[Dangerous states and mental health disorders: perceptions and reality].

Image of Madness was always strongly linked with the notion of "dangerousness", provoking fear and social exclusion, despite the evolution of psychiatric practices and organisation, and the emphasis on user's rights respect. Mediatization and politicization of this issue through news item combining crime and mental illness, reinforce and spread out this perception. This paper presents a review of the litterature on social perceptions associating "dangerousness", "Insanity" and "mental illness", available data about the link between "dangerous states" and "psychiatric disorders", as well as the notion of "dangerousness" and the assessment of "dangerous state" of people suffering or not from psychiatric disorders. MAPPING OF SOCIAL REPRESENTATIONS: The French Survey "Mental Health in General Population: Images and Realities (MHGP)" was carried out between 1999 and 2003, on a representative sample of 36.000 individuals over 18 years old. It aims at describing the social representations of the population about "insanity/insane" and "mental illness/mentally ill". The results show that about 75% of the people interviewed link "insanity" or "mental illness" with "criminal or violent acts". Young people and those with a high level of education more frequently categorize violent and dangerous behaviours in the field of Mental illness rather than in that of madness. CORRELATION BETWEEN DANGEROUS STATE AND PSYCHIATRIC DISORDERS: in the scientific literature, all experts reject the hypothesis of a direct link between violence and mental disorder. Besides, 2 tendencies appear in their conclusions: on one hand, some studies establish a significative link between violence and severe mental illness, compared with the general population. On the other hand, results show that 87 to 97% of des aggressors are not mentally ills. Therefore, the absence of scientific consensus feeds the confusion and reinforce the link of causality between psychiatric disorders and violence. OFFICIAL FIGURES BY THE MINISTRY OF JUSTICE: according to the French Ministry of Justice, there is a lack of significative data in general population, that would allow the accurate evaluation of the proportion of authors of crimes and offences presenting a "dangerous state", either of criminological order or related to a psychiatric disorder. FROM "DANGEROUSNESS" TO "DANGEROUS STATE": the vagueness of the notion of "dangerousness" aggravates the confusion and reinforce the negative social representations attached to subjects labelled as "mentally ills". A way to alleviate this stigmatisation would be to stop using the word "dangerous", and rather use those of "dangerous states". Assessment of dangerous states is complex and needs to take into account several heterogeneous factors (circumstances of acting, social and family environment...). Besides, it is not a linear process for a given individual. Those risk factors of "dangerous state" lead to the construction of evaluation or prediction scales, which limits lay in the biaises of over or under predictive value. The overestimation of dangerousness is harmful, not only to individuals wrongly considered as "dangerous", but also to the society which, driven by safety concerns, agrees on the implementation of inaccurate measures. A FEW TRACKS FOR REMEDIATION: the representations linking "mental illness" and "dangerousness" are the major vectors of stigma, and deeply anchored in the collective popular imagination. They are shared by all population categories, with no distinction of age, gender, professional status or level of education. To overcome those prejudices, one has to carefully study their basis, their criteria, document them with statistical data, look for consistency and scientific rigour, in the terminology as well as in the methodology. Moreover, one has to encourage exchanges about this topic, between users, relatives, carers, local elected, politicians, media and health professional.