Does hospitalisation affect hypnotic and anxiolytic drug prescribing?

OBJECTIVE: The aim of this study was to measure the impact of hospitalisation on hypnotic and anxiolytic (HA) drug prescription, during and after hospitalisation. METHOD: A descriptive study was carried out over three periods: before, during and after hospitalisation (three-month follow-up), examining the presence or absence of HA treatment at each stage. The HA drug list studied was selected using the World Health Organisation (WHO) Anatomical Therapeutic and Chemical (ATC) classification system. Trained final-year pharmacy students asked a series of structured questions during hospitalisation and postal questionnaires were sent to included patients one and three months after discharge. All the in-patient departments in the University Hospital of Besançon-France-were included, except units with pre-, peri-, and post-operative HA treatments. All in-patients present in the selected units on February 12, 2003, aged over 18, who gave their consent and were considered able to answer by the nursing team, were finally included. MAIN OUTCOME MEASURE: An eight-branch descriptive model, including the three study periods with two states (presence or absence of HA treatment) at each stage. RESULTS: A total of 260 in-patients were included, and a further 112 (43%) completed the whole study (alive, non re-hospitalised, one- and three- months post discharge response). 48% (n = 260), 64% (n = 260) and 58% (n = 112) of the included patients had sleep disorder complaints respectively before, during and after hospitalisation. HA usage increased when comparing pre- and during hospitalisation (33% vs. 51%; n = 112; p < 0.0001) and decreased when comparing during hospitalisation and post-discharge (51% vs. 43%; n = 112; p < 0.0001). The descriptive model showed an overall persistence of treatment induced by hospital stay in 5.35% of the patients. CONCLUSION: Hospital appeared to have a significant impact on delayed HA use in the French general population. Our results should incite hospital prescribers to transversally reconsider the whole sleep disorder treatment strategy in hospital settings, from improving patient's accommodation conditions, to working out a consensus on the justification of prescription of HA and precising the exact place of nursing team in sleep disorders management.

[Prevention of preparation errors of cytotoxic drugs in centralized units: from epidemiology to quality assurance]. / Maîtrise des erreurs de préparation des médicaments anticancéreux en unité centralisée: de l'épidémiologie à l'assurance qualité.

The centralized preparation of cytotoxic drugs is supposed to help preventing medication errors. Nevertheless, the residual risk of preparation errors has been demonstrated. Above all, prevention is based on the quality assurance policy. Following a previous work, this study valuates the effectiveness of a quality assurance policy and the impact of corrective actions. The study compares the rates of preparation errors observed before and after the corrective actions. The risk factors have also been studied. The study included 84,017 consecutive preparations. The results showed a significant reduction of overall (0.2% versus 0.4%) and major (0.1% versus 0.2%) errors between the two periods. This period impact has been validated by a multivariate analysis (OR were respectively 0.49 and 0.6, p < 10(-3)). The daily workload (> 60 preparations) appeared to be the most important risk factor (OR were respectively 2.4 and 3.2, p < or = 10(-3)). These results demonstrated the effectiveness of simple corrective actions. The follow-up on preparation errors appeared to be useful to raise technicians' awareness and to motivate them. An epidemiological approach allows to identify risk factors, and to definite relevant corrective actions.

Observational study of potential risk factors of medication administration errors.

OBJECTIVE: Medication administration errors (MAEs) are the second most frequent type of medication errors, as has been shown in different studies in the literature. The aims of this observational study were to assess the rate and the potential clinical significance of MAEs and to determine the associated risk factors. DESIGN: In two departments, Geriatric Unit (GU) and Cardiovascular-Thoracic Surgery Unit (CTSU) of Besançon University Hospital (France), MAEs were identified using the undisguised observation technique and classified according to the definitions of the American Society of Health-System Pharmacists. Injectable administration, lack of nurses's standardized protocol for the preparation and administration of drugs, incomplete or illegible prescription and nurse's workload were analysed as potential risk factors of MAEs in multivariate logistic regression analysis. RESULTS: During a period of 20 days, opportunities for error concerning 56 patients and 78 MAEs (58 in CTSU and 26 in GU) were observed. The medication administration error rate was 14.9%. Dose errors were the most frequent (41%) errors, followed by wrong time (26%) and wrong rate errors (1996). No potential fatal errors were observed, 8 (10%) were estimated as potentially life-threatening, 20 (26%) potentially significant and 50 (64%) potentially minor. Nurse workload and incomplete or illegible prescriptions were two independent risk factors of MAEs. CONCLUSION: According to these data, the quality of the medication administration process needs to be optimized in hospitals in order to minimize the incidence of iatrogenic preventable diseases.