Aim The aim of the pharmacy intervention audit was to prospectively record the number and type of interventions made to paediatric oncology chemotherapy prescriptions. This baseline data will be used in the future to assess the impact of electronic prescribing (EP) on prescribing error or intervention rates.
Independently from the EP project research, I interrogated the data to establish if there was a correlation between prescribing workload and rate of errors or interventions. I predicted that an ‘overworked’ prescriber would make more mistakes due to the volume of the workload and a less frequent prescriber would make more mistakes due to scarce use of these skills.
Intervention rates have been found to be as high as 66% for chemotherapy prescriptions, including interventions for missed information, wrong doses and protocol breach1. This intervention rate demonstrates the importance of pharmacy verification. Another source demonstrated that 80% of errors were due to poor prescription writing.2
Method An audit tool was created to collect the data, included fields for date, prescriber type, number of drugs prescribed, number of interventions made and intervention type. Data was collected from 5 Jan 2015 to 12 Jun 15.
Results The most common interventions required were addition of diluent volume, addition of start date and dose amendments to ensure doses could be accurately measured.
The staff grade doctor prescribed on average 75% of the prescriptions each week, with an intervention rate of 19%. The registrar was responsible for 23% each week and had an error rate pf 24%. Consultants were responsible for only 2% of the weekly prescription workload and had the lowest rate of interventions at 7%. There was no clear correlation between percentage of chemotherapy prescribed per week and rate of errors.
Conclusion The most common types of errors expected from the background reading are demonstrated by this audit, as the three common interventions are related to poor prescribing. EP should eliminate all three of these interventions as all these are either mandatory fields for a prescription to be ordered or measurable dose rounding will be in inbuilt into each drug field, and therefore calculated automatically by our prescribing system.
There was no clear correlation between error rate and proportion of prescribing. Errors are therefore independent of prescribing workload. Alternative reasons for errors could include external factors such as environment or bad habits of the prescriber. I believe the low rate of errors from the consultants is due to the types of prescriptions they often prescribe. Which were more frequently for single agents such as intrathecals. This suggests further data interrogation could identify whether there is a relationship between prescription complexity (or length) and error rate.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.