Consistency among two drug interaction compendia in onco-haematological inpatients

• Autores: Mª. Ángeles Fernández de Palencia Espinosa, Alberto Espuny Miro, María Sacramento Díaz Carrasco
• Localización: European journal of clinical pharmacy: atención farmacéutica, ISSN 2385-409X, Vol. 18, Nº. 2, 2016, págs. 90-97
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Background and purpose: Drug-drug Interactions (DDI) are an important subset in medication safety and can be considered a preventable error. Health professionals can check over different databases to research detailed information, such as severity, scientific evidence, pharmacologic effect, mechanism of action and management of DDI. The objectives of this study are to evaluate the consistency of two commonly used drug interaction compendia by making a comparative assessment on the ability to detect the presence of any potential DDI, and discrepàncies on DDI listing and severity and scientific evidence ratings, in treatments from onco-haematological inpatients, both adults and children.

    Method: This was a prospective, observational and descriptive study that was conducted over a 12-week period. Twice a week, every patient’s treatment sheet was collected and each medication profile was screened through two databases: MicromedexTM and Drug Interaction FactsTM (DIFTM). All identified DDI were recorded and graded by their level of severity and scientific evidence. The agreement among databases, regarding to the ability to detect the presence of any potential DDI, was determined by kappa statistic. Both qualitative discrepancies on DDI listings and severity and scientifc evidence ratings among databases were assessed using descriptive statistics. For DDI commonly detected by both compendia, Spearman’s rank correlation coeficient (ƿ rank) was used to evaluate the agreement between severity and evidence ratings.

    Results: A total of 1,166 treatments sheets were analyzed, belonging to 341 onco-haematological patients, and 14,401 medications were prescribed (329 antineoplastic agents). Overall, 5,144 DDI were identified and distributed in 3,155 by MicromedexTM and 1,989 by DIFTM. Regarding to the ability to detect the presence of any potential DDI, a weak congruence was determined (kappa = 0.372; p value <0.0001) among databases. In fact, there were 2,057 DDI included by MicromedexTM and not provided by DIFTM; similarly, 891 DDI were collected by DIFTM and not recorded in MicromedexTM. Regarding to DDI commonly detected by both databases, 1,098 DDI were identified. Discrepancies in the severity rating were found in 43.5% of DDI. Moderate and minor DDI were similar among databases, but MicromedexTM was likely to detect a higher proportion of major DDI. There was a null concordance between these compendia in respect of the severity and scientific evidence ratings (ƿ rank = 􀂱0.495, p value <0.001; and ƿ rank = -0.098, p value = 0.001, respectively).

    Conclusions: There was a lack of agreement in the ability to detect the presence of potential interactions, as well as in the inclusion and grading of them across these two drug interaction compendia. An effort is needed to identify clinically relevant interactions in order to provide a better care for onco-haematological inpatients