The management of patients with opioid addiction is a challenge insofar as many distractors or variability factors can interfere with the control of the addiction, whether they are psychological, psychiatric, environmental, pharmacokinetic or pharmacodynamic. Understanding this variability is potentially to be able to adjust a priori a dosage and to identify the factors of clinical response. Few population pharmacokinetic models exist for methadone and they generally concern the management of pain in palliative care patients or the management of opioid withdrawal syndrome in neonates. The hypothesis is therefore that the creation of such a model would make it possible to reduce patients' withdrawal periods, to set a target for plasma concentrations with a view to reducing dosages, and to empower the patient in his choice to monitor blood concentrations facilitated by a minimally invasive sampling device.

Inclusion Criteria:

• Major patients on methadone on the same dosage for at least 7 days.
• Hospitalization for withdrawal in the addictology department of La Colombière or consultation within the unit for the treatment of drug addiction and dependence (Addictology Department).
• 24 hours monitoring is possible and accepted

Exclusion Criteria:

• Patients that cannot be sampled because of weakened veins
• Unable to received information about the study
• Exclusion period determined by previous study
• Adult protected by law or patient under guardianship
• Not affiliated to french social security system
• Not able to give written inform consent
• Pregnant or breastfeeding woman
• Underaged patients (under 18 years old)
• Patient under court protection