In Reply We appreciate the letter from Bradley and Williams, which highlights 2 issues. We wish to clarify that the primary outcome is abstinence from both alcohol and opioids. We recognize the way it was presented in our Original Investigation1 was confusing and wish to correct it.
With regards to the diagnostic composition of the sample in our study,1 94% of the sample had an alcohol use disorder. Fifty-four percent had an alcohol use disorder alone, and 40% had both an alcohol use disorder and an opioid use disorder. Six percent had only an opioid use disorder. Among the alcohol-only group, the collaborative care intervention was significantly associated with both primary outcomes and with the Healthcare Effectiveness Data and Information Set initiation and engagement measures, abstinence from alcohol, opioids, cocaine, methamphetamines, and marijuana and is borderline significant for no heavy drinking in the past 30 days. Among the any opioids group, the collaborative care intervention was associated with receiving evidence-based treatment, and specifically brief therapy but was not associated with abstinence. These results suggest that the collaborative care intervention was effective for individuals with alcohol use disorders that were not comorbid with opioid use disorders and worked by increasing access to brief therapy rather than medication-assisted treatment. They also suggest that the increased brief therapy treatment access that resulted from the collaborative care intervention was either insufficient or ineffective in improving outcomes for individuals with opioid use disorders, relative to usual care. Our findings are consistent with policy recommendations that have emphasized increased access to medication-assisted treatment as a key approach for addressing the opioid epidemic,2,3 and underscore the importance of considering ways to engage individuals with opioid use disorders in medication assisted treatment.