How do prescribing error rates in digital care compare to traditional care settings?
A new study published in the Journal of Healthcare delivered a world-first analysis of prescribing errors in a real-world digital obesity setting.
Prescribing errors, such as failing to adequately discuss possible side effects with a patient before issuing a script, occasionally happen in many parts of the health system. However, very few benchmarks exist to indicate an acceptable rate of such errors.
Within the virtual care sector in Australia, there are no such benchmarks (let alone uniform safety and quality standards more generally). In this study, Eucalyptus sought to lay a foundation for the development of prescribing safety standards for obesity services and digital care in general.
The study detected errors in 4.4% of the audited prescriptions — a rate that is lower than those reported in Australian hospital settings and overseas digital care models.
Understanding the results
The study aimed to determine the rate and types of prescribing errors made in GLP-1 prescriptions within the Juniper service. Investigators analysed GLP-1 prescription errors over a 6-month period in 2023 which had been identified by the Eucalyptus clinical governance team’s four auditing methods: automated queries, new prescriber audits, ad-hoc audits and random audits.
The analysis detected errors in 1,654 (4.4%) of the 37,323 audited prescriptions.
Nearly half of all errors (49.15%) involved a failure to provide proper safety counselling, such as not discussing the potential medication side effects with patients. A further 30.29% of errors were related to inadequate assessments of possible contraindications, and 11.31% pertained to incorrect dosing.
Interestingly, patients with a body mass index (BMI) over 40 kg/m² were found to be 25% more likely to experience prescribing errors than those with lower BMIs, highlighting the group’s need for additional care.
Studies of electronic prescribing in Australian hospital settings have reported error rates between 9 and 27%, while rates in European and American digital community settings appear to range from 6.6 to 51.4%.
Although the 4.4% error rate detected in the Juniper is relatively low, the authors argue that a reasonable comparison cannot be made until data from other digital weight-loss services becomes available.
Significance of this research
This is the first Australian study to analyse electronic prescribing error rates outside of a hospital setting and the first study in the world to report these rates in an obesity programme.
The study’s results not only lay a vital foundation for prescribing error standards but also shed light on the clinical governance possibilities of digital healthcare. Specifically, they suggest that multi-method auditing approaches, such as those used by Eucalyptus, can be effective in identifying prescribing errors and ensuring patient safety.
