UNLABELLED: Computerized hospital databases are used for clinical and economic research. In France, the hospital administrative database, Programme de médicalisation des systèmes d'information (PMSI), could be an interesting means for identifying cases of abuse and dependence in hospitals. PURPOSE: To assess the capability of PMSI to identify cases of abuse and dependence (medicines or illicit drugs; tobacco and alcohol not included). METHODS: Cross-sectional study, from October 1 to December 31, 2008, in teaching hospitals of Bordeaux. All hospitalizations with an ICD-10 code related to possible abuse or dependence were selected. Cases were validated by a committee composed of three pharmacologists using discharge summaries. RESULTS: Among the 34 816 patients registered in the PMSI during the study period, a total of 227 patients were pre-selected as potential cases; 21 patients, hospitalized for abuse or dependence, or complications of which, were included in the analysis. Mean age was 35 years. Substances implicated were buprenorphine (n = 8), benzodiazepines (n = 7), cannabis (n = 6), cocaine (n = 4), heroin (n = 3), amphetamine, ecstasy, morphine, codeine, and tramadol (n = 1, respectively); there was polydrug use in six cases. CONCLUSIONS: The PMSI database can be useful to identify certain cases of abuse and dependence. This pilot study has been conducted at a local level; as the PMSI is available in all hospital settings in France, further analysis could be done at the regional and national levels. Such data could be a valuable indicator to analyze trends and assess the medical consequences of substance abuse.
UNLABELLED: Computerized hospital databases are used for clinical and economic research. In France, the hospital administrative database, Programme de médicalisation des systèmes d'information (PMSI), could be an interesting means for identifying cases of abuse and dependence in hospitals. PURPOSE: To assess the capability of PMSI to identify cases of abuse and dependence (medicines or illicit drugs; tobacco and alcohol not included). METHODS: Cross-sectional study, from October 1 to December 31, 2008, in teaching hospitals of Bordeaux. All hospitalizations with an ICD-10 code related to possible abuse or dependence were selected. Cases were validated by a committee composed of three pharmacologists using discharge summaries. RESULTS: Among the 34 816 patients registered in the PMSI during the study period, a total of 227 patients were pre-selected as potential cases; 21 patients, hospitalized for abuse or dependence, or complications of which, were included in the analysis. Mean age was 35 years. Substances implicated were buprenorphine (n = 8), benzodiazepines (n = 7), cannabis (n = 6), cocaine (n = 4), heroin (n = 3), amphetamine, ecstasy, morphine, codeine, and tramadol (n = 1, respectively); there was polydrug use in six cases. CONCLUSIONS: The PMSI database can be useful to identify certain cases of abuse and dependence. This pilot study has been conducted at a local level; as the PMSI is available in all hospital settings in France, further analysis could be done at the regional and national levels. Such data could be a valuable indicator to analyze trends and assess the medical consequences of substance abuse.