioid pharmacotherapy, 27.1% used buprenorphine, and 28.8% methadone. Fewer than half of women had any behavioral health counseling during pregnancy. The adjusted prevalence of methadone use declined from 31.6% (95% CI 29.3-33.9%) in 2009 to 25.2% (95% CI 23.3-27.1%) in 2015, whereas the adjusted prevalence of buprenorphine use increased from 15.8% (95% CI 13.9-17.8%) to 30.9% (95% CI 28.8-33.0%). Greater increases in buprenorphine use were found in geographic regions with large metropolitan centers, such as the Southwest (plus 24.9%) and the Southeast (plus 12.0%), compared with largely rural regions, such as the New West (plus 5.2%). In 2015, the adjusted number of behavioral health counseling visits during pregnancy was 3.4 (95% CI 2.6-4.1) among women using buprenorphine, 4.0 (95% CI 3.3-4.7) among women who did not use pharmacotherapy, and 6.4 (95% CI 4.9-7.9) among women using methadone. CONCLUSION: Buprenorphine use among Medicaid-enrolled pregnant women with opioid use disorder increased significantly over time, with a small concurrent decline in methadone use. Behavioral health counseling use was low, but highest among women using methadone.