Does self-reported consumption of heroin correlate with dose of methadone among pregnant people with opioid use disorder?

BACKGROUND: The prevalence of opioid use disorder during pregnancy is increasing in the United States. However, evidence to guide appropriate dosing of pharmacotherapy for the treatment of opioid use disorder, such as methadone, based on self-reported opioid use during pregnancy is limited. OBJECTIVE: This study aimed to examine the relationship between self-reported consumption of heroin and methadone dose among pregnant people with opioid use disorder admitted to the hospital for methadone titration. STUDY DESIGN: This was an analysis of a single-site retrospective cohort of pregnant people admitted for the management of opioid use disorder at an urban, tertiary care center between 2013 and 2020. Patient-reported consumption of heroin was evaluated by clinical staff and described as a self-reported dollar amount of heroin consumed per day. The primary outcome was methadone dose at the time of discharge. The secondary outcome was methadone dose at the time of delivery. Bivariate and linear regression modeling were performed. A priori covariates included in the model were an epoch of admission, Clinical Opioid Withdrawal Scale score at the time of admission, gestational age at admission, and need for readmission because of recurrent opioid use. A sensitivity analysis was performed using propensity score matching. Statistical significance was set at P<.05. RESULTS: Of 100 people admitted during the study period, 53 and 47 individuals met the inclusion criteria for the primary and secondary outcomes, respectively. On bivariate and multivariate linear regression modeling, the self-reported dollar amount of heroin consumed per day was independently associated with the dose of methadone at the time of discharge. For every $10 of heroin consumed, the dosage of methadone increased by 1.3 mg (95% confidence interval, 0.4–2.2). On sensitivity analysis with the use of a propensity score, the self-reported dollar amount of heroin consumed per day was independently associated with the dose of methadone at the time of discharge. There was no significant relationship between self-reported dollar amount of heroin consumed per day and methadone dose at the time of delivery. CONCLUSION: Self-reported dollar amount of heroin consumed per day was independently associated with the dose of methadone at the time of discharge, but not at the time of delivery. These data can be useful in clinical counseling and management of pregnant people with opioid use disorder admitted for initiation of methadone.