Introduction The Medication Prescription and Administration Record (Kardex) must always be legible and complete. It is this initial point of prescription writing that opens the way for errors and delays to delivery of treatment to occur due to omitted or misinterpretation of the prescription.¹

A Regional Safety Forum introduced a Paediatric Safety, Quality and Improvement Collaborative in November 2013, where the local Trusts could come together to learn about new initiatives and discuss ongoing projects. One area our Trust felt was important and decided to explore was medication safety, starting with medication prescribing.

Overall aim The aim of this project is to examine how well the kardex is being completed and how it can be improved.

Methods Data collection was commenced in September 2014 by the paediatric pharmacist.

A weekly kardex review audit was carried out looking at two areas:

1. The correct completion of the front of the kardex (against 7 criteria – patient name, hospital number, date of birth, allergy status, ward, weight, date of weight).

2. The overall prescription writing error per drug (against 17 criteria).

Five kardexes each week were audited and were selected randomly from the current in-patients.

An education intervention was a poster for the ward to highlight best practice prescribing guidelines.

The initial 4 week results were presented at the next Paediatric Learning Session. Any particular areas of concern were fed back to the medical staff on the ward. Data collection continued to determine if there was any improvement with medical staff being aware of the audit and also with the poster being prominently displayed on the ward.

Results In September 2014 the accurate completion of the front of the kardex was 94.3% (week 1); 94.3% (week 2); 93.9% (week 3) and 94.3% (week 4). This improved in October to 94.3% (week 5); 100% (week 6); 94.3% (week 7) and 100% (week 8).

The mean prescription writing error per drug in September was 1.375 error per drug with the weekly results documented as 1.8 (week 1); 0.8 (week 2); 1.7 (week 3); 1.2 (week 4), compared to October which showed the overall mean prescription writing error per drug to be 0.3275. The weekly results were 0.62 (week 5); 0.2 (week 6); 0.36 (week 7) and 0.13 (week 8).

Conclusion The completion of the front page of the kardex has shown improvement from September to October. It is now being completed 100% accurately, compared to initially only 94.3%.

There has been a reduction in the prescription writing error rate, as is shown in the results, but there is room for further improvement.

At the most recent Paediatric Safety Forum meeting it was discussed how the different Trusts were auditing prescribing and that we could consider the possibility of both a regional audit tool and also look at a Regional Paediatric Pharmacy Training on Prescription Writing for the paediatric doctors rotating.

Reference

1. Maxwell SRJ, Wilkinson K. Writing safe and effective prescriptions in a hospital kardex. J R Coll Physicians Edinb2007;37:348–351.