Background Antineoplastic preparation presents unique safety concerns because of its toxicity and narrow therapeutic window. Moreover, the antineoplastic use system includes several stages that are vulnerable to opportunities for potentially harmful medication errors.

Purpose To identify the type and underlying factors of medication errors during antineoplastic drug preparation, to identify improvement strategies.

Material and methods Design: prospective, observational study, from April to June 2016, using a disguised observation technique. A team of four pharmacy students were specifically trained to make the observations unobtrusively for this study.

Setting: Hazardous Preparation Unit of the Hospital Pharmacy Department in a 1300-bed tertiary teaching hospital. In this Unit, quality control of the final products is made by a nurse, who compares the preparation order instructions with all vials used to make the preparation.

Definitions: a preparation error was considered when the dose was >5% variation from the prescribed, or the preparation did not meet pharmacy quality standards, such as wrong fluid or wrong final volume that implied instability of the drug. Otherwise, they were considered discrepancies.

An independent team classified errors, discrepancies and their potential causes according to the Ruiz Jarabo 2008 classification. Potential clinical severity for errors was assigned as minor, moderate and serious.

Results Eight (1.41%) errors and 15 (2.65%) discrepancies were intercepted in 566 preparations observed. The errors detected were classified as: wrong dose (0.53%), wrong protection from light (0.35%), wrong drug (0.18%), wrong fluid (0.18%) and wrong label (0.18%). The discrepancies documented were: wrong preparation technique (1.24%), wrong dose (1.06%), and wrong fluid (0.35%).

Potential severity of preparation errors was: 33% minor (no damage or monitoring required) and 67% moderate (temporary damage that required monitoring or treatment). All of them were detected and corrected at the pharmacy and did not reach the patients.

Potential error causes detected were: lapse of concentration (78%) and lack of standardised procedures (22%).

Conclusion Although the identified error rate is very low and consistent with previous studies, the high intrinsic risk of antineoplastic drugs calls for a zero rate target. For this reason, and in order to improve preparation accuracy, new strategies such as automised workflow management systems will be of use in the near future.

No conflict of interest