Substance Use
Buprenorphine in Substance Use Disorder (SUD) Specialty Settings versus Non-SUD Settings: Trial Emulation using the Veterans Health Administration Data Haidong Lu* Haidong Lu Thomas Thornhill Sandra Springer Ismene Petrakis Gregg Gonsalves
Background: Office-based visits involving buprenorphine prescriptions have increased significantly among non-substance use disorder (SUD) specialty and primary care clinics since 2006. We investigated whether buprenorphine prescribing in non-SUD settings and SUD specialty settings has comparable effects on retention in treatment and all-cause mortality among veterans.
Methods: We emulated a target trial using electronic health records of veterans with opioid use disorder (OUD) in the Veterans Health Administration (VHA) who initiated buprenorphine in either VHA SUD specialty clinics or non-SUD clinics between 2006 and 2019. The primary outcome was treatment discontinuation, defined as a gap of more than 30 days between the end of one prescription and the subsequent dose. The second outcome was all-cause mortality. We estimated observational analogs of intention-to-treat risk differences (RDs) and hazard ratios (HRs) while accounting for baseline confounding including patient sociodemographic factors and co-morbidities.
Results: Of 33,107 veterans meeting the eligibility criteria, 8,508 (25.7%) initiated buprenorphine treatment in non-SUD clinics, and 24,599 (74.3%) initiated treatment in SUD specialty clinics. The estimated 1-year intention-to-treat risks of buprenorphine discontinuation were 70.6% in non-SUD clinics and 63.7% in SUD clinics, resulting in a RD of 6.9% (95% CI: 5.6% to 8.1%) and a HR of 1.22 (1.18-1.26). The estimated 3-year risks of buprenorphine discontinuation were 85.9% in non-SUD clinics and 82.3% in SUD clinics, yielding a RD of 3.6% (2.6% to 4.6%) and a HR of 1.19 (1.15-1.23). The estimated 3-year intention-to-treat risks of all-cause mortality were 8.1% in non-SUD clinics and 6.9% in SUD clinics, resulting in a RD of 1.2% (0.5% to 1.9%) and a HR of 1.18 (1.07-1.29).
Conclusions: Buprenorphine prescriptions in non-SUD VHA clinics were associated with a higher risk of treatment discontinuation and all-cause mortality compared with those in SUD VHA clinics. However, both groups showed low retention on buprenorphine treatment.