Background Medicines Errors (MEs) in patients with cancer are a known safety concern.

Purpose To analyse potential and real MEs detected in the healthcare process in oncology patients.

Materials and methods We conducted a retrospective, observational study, from January 2009 to March 2013 in a general hospital, using an electronic prescribing system. The study examined the oncology prescriptions for adult patients receiving Intravenous Antineoplastic (IA) and/or Supportive-care Treatment (ST). We identified MEs at the stages of prescription, pharmacist validation of the prescription, preparation, dispensing and administration of IA and/or ST in cancer patients. Data collected: stage in the drug treatment process and drug involved. We used an error seriousness scale, described by Serrano, to classify the errors that reached the patient: MEs with a score of 4 or more were considered to be clinically significant.

Results A total of 37,425 IAs and STs were prepared and administered to cancer patients, 167 MEs were recorded (accounting for 0.004% of the total preparations). Prescription errors were the most common (n = 80), followed by preparation errors (n = 45), pharmacist validation errors (n = 29) and dispensing/administration errors (n = 13). Regarding the type of drug involved in MEs, 133 were IAs (42% were 5-fluorouracil, trastuzumab or cetuximab) and 34 were ST (68% were filgrastim or zoledronic acid). 13% of errors reached the patient with 4% being clinically significant: 6 MEs were caused by extravasations (which were quickly resolved with extravasation protocols) and one was a dispensing error (omission of filgrastim) where the patient suffered from neutropenia and he had to be hospitalised for one week. Each error was systematically reviewed by a multidisciplinary team in order to minimise future incidents.

Conclusions The majority of potential MEs were detected and resolved by pharmacists and nurses at the prescription’s validation and preparation stages. A double check in the dispensing/administration stage is necessary in order to prevent MEs that could have a serious impact on patient morbidity.

No conflict of interest.