Poor Identification of Emergency Department Acute Recreational Drug Toxicity Presentations Using Routine Hospital Coding Systems: the Experience in Denmark, Switzerland and the UK.

Background: Understanding emergency department and healthcare utilisation related to acute recreational drug toxicity (ARDT) generally relies on nationally collated data based on ICD-10 coding. Previous UK studies have shown this poorly captures the true ARDT burden. The aim of this study was to investigate whether this is also the case elsewhere in Europe.

Methods: The Euro-DEN Plus database was interrogated for all presentations 1st July to 31st December 2015 to the EDs in (i) St Thomas' Hospital, London, UK; (ii) Universitätsspital Basel, Basel, Switzerland; and (iii) Zealand University Hospital, Roskilde, Denmark. Comparison of the drug(s) involved in the presentation with the ICD-10 codes applied to those presentations was undertaken to determine the proportion of cases where the primary/subsequent ICD-10 code(s) were ARDT related.

Results: There were 619 presentations over the 6-month period. Two hundred thirteen (34.4%) of those presentations were coded; 89.7% had a primary/subsequent ARDT-related ICD-10 code. One hundred percent of presentations to Roskilde had a primary ARDT ICD-10 code compared to 9.6% and 18.9% in Basel and London respectively. Overall, only 8.5% of the coded presentations had codes that captured all of the drugs that were involved in that presentation.

Conclusions: While the majority of primary and secondary codes applied related to ARDT, often they did not identify the actual drug(s) involved. This was due to both inconsistencies in the ICD-10 codes applied and lack of ICD-10 codes for the drugs/NPS. Further work and education is needed to improve consistency of use of current ICD-10 and future potential ICD-11 coding systems.