Background Adverse drug events (ADEs) can lead to increased morbidity, mortality and costs. Previously, the analysis of ADEs in North Denmark Region (NDR) has not led to comprehensive initiatives, and no preventive regional strategies had been developed within the medicines area. To ensure the learning potential of reported ADEs, a thorough analysis needs to be performed.

Purpose To re-categorise the reported ADEs in the NDR, and to assess whether the current classification system did identify the underlying causes of the ADE.

Materials and methods The project was designed as a retrospective, descriptive study. The ADEs were from January 2012 to January 2013.

The ADEs were re-categorised according to the WHO classification system by a clinical pharmacist, and the ADEs related to ‘prescribing’ were also categorised according to the Ferner and Aronsons (FE) classification model.

Results Of the 463 ADEs, all were categorised by the rapporteur within the ‘prescribing’ process, whereas the distribution of ADEs categorised by the clinical pharmacist included seven categories: prescribing (88.7%), prescription management (0.9%), dispensing (1.7%), administration (3.2%), documentation (3%), monitoring (0.02%) and other process (1.5%).

Categorising according to FE showed that the majority (91.8%) of ADEs, were categorised as ‘Good rules misapplied’ (40.2%), ‘Knowledge-based errors’ (30.9%) and ‘Action-based errors’ (20.7%).

Within ‘Knowledge-based errors’, the most common ADE (14.6%) was ‘Ignorance of drug/dose/treatment’. Within ‘Good rules misapplied’, 12.4% of the ADEs were categorised as ‘medicines reconciliation not performed/performed incorrectly’. ‘Wrong dose’ (7.8%) was the most common ADE within ‘Action-based errors’.

Conclusions The categorisation according to WHO classification system was not consistently performed by the rapporteur and by the clinical pharmacist. The WHO classification system points out where ADE happens within the medication process, but it does not identify the underlying cause of the ADE. The FE classification model provides more detailed information on the underlying cause of the ADE and is recommended for future use in NDR.

No conflict of interest.