Patients receiving buprenorphine for opioid use disorder are typically also referred for psychotherapy, either individual or group. However, several studies have failed to show that therapy improves outcomes in buprenorphine treatment program

Review of: Moore B et al. Cognitive behavioral therapy improves treatment outcomes for prescription opioid users in primary care buprenorphine treatment. Journal of Substance Abuse Treatment 2016;71:54–57.

Study type: Secondary analysis of a randomized clinical trial

Patients receiving buprenorphine for opioid use disorder are typically also referred for psychotherapy, either individual or group. However, several studies have failed to show that therapy improves outcomes in buprenorphine treatment programs (Amato L et al, Cochrane Database Syst Rev 2011;(10):CD004147.doi:10/1002/14651858.pub4).

In spite of previous negative results, Moore and colleagues theorized that patients with opioid use disorder who were primarily prescription opioid users might respond differently to CBT than those who were primarily heroin users. To test this hypothesis, they reviewed a previous randomized clinical trial by Feillin et al (Am J Med 2013;126(1):74e11–17) in which 141 patients with opioid use disorder being treated in a primary care clinic were randomized to physician management (PM) alone or physician ­management plus CBT (PM-CBT). ­Self-reported opioid use and urine drug tests were obtained weekly for 24 weeks, and primary outcomes were use of illicit opioids and weeks of abstinence. Feillin’s study failed to detect a difference between the two groups. Moore and colleagues reanalyzed those data by further subdividing the subjects into those who primarily used prescription opioids and those who primarily used heroin, something the original study had not done.

Results
When subdivided, there were 23 PM-CBT and 26 PM-only prescription opioid users, and there were 47 PM-CBT and 44 PM-only heroin users. When the authors reanalyzed the data based on this new classification, once again there was no significant difference between the groups in the number of urines negative for opiates. However, subjects who used primarily prescription opioids and received both CBT and PM had more urines negative for all drugs (p = .04) than those who received PM only. In addition, the subjects who used primarily prescription opioids and received PM-CBT also appeared to have more consecutive weeks of opioid abstinence, but this difference was not statistically significant. On the other hand, assignment to PM or PM-CBT made no difference in either outcome among subjects who used primarily heroin. Based on demographic data, the prescription opioid users had fewer years of opioid use, were less likely to report past treatments, and were less likely to report IV drug use compared with the group that primarily used heroin.

CATR’s Take
The small number of subjects, particularly in the prescription opioid user group, is a strong limiting factor in this study. There was a significant dropout rate (around 50%) for all groups, making the numbers in the final analysis even smaller. This limits the power of the study to detect true differences between the groups. Subjects were classified as primarily prescription opioid or heroin users based on self-reporting, which, as we know, may not be totally reliable. The most important finding of the study, that the prescription opioid PM-CBT group had more urines negative for all drugs, is only modestly significant. A larger randomized controlled trial is needed before any reliable conclusions can be drawn.

Practice Implications
Adding CBT to buprenorphine treatment seems to confer no additional benefit for heroin users. However, rather than throwing out CBT altogether, we should consider whether CBT can help prescription opioid users—whose addiction may be less severe—abstain from opioids longer.