Poster presented at the College on Problems of Drug Dependence (CPDD) annual meeting, Palm Springs, CA, June 11-16, 2016.
Abigail Matthews, PhD (Clinical Coordinating Center, EMMES), Maureen Hillhouse, PhD (University of California, Los Angeles, PR Node), Christie Thomas, MPH (University of California, Los Angeles, PR Node), Walter Ling, MD (University of California, Los Angeles, PR Node), Dikla Shmueli-Blumberg, PhD (Clinical Coordinating Center, EMMES).
The use of blinding in trials is an established element of study design, intended to minimize bias and expectation effects and strengthen the internal validity of the results. The goal of the current analysis was to assess participants' perceptions of their blinded treatment assignment and examine whether these perceptions were associated with the primary outcome results of the trial. Perceived treatment assignment was evaluated in the NIDA Clinical Trials Network (CTN) trial "Cocaine Use Reduction with Buprenorphine (CURB)" at the the end of active medication. Participants were randomly assigned to 1 of 3 conditions: 16mg buprenorphine+naloxone (BUP16), 4mg (BUP4), placebo (PLB), plus cognitive behavioral therapy and extended-release naloxone (XR-NTX). Data was available for 281/302 participants (93%). 57% of participants had an opinion regarding their assignment and 43% were unsure. Of those who had an opinion, 46% guessed correctly. There was no association with actual treatment group. In the BUP16 arm, 55% guessed correctly, 44% in the BUP4 arm, and 39% in the PLB arm. Perceived treatment assignment was not related to the primary outcome (self-reported cocaine use combined with urine drug screens during last 30 days of medication) or the number of cocaine-negative UDS collected during that period. The fewest average number of days of cocaine use was 6.4 for participants who believed they got BUP16; the greatest was for PLB at 8.0 days. These differences were not significant. There was no difference in the number of cocaine-negative UDS across perceived treatment.
Conclusions: This secondary analysis showed that the blind was maintained in CURB. Participants who speculated about their arm were no more likely to be correct than by chance. This finding increases the confidence in the validity of trial results. Further examination revealed no association with cocaine abstinence. (Poster, PDF, English, 2016)
Keywords: Buprenorphine/Naloxone | Cocaine | CTN platform/ancillary study | Heroin | Naltrexone | Pharmacological therapy | Prescription-type opiates | Research design | College on Problems of Drug Dependence (CPDD) annual meeting, 2016
Document No: 1210.
Submitted by Dikla Blumberg, PhD, Clinical Coordinating Center, 6/29/2016.