Kontrollierte Interventionsstudie zum Nutzen eines klinischen Pharmazeuten auf einer deutschen chirurgischen Intensivstation

BACKGROUND: The benefit of a clinical pharmacist in the intensive care unit (ICU) has been shown by numerous international studies. Previous inves-tigations found an impact on the reduction of medication errors, preventable adverse drug events, drug costs and improvement of clinical outcomes such as the ICU length of stay or mortality. These findings are not directly transferable to the German health care system. On one hand establishing clinical pharmacy services on the ward is not as well-advanced in Germany compared to the USA or the UK. On the other hand, the countries’ requirements differ concerning education and qualification of the clinical pharmacists. Valid data on the over-all incidence of prescribing errors and the number, impact and acceptance of pharmaceutical interventions in German ICUs is very limited. The high risk of medication errors occurring in the intensive care setting has been revealed by various studies. Due to the scarcity of data, a controlled interventional study was designed to investigate the benefits of a clinical pharmacist in a German surgical ICU. METHODS: This study investigates the influence of clinical pharmacists’ medication review on the prescribing error rate in a German surgical ICU. Sec-ondary outcomes were the reduction of potentially severe prescribing errors, the number of days without systemic anti-infective therapy, a fictitious ICU length of stay and the drug- and staff costs. A controlled interventional study was conducted in a surgical ICU with one control phase (P0) and two interven-tion phases (P1 and P2). The investigation aimed to determine if the medication review by clinical pharmacists results in a significant reduction of prescribing errors related to a control period. The prescribing errors were analysed descrip-tively. Throughout P0 the data was collected retrospectively without any inter-vention. During the intervention periods P1 and P2, two clinical pharmacists screened the medical records for prescribing errors and discussed them with the senior physician in charge. During P2 one clinical pharmacist attended ward rounds additionally. Prescribing errors detected by the clinical pharmacists were only taken into account, if consent with the physicians was achieved. RESULTS: The medication orders of 336 patients were reviewed. The inci-dence of prescribing errors was significantly reduced from 1660 in P0 to 622 in P1 respectively 401 in P2 (P0 vs. P1/P2 respectively; both p<0.001; Fisher’s Ex-act Test) in total, respective 14.12% in P0 vs. 5.13% in P1 and 3.25% in P2 re-lated to the monitored medications (P0:11,755; P1:12,134; P2:12,329). The total rate of potentially severe prescribing errors was found to be 9.9% in P0 and 3.1% and 1.8% in P1 and P2 respectively related to the monitored medications. Clinical pharmacists performed 1874 interventions during P1 and P2. Consen-sus with the physicians was reached in 77.6% (P1) and 80.7% (P2) of all inter-ventions. Most of the prescribing errors throughout the study were owed to “drug indicated but not prescribed”. Regarding the number of monitored days without systemic anti-infective therapy medians of 1.0 day per patient were de-tected in all three phases. The 75th percentiles were 3.0 days (P0, P1) and 4.0 days (P2). During P0 237 days without systemic anti-infective therapy were counted, in P1 and P2 343 and 328 days respectively. Relating the number of all monitored days without systemic anti-infective therapy per study-phase to the total number of monitored patient days, significantly (p < 0.01; Fisher’s exact test) more days without systemic anti-infective therapy occurred in P1 and P2. The median fictitious length of stay was found to be 8.0 days in P0, 7.0 days in P1 and 8.0 in P2. Compared to P0, the drug costs of P1 and P2 were reduced. The clinical pharmacists’ interventions resulted in € 4.16 staff costs per patient in P1 and € 7.19 in P2 respectively. CONCLUSION: Clinical pharmacists’ interventions led to a significant re-duction of prescribing errors in the ICU, contributing to a safer medication process. Potentially severe prescribing errors were reduced, as well as drug costs and the use of systemic anti-infective therapy. The savings concerning the drug costs were overcompensating the additional staff costs during the in-tervention periods. A broad implementation of clinical pharmacists in German ICUs is strongly recommended.