[Quality of the diagnoses' coding (main or associated diagnoses) in the medico-administrative database for psychiatric care (RIM-P) in 2015 and 2016, France]. / Qualité du codage des diagnostics et motifs de prise en charge (principal et associés) dans le recueil d'informations médicalisé en psychiatrie (RIM-P) en 2015 et 2016, France.

BACKGROUND: Based on the observation of the misuse of ICD-10 to code the diagnoses in the RIM-P (lack of completeness, conformity and diversity), the Technical Agency for information on Hospital Care (ATIH), which provides tools for collecting medical information, conducted two actions in 2016. First, a chapter devoted to the instructions of coding has been written in the methodological guide of production of the RIM-P, second, a variable "type psy" was added to the ICD-10 nomenclature's file framing ICD-10 coding in the RIM-P. The purpose of this study is to describe the quality of diagnosis coding using ICD-10 in the RIM-P in 2015 and 2016. METHODS: The quality of diagnosis coding using ICD-10 in the summaries of activity of the RIM-P national databases was described in 2015 and 2016. The study focused on the completeness, the conformity and the diversity of coding. RESULTS: Between 2015 and 2016, the percentage of summaries without primary diagnosis ("DP") decreased slightly for full-time (5.2% vs. 3.8%), part-time (6.3% vs. 4.9%) inpatient stays and outpatient care (9.9% vs. 8.9%). ICD-10 codes used to code DP or associated diagnosis ("DA"), while prohibited, mainly belong to Chapter V Mental and behavioral disorders. Per year, only one-third of the summaries and one-half of patients had two or more ICD-10 codes reported for inpatient stays (one-fifth of the summaries and one-fourth of the patients for outpatient care). In addition, per year and per facility, the average number of distinct ICD-10 codes used to fill "DP" or "DA" was approximately half as important in part-time hospitalization, as in full-time hospitalization or for outpatient care. Moreover, 90% of the health facilities used<550 distinct ICD-10 codes in full-time inpatient stays,<270 in part-time inpatient stays and<950 for outpatient care to code the "DP" or the "DA". The diversity of ICD-10 codes used was low and similar between 2015 and 2016, especially to describe the socio-economic environment, resistance to treatment or non-compliance. CONCLUSION: This study emphasizes the need for a collective effort to improve the diversity of the diagnoses' coding in the RIM-P.