Prescription drug abuse is a major challenge in public health, in particular in countries such as France, one of the main consumers of medication in Europe. A single source of information is rarely sufficient to measure such a complex phenomenon. This project entitled MEGADOSE (Monitoring prEscriptiondruG Abuse using DOctor Shopping bEhavior) may complete efficiently the available data from the addictovigilance centres allowing the health authorities (French Drug Agency) to answer to the main general questions (What are the main prescription drug abused in France? What are the national trends in prescription drug abuse? What is the impact of a public health measure on prescription drug abuse). This project constitutes the opportunity of "extending" the application of our tool (the DS method) on a French national database with a minimal risk since this method has already been validated and applied on different geographic areas, on different drugs and at different levels of analysis. The project would also be integrated in routinely as a useful tool included in the addictovigilance monitoring, also allowing a multi-approach approach and the monitoring of policy and practice interventions implemented by the French Drug Agency.

The project MEGADOSE is also in accordance with the main topic of the projects funded by ANSM related to drug misuse in particular: development and identification of systems and tools able to detect earlier misuse phenomenon; analysis of the impact of communications, guidelines, decisions to health authorities and patients

Medicines such as analgesics, opioids substitution drugs, sedatives, hypnotics are increasingly being misuse to induce psychoactive effects or to alter the effects of other consumed drugs, potentially resulting in abuse and dependence (24). Awareness of medicine misuse, especially prescription drugs misuse has been increasing in the past few years and organizations such as the United Nation's International Narcotics Control Board predict that worldwide misuse of prescription drugs will exceed illicit drug use. A major potential part of psychoactive drug misuse is related to drug abuse. Assessment of misuse, abuse and diversion of prescription drugs must address its occult nature and involves fundamentally different monitoring methods compared to other pharmaceutical products. Current best practice is to use multiple detection systems to assess misuse, abuse, and diversion of CNS active drugs by various populations in a timely, sensitive and specific manner (20). In order to detect earlier the prescription drug abuse related to misuse, the investigators have developed a specific method, the doctor shopping method, at the regional level. Doctor shopping is a well-known behavior defined as obtaining prescriptions from several physicians for the same medication at the same time. By cumulating the prescriptions of different physicians during the same period of time, doctor shopping enables patients to be delivered a daily dose superior to that prescribed by each physician on an individual basis. Doctor shopping behavior is regarded as not only a means of obtaining high-dose prescriptions but as the principal method of diversion of prescription medications to street markets.

Using health Insurance Database, the investigators have developed and validated a specific method "the doctor shopping method" providing two indicators: the DSI (Doctor Shopping Indicator expressed in %) and the DSQ (Doctor Shopping Quantity expressed in DDD), reflecting respectively the risk of abuse and the extent of abuse. This previous large work, based along ten years of development and validation through several different studies, has i) been always performed at the regional level on SNIIR-AM (several French areas - Paca Corse, Rhône Alpes, Midi Pyrenées), ii) focused on one product (buprenorphine, clonazepam, methylphenidate) or on one pharmacological class (opioids, benzodiazepines) iii) measured the impact of national prescription monitoring programme (buprenorphine). More recently the investigators have performed a global overview of the extent and risk of abuse of more than 100 psychoactive medications among 11.7 million inhabitants according the doctor shopping method, used as a proxy of drug abuse. Based on this experience, and because there is a need to measure prescription drug abuse, to understand trends over the time and to compare abuse of one product to another, the aim of this project is to provide an up-to-date and nationally consolidated estimation of the risk and the extent of abuse associated with more of 100 psychoactive drugs in the French population, using doctor shopping method on National SNIIR-AM, covering the entire French population (65 million inhabitants). To increase the accuracy of its estimation and its description, the study will take into account relevant factors and determinants of drugs abuse such as i) formulation (dosage, extended or immediate release, route of administration) ii) geographical level iii) time factor (trends analysis between 2010 and 2015).

Prescription drug abuse is a major challenge in public health, in particular in countries such as France, one of the main consumers of psychoactive medications in Europe. A single source of information is rarely sufficient to measure such a complex phenomenon. This project entitled MEGADOSE (Monitoring prEscriptiondruG Abuse using DOctor Shopping bEhavior) may complete efficiently the available data from the addictovigilance centres allowing the health authorities (French Drug Agency) to answer to the main general questions (What are the main prescription drug abused in France? What are the national trends in prescription drug abuse? What is the impact of a public health measures on prescription drug abuse).

Inclusion Criteria:

• patients registered in the SNIIR-AM with at least one reimbursement of psychoactive drugs during the study period.

Exclusion Criteria:

• French population not benefiting from social security

The target population is the whole French population. The SNIIR-AM will be used as the source population. The SNIIR-AM (Systeme National d'Information Inter-Régime de l'Assurance Maladie) is the information system of the French National Health Insurance Scheme, containing individualized, anonymous, and exhaustive data on all health spending reimbursements of the overall French insured population (almost 65 million inhabitants).

The information contained in this database covers the health care services reimbursed including:

• Characteristics of medication dispensed (name, formulation, dates of prescription, date of dispensing, quantity dispensed and cost)
• Patients characteristics including age, gender, area of residence, type of affiliation, dependency on the universal health care coverage or CMU (Couverture Médicale Universelle Complémentaire, CMUC, which is attributed to the unemployed and low income insurees, and which can be used as a proxy of the income level),