[Differences in psychiatric expertise of responsibility for schizophrenic persons accused of murder: Study with experts of the Court of Appeal of Aix-en-Provence].

INTRODUCTION: In France, forensic psychiatric assessment plays a central role in the relationship between psychiatry and justice. The psychiatric expert is commissioned to determine whether or not the accused has a mental disorder and to specify whether or not it affected discernment at the time of offense. Nowadays, psychiatric expertise is coming under more and more criticism, particularly regarding divergences between experts.

OBJECTIVES: Our objectives were to find points of divergence between experts, try to understand causes and suggest ways to try to reduce them.

METHODS: For this we conducted a study, between July 2012 and January 2013, with psychiatric experts of the Court of Appeal of Aix-en-Provence through semi-structured interviews. We focused on a limited context: psychiatric expertise of responsibility for schizophrenic persons accused of murder. We questioned the experts about the issue of criminal liability of a person with schizophrenia in general but also in clinical situations we thought particularly involved in disagreements.

RESULTS: We recruited a population of 17 psychiatrists, mostly males of average age of 58 years, working mostly in the department of adult psychiatry of a hospital. We highlighted the differences between the experts, first with regards to the issue of liability in general. Experts divided seemed to keep in majority (52.9 %) the alternative between abolition and alteration of discernment when faced with a schizophrenic person accused of murder. The differences were even more pronounced in specific contexts. Thus, the fact that the person had suffered from delirium at the time of the offense led half of the experts (47.1 %) to conclude a systematic abolition of discernment, while the other half made such a conclusion when the delirium was directly linked to the facts. Discontinuation of neuroleptic treatment, drug abuse or existence of premeditation changed the conclusions of the experts in half the cases, more in the sense of an increased accountability in the cases of drug abuse or premeditation, and in the direction of a reduction of liability in case of cessation of treatment. The denial of facts by the accused caused fewer disagreements between experts. Among experts, 76.5 % had already observed differences, which, according to them, were based primarily on schools of thought, or personal views (64.7 %), which could distort clinical evaluation and especially forensic interpretation of the relationship between pathology and facts. The experts thought it was possible and desirable to reduce differences and proposed different solutions for this, especially the return to dual expertise and colleges of experts.

DISCUSSION: Our results were consistent with those in the literature. Based on proposals from experts and data from the literature, we identified five perspectives likely to reduce differences: first it would be useful to put in place a better specific training in forensic psychiatry and expertise, not only theoretical but also in terms of practical training through tutoring. We would identify a jurisprudence in forensic psychiatric assessment and identify consensual points. It would be good to allow experts to acquire sufficient experience not just through tutorials but also by statutory changes. Moments of exchange between experts - including a return to dual expertise and the organization of work meeting - could also reduce differences. Finally, we propose legislative changes: not only to rewrite the paragraph 2 of Article 122-1 of the French Penal Code, but also to give priority to the expertise of responsibility on the expertise of dangerousness.

CONCLUSION: We showed that there were differences between the experts mainly concerned with the forensic interpretation, and that they seemed linked to schools of thought or to personal views of each expert. To reduce the differences, we discussed five perspectives.