Prevention of adverse drug reactions in intensive care patients by personal intervention based on an electronic clinical decision support system

Abstract

We investigated the effect of written drug information for senior clinicians on the incidence of drug-drug interactions (DDIs) and DDI-related adverse events in intensive care patients. A prospective controlled intervention cohort study was conducted in a medical intensive and intermediate care unit in a university hospital. From 1,062 consecutive intensive care patients, those 265 (control: 136, intervention: 129) with > or =8 concurrently prescribed drugs were included in the study (to include high-risk patients with polypharmacy). The DDI information for senior clinicians during an intervention period of 3 months was based on a computerised clinical decision support system (CDSS) containing information on risk and management of 9,453 drug combinations. The number of patients with at least one DDI at the end of the respective study phase decreased by 18% (relative risk reduction) from 90 (66%) patients in controls to 70 (54%) in the intervention group (p = 0.02). The relative risk of a patient suffering from at least one DDI-related adverse event decreased by 43% from 60 (44%) patients in controls to 32 (25%) in the intervention group (p < 0.01). Among these events, the incidence of QT(C) prolongation was reduced by 64% from 15 (11%) patients in the control group to 5 (4%) in the intervention group (p = 0.04), and the incidence of hypokalemia by 80% from 14 (10%) to 2 (2%, p < 0.01). Written drug information based on a CDSS considerably decreased DDIs and DDI-related adverse events in routine practice.