[Reducing prescription errors in oncological systemic therapies by implementing software-assisted therapy planning]

Chemotherapy and antibody-based therapies are cornerstone treatments in modern oncology, but they carry a substantial risk of therapy-related morbidity and mortality. Prescription errors-including incorrect patient information, inaccurate dosing, inappropriate treatment regimens, or insufficient consideration of patient-specific factors-can further exacerbate this risk.

Investigation of the impact of a software-based electronic therapy planning system (CPOE, Computerized Physician Order Entry) on the frequency and spectrum of prescription errors, which were previously prioritized by an interdisciplinary team according to their potential severity.

Retrospective analysis of all oncological systemic therapy orders processed in the pharmacy of the University Medical Center Göttingen between January 1 and December 31, 2021. Therapies were ordered either via the Excel-based prescription system (TKV) or using CPOE. Medication errors were systematically identified during pharmacist-led medication review, prioritized according to potential clinical relevance, and classified using an adapted application of the NCC MERP Index, which was used in this study to assess the potential risk of prevented errors.

Among 27,898 oncological systemic therapies, TKV prescriptions showed a higher error rate (0.73 %) than CPOE prescriptions (0.48 %, p = 0.0078). Errors involving incorrect patient data, which were classified as particularly relevant in the internal prioritization process, occurred exclusively in the TKV group. All other prescribing errors were consistently less common, although the difference was not statistically significant.

The implementation of a fully integrated CPOE-based therapy planning system significantly reduced the overall number of prescribing errors, altered the error spectrum, and specifically prevented the error of incorrect patient data, which was rated as particularly critical. Considering the limitations of the study, these findings suggest an improvement in care quality and patient safety, provided that the software is embedded within a continuously monitored quality management system.