Aims To compare time to antibiotics in febrile neutropenia cases, with and without use of a patient group direction (PGD). The PGD empowers nurses to administer tazocin, within strict criteria, without a doctor’s prescription.

Methods In a district general hospital, a PGD was created allowing specifically trained nurses to administer tazocin to children with cancer at risk of febrile neutropenia. Data was collected retrospectively over six months (January to June 2016) using computer records and clinical notes. Time to antibiotics and use of PGD was recorded. Two-tailed Mann-Whitney test was used to assess statistical significance.

Results There were 31 episodes of suspected febrile neutropenia in 14 patients. 50% of patients were male, median age was 8 years and 64% had haematological malignancy. Overall, mean time to antibiotics was 43 min (83% received antibiotics within an hour). Nurses used the PGD in 45% of episodes (14/31). Mean time to antibiotics was 25 min using PGD, and 59 min with doctor prescription (p=0.0004, see table 1). One third of episodes with doctor prescription (5/15) resulted in antibiotics given outside the golden hour. No episodes of patient harm, or risk of harm, were identified through the use of the PGD. 26 senior nurses have been trained on this PGD over the last two years.

Conclusion Training nurses to administer tazocin to children with suspected febrile neutropenia, via PGD, results in improved time to antibiotics, with less variability and increased number receiving treatment within the golden hour. Sample size is relatively small but despite this, the relationship was strongly statistically significant. These results show the power of this approach to improve outcomes in febrile neutropenia. The National Institute for Health and Care Excellence (NICE) recommend that PGDs are used to improve patient care without compromising patient safety, which is supported by this data. A strong multidisciplinary approach is required with close attention to governance arrangements and accountability.