Adherence to Anti-Infective Prescribing Recommendations in Institutional Febrile Neutropenia Guidelines

Introduction: Infection is a major cause of morbidity and mortality in patients with severe neutropenia. In order to optimize patient outcomes and prevent the development of antibiotic resistance, it is important to select optimal anti-infective therapy when appropriate. In 2012, the clinical practice guideline for patients with febrile neutropenia at a free-standing children’s hospital was updated to reflect the most current medical evidence. It is unknown whether anti-infective prescribing behaviors are consistent with guideline recommendations or if deviation from the guideline is well-documented by prescribers. The objective of this study was to determine prescribing adherence to the updated febrile neutropenia clinical practice guideline. Secondary objectives included characterizing factors that influenced non-adherence and describing prescribers’ practices for documenting rationale for occurrences of non-adherence.

Methods: This was a retrospective cohort study of patients admitted to a free-standing children’s hospital who had febrile neutropenia. Patients were identified using pharmacy reports to select patients with an absolute neutrophil count < 500 cells/mm3 who received at least one of the antimicrobials included in the institutional clinical practice guideline from January 1, 2013 to December 31, 2013. Descriptive statistics were used for frequencies of empiric drug selection, de-escalation, and/or discontinuation that were adherent to the clinical practice guideline. Spearman correlation was utilized to determine association between patient variables and guideline non-adherence.

Results: One hundred thirty-three patients with febrile neutropenia were included. Twenty incidents (15%) of guideline non-adherence were noted, resulting in overall adherence of 85%. There was no correlation between specific patient variables and guideline nonadherence. Cefepime and vancomycin were the drugs most often associated with guideline non-adherence, accounting for 35% (n=7) and 30% (n=6) of the incidents, respectively. For all instances of guideline nonadherence involving vancomycin, rationale for empiric addition of vancomycin could not be located in the electronic medical record. It could not be determined if the patient truly did not meet guideline criteria for empiric vancomycin or if empric vancomycin was appropriate, but documented rationale was absent.

Conclusion: Overall adherence to the institutional clinical practice guideline for febrile neutropenia was good at 85%. Cefepime and vancomycin were responsible for most inconsistencies, so it may be beneficial to focus educational efforts on recommended use of these specific antibiotics. Since documentation for the empiric use of vancomycin was inadequate, encouraging prescribers to document clinical justification for the empiric use of vancomycin may increase ability to determine adherence in the future. For future guideline updates, it may be useful to include objective measures for the determination of factors including measures for degree and severity of mucositis and hemodynamic instability.