Go back or Close Window
7A.1.2 Treatment response after 12 months among the treatment completers

The 138 treatment completers in the trial on injectable heroin - consisting of patients who still participated in treatment (i.e. received methadone in the control condition, and co-prescribed heroin in the experimental condition) in month 12 - involved 83 subjects in the methadone alone treatment group, and 55 subjects in the co-prescribed heroin group (see Chapter 5A). Analogous to the primary outcome analysis in the intention-to-treat population - applying LOCF for the missing endpoint-assessments - the percentage responders after 12 months was determined among the treatment completers. The response rate among the treatment completers in the co-prescribed heroin group amounted to 58.2%, compared to 34.9% in the methadone alone group. The difference of 23.3% was statistically significant in a good fitting logistic regression model, in which "treatment site" was the only covariate, and in which neither treatment site, nor treatment-by-site interaction were significantly related to response (adjusted OR=2.95; 95%-CI: 1.41-6.18; p=0.00041).

  • After twelve months of treatment, the experimental treatment with co-prescribed injectable heroin was significantly more effective than the control treatment with methadone alone. The observed difference in percentage responders between the experimental group and the control group amounted to 25.0%.

  • The observed difference in percentage responders decreased from 25.0% to 15.5% when a worst-case scenario was applied for the missing endpoint-assessments. However, the difference was still statistically significant.

  • The observed difference found after twelve months of treatment was also statistically significant two months earlier, after ten months of treatment. Hence, anticipation effects, with regard to a possible reinstatement of the heroin treatment after the experimental phase for responders in the heroin condition, can not be held responsible for the observed effect after twelve months of treatment.

  • In the sub-population of treatment completers, the difference in percentage responders between the treatment groups was similar to the difference observed in the intention-to-treat population.

  • The observed difference in percentage responders between the treatment groups after twelve months of treatment was not significantly different across the six participating sites.