Background: Although methadone maintenance treatment (MMT) is recognized as the treatment of reference for opioid dependence, little information is available regarding the dynamic adherence to methadone and its determinants. With data from the Methaville trial we investigated the evolution of non-adherence to methadone and the effect of pre-treatment and in-treatment factors on longterm non-adherence to methadone.
Methods: We selected 145 patients to study adherence to methadone at 3, 6 and 12 months (M3, M6 and M12, respectively) using a multidimensional questionnaire and a 3-level variable "adherent"/"non adherent"/"highly nonadherent". We then identified the pre-treatment and in-treatment variables associated with long-term non-adherence to methadone at the M12 visit using a univariate logistic regression and two different multivariate models: the first incorporating only the pre-treatment variables, the second adding the in-treatment variables to the pre-treatment ones.
Results: At the M12 visit, 35.2% of the participants remained adherent, 55.9% and 9% were non-adherent and highly non-adherent, respectively. The multivariate analysis of long-term non-adherence to methadone showed 4 pre-treatment predictors and 1 in-treatment predictor as follows: being female, not having stable housing, alcohol consumption, cocaine use and perceiving methadone dose as inadequate.
Conclusions: Our findings highlight that pretreatment predictors are important to consider when starting maintenance treatment for opioid dependence, such as cocaine use and problematic alcohol consumption but also low socio-economic levels. In addition, during maintenance treatment, in-treatment predictors such as methadone dose adequacy is a crucial issue to achieve good adherence to MMT.